Intelligent Manufacturing and Energy Sustainability: Proceedings of ICIMES 2019 (Smart Innovation, Systems and Technologies, 169) 9811516154, 9789811516153

This book includes selected, high-quality papers presented at the International Conference on Intelligent Manufacturing

129 40 40MB

English Pages 883 [844] Year 2020

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Conference Committee
Chief Patron
Patrons
Conference Chair
Honorary Chair
Publication Chair
Convener
Organizing Chair
Organizing Secretary
Coordinators
Session Chairs
Organizing Committee
Web Developer
Proceedings Committee
Technical Program Committee
Publicity Committee
Registration Committee
Hospitality Committee
Sessions Committee
Certificate Committee
Decoration Committee
Transportation Committee
International and National Advisory Committee
Preface
Contents
About the Editors
Performance Evaluation of Multi-layer Barriers for Machine-Induced Low-Frequency Noise Attenuation
1 Introduction
2 Mathematical Model
3 Experiment Methodology
4 Results and Discussion
5 Conclusion
References
Planing Process on AISI S-1006, S-7, and S-4340, Based on Johnson–Cook Model Using Numerical Technique
1 Introduction
1.1 Modeling Techniques and Assumptions
2 Results and Discussion
3 Conclusions
References
An Analytical Study of Diametral Error in Simultaneous Turning Process
1 Introduction
2 Formulation of Diametral Error of Workpiece with Cutting Tools kept Opposite to Each Other
3 Comparison of Analytical Model with Literature
4 Conclusion
References
Improvement of Electric Heater Design for Household Cooking Application in Developing Countries
1 Introduction
2 Experiment Methodology
3 Results and Discussion
4 Conclusions
References
Embodiment of an Efficient Brown’s Gas Compound Fuel Tank
1 Introduction
2 Literature Survey
3 Materials and Methods
3.1 HHO Module Design
3.2 Tank Design and Development
4 Water Flow Sensor Results and Discussions
5 Conclusions
References
Automated Solar Photovoltaic Panel Cleaning/Cooling System Using Air–Water Mixture and Sustainable Solutions to Off-Grid Electrification
1 Introduction
2 Proposed Methodology
2.1 Automation of Solar Panel Cleaning and Cooling System
3 Results and Discussion
4 Conclusions
References
Design and Fabrication of Four-Way Multi-hacksaw Cutting Machine
1 Introduction
2 Fabrication
3 Working
3.1 Working Principle
4 Design Calculations
5 Cost Analysis
6 SWOT Analysis
7 Conclusions
References
Adhesion Strength of Plasma Sprayed Coatings—A Review
1 Introduction
1.1 Selection Criterion in Plasma Sprayed Coatings
2 Adhesion Strength of Plasma Sprayed Coatings
3 Conclusions
References
An Experimental Studies on the Polymer Hybrid Composites—Effect of Fibers on Characterization
1 Introduction
2 Materials and Methods
2.1 Fabrication of Composites
3 Results and Discussions
4 Conclusion
References
Design Analysis and Pressure Loss Optimization of Automobile Muffler
1 Introduction
1.1 Backpressure
1.2 Effects of Increased Back Pressure
2 Methodology
3 Geometry
3.1 Meshing
3.2 Assumption and Boundary Conditions
4 Results and Discussions
4.1 Inlet Velocity: Case 1: 39 M/S
4.2 Inlet Velocity: Case 2: 44 M/S
4.3 Inlet Velocity: Case 3: 49 M/S
5 Conclusion and Future Work
References
Different Modules for Car Parking System Demonstrated Using Hough Transform for Smart City Development
1 Introduction
2 Technology Gap
3 Background
4 Hough Transform
4.1 Find Circle Using Hough Transform
5 IP Webcam
6 Proposed Method
6.1 Step-by-Step Procedure
7 Simulation Results
7.1 Circle with Different Radius—Parallel Parking
7.2 Circle with the Same Radius, Angle Parking
8 Conclusion and Future Work
References
Effect of Arrangement and Number of Water Mist Spray Nozzles on Air Humidity
1 Introduction
2 Equipment Modification
3 Results and Discussion
4 Conclusions
References
Vibration Condition Monitoring of Spur Gear Using Feature Extraction of EMD and Hilbert–Huang Transform
1 Introduction
2 Gear Vibration Signals
3 Time–Frequency Representations
3.1 HHT
4 Results and Discussions
4.1 EMD
4.2 HSA and FFT
5 Conclusion
References
Comparative Study of PWM Technique for Switching Loss Reduction and Acoustic Noise Reduction in VSI-Fed Drives
1 Introduction
2 Advanced PWM Scheme
3 Switching Loss
4 Acoustic Noise Prediction
5 Simulation Results and Discussions
6 Conclusion
References
An Improved Genetic Algorithm for Production Planning and Scheduling Optimization Problem
1 Introduction
2 Research Methodology
2.1 Problem Modelling
2.2 Chromosome Encoding
2.3 Notations
2.4 Evaluation
2.5 Selection
2.6 Crossover
2.7 Mutation
3 Case Study
4 Experimental Results and Discussion
5 Conclusion
References
Automatic Calibration for Residential Water Meters by Using Artificial Vision
1 Introduction
2 Equipment Description
3 Instrument Measurement
4 Results
5 Conclusions
References
Hardware-in-the-Loop of a Flow Plant Embedded in FPGA, for Process Control
1 Introduction
2 Development of the HIL
2.1 The Flow Plant Design
2.2 Implementation of the Flow Plant
3 Controller Design
3.1 Design of the Continuous PID Controller
3.2 Design of the Discrete PID Controller
4 Tests and Results
4.1 Tests
4.2 Results
5 Conclusions
References
Error Diagnosis in Space Navigation Integration Using Wavelet Multi-Resolution Analysis with General Regression Neural Network
1 Introduction
2 Inertial Navigation Systems
2.1 Prediction of Errors in GPS
2.2 Parameters Obtained
3 Wavelet Analysis
4 Neural Networks
5 Result and Discussion
6 Conclusion
References
Hydromagnetic Squeeze Film Performance of Two Conducting Longitudinally Rough Elliptical Plates
1 Introduction
2 Analysis
3 Results and Discussions
4 Conclusion
References
Performance and Emission Characteristics of Biodiesel from Rapeseed and Soybean in CI Engine
1 Introduction
2 Literature Review
3 Methodology
4 Results and Discussion
4.1 Brake Power
4.2 Specific Fuel Consumption
4.3 Brake Thermal Efficiency
4.4 NOx Emission
4.5 CO2 Emission
4.6 Particulate Matter Emission
5 Conclusion
References
A Practical Approach to Camera Calibration for Part Alignment for Hybrid Additive Manufacturing Using Computer Vision
1 Introduction
1.1 Problem Identification
1.2 Camera Calibration
2 Literature Review
3 Experimentation
3.1 Setup
3.2 Results
3.3 Result Table
4 Conclusion
5 Future Work
References
Development of a Colour and Orientation Detection System for Small Part Feeding
1 Introduction
2 Experimental Setup
2.1 Colour Sensor
2.2 Infrared Sensor
2.3 Servo-Based Actuator
2.4 Microcontroller
3 Description of Algorithm Used in Code
3.1 Mean of a Set of 10 Readings of Colour Sensor to Reduce the Error
3.2 Identification of the Colour from the Mean RGB Values
3.3 Taking the Mean of 10 Infrared Sensor Values for More Accuracy and Sensing the Orientation of the Parts
3.4 Condition for Rejection of the Undesirable Part by the Actuator
4 Working
5 Conclusion
References
Comparative Analysis on Battery Used in Solar Refrigerated E-Rickshaw in India
1 Introduction
2 Smart Mobility System for Indian Megacities
3 BMS Duty for Li-Ion Battery Use in E-Rickshaw and SHRERs
4 The Relation Between State of Function with SOC and SOH
5 Conclusion
References
Application of Value Stream Mapping (VSM) in a Sewing Line for Improving Overall Equipment Effectiveness (OEE): A Case Study
1 Introduction
1.1 General
1.2 Objectives
2 Literature Review
2.1 Case Studies
2.2 Conceptual Work
2.3 Modeling Work
2.4 Survey Articles
3 Methodology
3.1 Current State VSM
3.2 Rough Set Theory
3.3 Future State VSM
4 OEE and Other Performance Parameters
4.1 Formula Used
4.2 OEE Calculation
4.3 Existing Scenario
4.4 Proposed Scenario
4.5 Calculation of Other Performance Parameters
5 Result and Conclusion
References
CFD Analysis to Increase Heat Transfer in Pipe with Combined Effect of Circular Perforated Ring and Twisted Tape Insert
1 Introduction
2 Numerical Model Description
2.1 Geometrical Model
2.2 Governing Equations
3 Results and Discussion
4 Conclusion
References
Study on Solar Parabolic Trough Collector with Different Copper Absorber Tubes
1 Introduction
2 Experimental Setup
2.1 Absorber Tubes
3 Experimentation
4 WASPAS Method
5 Results and Discussions
6 Observations
References
Harmonic Analysis for Bidirectional Grid-Connected Converter for Electrical Vehicle During Charging and Discharging Operations
1 Introduction
2 Grid-Connected Converter
3 Simulation Results
4 Conclusion
References
Effect of Tribo-layer on the Sliding Wear Behavior of Detonation Sprayed Alumina–Titania Coatings
1 Introduction
2 Methodology
2.1 Sample Preparation
2.2 Powder Preparation
2.3 Process Parameters
2.4 Coating Characterization
3 Results and Discussion
3.1 Surface Roughness, Microstructure and Porosity Values of Coatings
3.2 XRD Analysis
3.3 MicroHardness Values
3.4 Sliding Wear Behavior
3.5 Coefficient of Friction Values
3.6 Wear Mechanisms
3.7 EDS Analysis of Worn Surfaces
4 Conclusions
References
Development of Sustainable Cementitious Binder Utilizing Silicomanganese Fumes
1 Introduction
2 Methodology of Research
2.1 Materials
2.2 Experimental Program
2.3 Evaluation Methods
3 Results and Discussion
3.1 Flow
3.2 Setting Time
3.3 Compressive Strength
4 Conclusion
References
Review on Sliding Wear of Ti–6Al–4V Alloy Concerning Counterface and Sliding Conditions
1 Introduction
2 Literature Review
2.1 Microstructures
2.2 Coatings
2.3 Heat and Surface Treatment
2.4 Ti–6Al–4V Against Same and Different Alloys
3 Discussion and Result
4 Conclusion and Future Scope of Work
References
Hardness and Microstructure Studies on the Effect of Solution-Treated Age-Hardened Al-6082 Alloy
1 Introduction
2 Material Selection
3 Heat Treatment
4 Hardness Test
5 Result and Discussion
5.1 Hardness and Microstructure Results of Experiment No. 1
5.2 Hardness and Microstructure Results of Experiment No. 2
5.3 Hardness and Microstructure Results of Experiment No. 3
5.4 Hardness and Microstructure Results of Experiment No. 4
5.5 Hardness and Microstructure Results of Experiment No. 5
6 Conclusion
References
Floating Photovoltaic Thin Film Technology—A Review
1 Introduction
2 Floating photovoltaic (FPV) Technology
2.1 Design Requirements
2.2 Concepts
3 Thin Film Technology
3.1 Commercial Thin Film Technologies in Practice
3.2 Future Thin Film Technologies
4 Conclusion
References
Effect of Ethanol Fumigation on Performance and Combustion Characteristics of Compression Ignition Engine Fuelled with Used Cooking Oil Methyl Ester in Dual-Fuel Mode
1 Introduction
1.1 Present Work
2 Materials and Methods
2.1 Source
2.2 Trans-esterification of UCO
3 Experimental Set-Up
3.1 Details of Ethanol Fumigation Rate
3.2 Details of Energy Sharing of Ethanol
4 Results and Discussions
4.1 Effect of Ethanol on Engine Performance
4.2 Effect of Ethanol on Combustion Analysis
5 Conclusions
References
Estimation and Moderation of Harmonics in Distribution Systems
1 Introduction
2 Estimation of Harmonics
3 Mitigation of Harmonics
3.1 Harmonic Reduction Using Pulse Converters
4 Proposed Method
5 Result and Analysis
6 Conclusion
References
Theoretical Evaluation of Energy Performance of a Vapour Compression Refrigeration System Using Sustainable Refrigerants
1 Introduction
2 Materials and Methods
2.1 Development of Sustainable Alternative Refrigerants
3 Methodology
3.1 Thermodynamic Analysis of Alternative Refrigerants
4 Results and Discussion
4.1 Energy Efficiency Ratio
4.2 Discharge Temperature of Compressor
4.3 Power Spent Per Ton of Refrigeration
4.4 Volumetric Refrigeration Capacity
5 Conclusions
References
Effect of Autofrettage on the Properties of the Aluminium Cylinders
1 Introduction
2 Experimental Studies
3 Result and Discussion
4 Conclusion
References
Influence of Graphene and Tungsten Carbide Reinforcement on Tensile and Flexural Strength of Glass Fibre Epoxy Composites
1 Introduction
2 Literature Review
3 Materials
4 Methodology
5 Experimental Study
5.1 Tensile Strength
5.2 Flexural Strength
6 Results and Discussions
6.1 Effect of Fibre Loading on Tensile Properties of Composites
6.2 Effect of Fibre Loading on Flexural Properties of Composites
7 Conclusion
References
Investigation of Partial Discharge Due to Copper Spherical Particle in Power Transformer Under Various Oil Flow Models Using CFD
1 Introduction
2 Simulation
2.1 Analysis Under Non-directed Oil Flow System
2.2 Analysis of Directed Oil Flow System
3 Discussions and Conclusions
References
A High-Power High-Frequency Isolated DC Power Supply for Electric Vehicle Charging Application
1 Introduction
2 Proposed Topology of EV Power Supply
2.1 T-type Converter Topology
3 Charge Controller
4 Efficiency Calculation
5 Simulation Results and Discussions
5.1 Dynamic Load Test
5.2 Charging of Battery Pack
6 Conclusion
References
Effect of Slicing Thickness and Increment on the Design of Patient Specific Implant for Total Knee Replacement (TKR) Using Magnetic Resonance Imaging (MRI)—A Case Study
1 Introduction
2 Methodology
2.1 Creating a 3D Model of Knee
2.2 Optimization of Data
3 Results and Discussion
4 Conclusions
References
Microstructure and Hardness Behaviour Study of Carbon Nanotube in Aluminium Nanocomposites
1 Introduction
2 Materials and Methodology
3 Synthesizing of Al–Cnt Metal Matrix Composites
4 Microstructural Analysis
5 Hardness Test
6 Results and Discussions
7 Conclusions
References
Comparison of Electric Fields with and Without Corona Ring for 66 kV Line Insulators
1 Introduction
2 Methodology
2.1 Numerical Methods
2.2 Calculation of Electric Field
3 Modelling
3.1 Properties of Different Materials
3.2 Model
4 Result and Discussion
4.1 Without Corona Ring
4.2 With Corona Ring
5 Conclusion
References
Experimental and FEA Simulation of Thermal-Fluid Interaction Between TIN Coated Tungsten Carbide Tool and Inconel-825 Workpiece
1 Introduction
2 Design of Experiment and Experimentation
3 Multiple Regression Model
4 Geometry and Mesh Model
4.1 Dry and Wet Condition
5 Results and Discussion
5.1 Dry and Wet Cutting Condition
6 Conclusions
References
The Effect of Recrystallization on Electrical Resistivity of Stir Casted SiCP/AA6061 Composite After Shot Peening
1 Introduction
2 Materials and Methods
3 Results and Discussion
3.1 Microstructural Examination
3.2 XRD Analysis
3.3 Electrical Resistivity
4 Conclusions
References
Design and Modeling of Wye Piece
1 Introduction
2 Modelling Using CATIA and Hypermesh
3 Meshing and Boundary Conditions Using Hypermesh
4 Analysis Results
5 Conclusions
6 Future Scope
References
Comparative Investigation on Gas Tungsten Arc Welding and Friction Stir Welding of Electrolytic Tough Pitch Copper Plates
1 Introduction
2 Materials and Methods
2.1 Material
3 Result and Discussion
4 Conclusion and Further Scope
References
Synthesis and Characterization of Functionally Graded Ceramic Material for Aerospace Applications
1 Introduction
2 Experimentation
2.1 Slip Casting
3 Results
3.1 Dielectric Constant
3.2 Density
3.3 Compressive Strength
3.4 Scanning Electron Microscope (SEM) Analysis
4 Conclusion
References
CI Engine Characteristic Investigation by Application of Metal-Based Additives with Biodiesel Blends
1 Introduction
2 Experimental Setup
3 Methodology
3.1 Production of Biodiesel from Waste Cooking Oil (WCO)
3.2 Determination of Fuel Properties
3.3 Preparation of Nanofluids
4 Results and Discussion
4.1 Performance Parameter
4.2 Combustion Parameter
5 Emission Analysis
5.1 Oxides of Nitrogen
5.2 Carbon Monoxide
5.3 Unburnt Hydrocarbon
5.4 Smoke Opacity
6 Conclusion
References
Simulation of Underground Cable Defects with the Detection of Partial Discharge
1 Introduction
2 Cable Defects
2.1 What Is Cable Defects?
2.2 Cause of Cable Defects
3 Results and Discussion
3.1 Void
3.2 External Cut
3.3 Termination and Left Out Completely
4 Conclusion
References
Comparison of C.I Engine Performance Parameters and Emissions by Varying Designs of Intake Manifolds
1 Introduction
2 Experimental Set-Up
2.1 Engine Specifications
3 Designs of Manifolds
3.1 Normal Manifold
3.2 Nozzle Inside the Inlet Manifold
3.3 Internal Threaded Manifold
4 Tabular Columns for Experimental Results
5 Results and Discussions
6 Conclusions
References
Variation of Time Lag, Decrement Factor and Inside Surface Temperature with Solar Optical Properties of Building Envelope in Different Climatic Zones of India
1 Introduction
1.1 Time Lag and Decrement Factor
2 Methodology
2.1 Sole Air Temperature
2.2 Climatic Zones of India
2.3 Time Lag and Decrement Factor Calculation
3 Results and Discussion
3.1 Time Lag and Decrement Factor Calculation
3.2 Electricity Saving
4 Conclusions
References
Study on Tensile and Hardness Properties of Aluminium 7075 Alloy Reinforced with Graphite, Mica and E-Glass
1 Introduction
2 Materials and Process
2.1 Materials
2.2 Process
2.3 Hardness Test
2.4 Tensile Test
3 Result and Discussion
3.1 Tensile Strength and Hardness
3.2 Signal-to-Noise Ratio
3.3 Analysis of Variance (ANOVA)
3.4 Microstructure Studies
4 Conclusion
References
ANN-Based Fault Classification and Section Identification Technique Using Superimposed Currents for Three-Terminal Transmission Line
1 Introduction
2 System Modelling
3 Feature Selection for Section Identification and Fault Classification
4 Results and Discussion
5 Conclusion
References
An Analytical Investigation for Combined Pressure-Driven and Electroosmotic Flow Without the Debye–Huckel Approximation
1 Introduction
2 Description of the Physical Problem
3 Mathematical Formulation
3.1 Poisson–Boltzmann Equation
3.2 Momentum Equation
3.3 Energy Equation
4 Solution of the Governing Equations
4.1 Electric Potential Distribution
4.2 Velocity Distribution
4.3 Temperature Distribution
5 Results and Discussion
6 Conclusion
References
Addressing the Green Tribology Advancement, Future Development, and Challenges
1 Introduction
1.1 Global Sustainable Development
2 Green Tribology Fields in Engineering
3 Utility of Green Tribology in Real Life
4 Green Tribology Development Direction
5 Current Status of Energy Loss Due to Friction and Wear Worldwide
6 Results and Discussions
7 Conclusions
References
Effect of Drying Temperature to the Thin Layer Drying Model of Sago Starch
1 Introduction
2 Materials and Methods
2.1 Sample Preparation
2.2 Drying Experiment
2.3 Data Analysis and Mathematical Modeling
3 Results and Discussions
3.1 Effect of Drying Temperature to the Drying Curve of Traditionally Processed Sago Starch and Drying Characteristic
3.2 Effect of Drying Temperature to the Fitting Drying Model of Traditionally Processed Sago Starch
4 Conclusions
References
Evaluating Mechanical Properties of Egg Shell, and Coco Peat Reinforced Epoxy Composite
1 Introduction
2 Methods
2.1 Preparation of Composite Material
2.2 Testing of Composite Material
3 Results and Discussions
3.1 Tensile Test
3.2 Compressive Test
3.3 Hardness Test
4 Conclusions
References
Design Modification of Rear Axle Housing by Fatigue Failure Analysis
1 Introduction
2 CAD and FE Model Description
3 Axle Housing Material
4 Loading Condition
5 Finite Element Analysis of Housing and Results
6 Theoretical Validation
7 Design Optimization
8 Results and Discussion
9 Conclusions
References
Mixture Design Using Low-Cost Adsorbent Materials for Decolourisation of Biomethanated Distillery Spent Wash in Continuous Packed Bed Column
1 Introduction
2 Materials and Methods
2.1 Collection and Bio-chemical Analysis of BDSW
2.2 Preparation and Characterisation of Biosorbent Carrier Materials
2.3 Batch Experiments
2.4 Continuous Column Flow Studies
2.5 Decolourisation Assay
2.6 Nutritional Analysis of Spent Modified Soil
3 Results and Discussion
3.1 Characteristics of BDSW
3.2 Properties of Biosorbent
3.3 Colour Degradation
3.4 Nutritional Quality of Spent Modified Soil
4 Conclusion
References
Prediction of the WPPO Biodiesel-Fuelled HCCI Engine Using Artificial Neural Networks
1 Introduction
2 Experimental Setup
3 Results and Discussion
3.1 Modelling with ANN
4 Conclusion
References
Dimensional Analysis of Form Drilling Parameters by Buckingham Pi Theorem and Optimization of Heat Generation in Form Drilling Process by Taguchi
1 Introduction
2 Literature Survey
3 Experimental Setup
4 Pi Terms Evaluated
4.1 Process Parameters
5 Results and Discussion
5.1 Analysis of Variance for Heat Generation
5.2 Analysis of Variance for Petal Height
5.3 Taguchi Analysis for Heat Generation
5.4 Taguchi Analysis for Petal Height
6 Conclusion
References
Comparison of Ductile, Flexural, Impact and Hardness Attributes of Sisal Fiber-Reinforced Polyester Composites
1 Introduction
1.1 Materials and Methods
1.2 Fabrication of Specimen
2 Mechanical Testing
2.1 Tensile Strength
2.2 Flexural Strength
2.3 Impact Strength and Hardness
3 Scanning Electron Microscope (SEM)
4 Results and Discussions
4.1 Ductile Properties
4.2 Flexural Properties
4.3 Impact Properties
4.4 Hardness Properties
5 Analysis of SEM
6 Conclusions
References
Optimization of EDM Process Parameters Using Standard Deviation and Multi-objective Optimization on the Basis of Simple Ratio Analysis (MOOSRA)
1 Introduction
2 Experimental Environment
3 Methodology
3.1 Optimization Problem
3.2 Standard Deviation Concept for Weight Calculation
3.3 MOOSRA Methodology
4 Results and Discussion
4.1 Allocation of Weights
4.2 Result Analysis
5 Conclusion
References
Modeling and Simulation of Trans Z-Source Inverter with Maximum Constant Boost PWM Control for Solar PV System
1 Introduction
2 T-Source Inverter
2.1 Equivalent Circuit and Operating States of TSI
3 Modeling of TSI-Based Solar PV System
4 Maximum Constant Boost PWM Control (MCB PWM)
5 Simulation Results and Discussion
5.1 Simulation Results of TSI for Solar PV System Using MCB PWM Algorithm
6 Conclusion
References
Driver Drowsiness Monitoring System
1 Introduction
2 Literature Survey
3 Proposed System
3.1 Eye Blink Detection
3.2 Yawn Detection
3.3 Head Location Analysis
4 Result
5 Conclusion
References
Detection and Control of Water Leakage in Pipelines and Taps Using Arduino Nano Microcontroller
1 Introduction
2 Methodology
3 Arduino Nano
4 Water Flow Sensor
5 Water Drop Sensor
6 Liquid Crystal Display
7 Relay
8 Solenoid Valve
9 Experimental Set-up
10 Conclusions
References
Smart Automated Processes for Bottle-Filling Industry Using PLC-SCADA System
1 Introduction and Background Theory
2 Proposed Methodology
2.1 SCADA Implementation of Bottle-Filling Industry
2.2 Warehouse Management Systems
2.3 Energy-Efficient Lighting Solutions
3 Results and Analysis
3.1 SCADA Implementation of Bottle-Filling Industry
3.2 Warehouse Management Systems
3.3 Energy-Efficient Lighting Solutions
4 Conclusion
References
Effect of Stirring Speed During Casting on Mechanical Properties of Al–Si Based MMCs
1 Introduction
2 Experimental Procedure
2.1 Casting
2.2 Impact Test
2.3 Tensile Strength Test
2.4 Compression Strength Test
2.5 Hardness Test
3 Results and Discussion
4 Conclusion
References
Exploration of Pillars of Industry 4.0 Using Latent Semantic Analysis Technique
1 Introduction
2 Methodology
2.1 Latent Semantic Analysis (LSA)
3 Results and Discussion
3.1 Data Selection and Analysis
3.2 Interpretation of Results
3.3 Pillars of Industry 4.0
4 Conclusions and Implications
References
An Overview of Different Topologies of DC/DC Bidirectional Converter for Different Applications
1 Introduction
2 Overview of Different Topologies
3 Conclusion
References
An Approach to Detect and Classify Defects in Cantilever Beams Using Dynamic Mode Decomposition and Machine Learning
1 Introduction
2 Methodology
2.1 Simulation, Data Acquisition, and Analysis
2.2 Dynamic Mode Decomposition
2.3 Support Vector Machines
3 Results and Discussions
4 Results and Discussions
References
Musculoskeletal Simulation and Analysis of Upper Limb Rehabilitation Device
1 Introduction
2 Materials and Methods
2.1 Upper Limb Rehabilitation Device Selection and Setting
2.2 Musculoskeletal Analysis using AnyBody Software
3 Results and Discussion
4 Conclusion
References
Fuzzy Regenerative Braking Strategy
1 Introduction
2 Vehicle Dynamics and Braking
3 Mathematical Model of Energy Regeneration
4 Braking Strategy
5 Design of Fuzzy Logic Controller
6 Simulation Results and Discussions
7 Conclusion
References
A New Framework for Secure Outsourcing of Medical Data
1 Introduction
2 Paper Organization
3 Related Work
4 Proposed System Architecture
5 Algorithms
5.1 Least Significant Bit
5.2 BlowFish Algorithm
5.3 SHA-256
6 Results
7 Security Challenges
8 Conclusion and Future Work
References
A Comprehensive View for Providing the Decision on Medicare Data
1 Introduction
2 Literature Survey
3 Recommended System for Providing the Decision on Medicare Data
3.1 Sources of Healthcare Claims Data
3.2 Analysis of Medicare Data
3.3 Development Phase
4 Experimental Results
5 Conclusion and Future Work
References
Predicting Cardiac Arrhythmia Using QRS Detection and Multilayer Perceptron
1 Introduction
2 Literature Survey
3 Prediction System Architecture
3.1 Data Preprocessing
3.2 Multilayer Perceptron
3.3 QRS Detection
4 Algorithms
4.1 Pan–Tompkins Algorithm
4.2 Multilayer Perceptron
5 Dataset
6 Implementation
7 Results
8 Conclusion
References
Full Length Driver Drowsiness Detection Model—Utilising Driver Specific Judging Parameters
1 Introduction
2 System Architecture
3 Implementation
3.1 Facial Landmarks
3.2 Eye Aspect Ratio
3.3 Mouth Aspect Ratio
3.4 Drowsiness Predication
4 Experimental Results
5 Conclusion and Future Work
References
Enterprise Reporting Solution on Integrating Business Intelligence for Operational and Financial Data
1 Introduction
2 Survey of Existing Models
3 Architecture of Analytical Reporting Using Business Intelligence
3.1 Detailed Design, ER (Entity Relationship) Diagram
4 Implementation
5 Conclusion
References
A Survey on Accelerating the Classifier Training Using Various Boosting Schemes Within Cascades of Boosted Ensembles
1 Introduction
2 Boosting Chain Learning
2.1 Booster Cascade
2.2 Cascade Training
2.3 Tuning the Boosting Cascade
2.4 Learning the Naive Booster Chain Framework
2.5 Using Bootstrap for Boosting Chain Learning
3 Optimization of Boosting Chain
3.1 Booster Optimization Based on a Linear Model
3.2 Adjusting the Classifier
3.3 Removal of Redundancy in Boosting
4 Performance Comparisons of the Three Detectors
5 Observations of Empirical Outcomes
6 Conclusion
References
Service Layer Security Architecture for IOT Using Biometric Authentication and Cryptography Technique
1 Introduction
2 Related Works
2.1 System Architecture
2.2 Algorithm (Honey Encryption)
3 AVISPA Code
4 Result Analysis
5 Conclusion and Future Work
References
Author Index
Recommend Papers

Intelligent Manufacturing and Energy Sustainability: Proceedings of ICIMES 2019 (Smart Innovation, Systems and Technologies, 169)
 9811516154, 9789811516153

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Smart Innovation, Systems and Technologies 169

A. N. R. Reddy · Deepak Marla · Milan Simic · Margarita N. Favorskaya · Suresh Chandra Satapathy   Editors

Intelligent Manufacturing and Energy Sustainability Proceedings of ICIMES 2019

Smart Innovation, Systems and Technologies Volume 169

Series Editors Robert J. Howlett, Bournemouth University and KES International, Shoreham-by-sea, UK Lakhmi C. Jain, Faculty of Engineering and Information Technology, Centre for Artificial Intelligence, University of Technology Sydney, Sydney, NSW, Australia

The Smart Innovation, Systems and Technologies book series encompasses the topics of knowledge, intelligence, innovation and sustainability. The aim of the series is to make available a platform for the publication of books on all aspects of single and multi-disciplinary research on these themes in order to make the latest results available in a readily-accessible form. Volumes on interdisciplinary research combining two or more of these areas is particularly sought. The series covers systems and paradigms that employ knowledge and intelligence in a broad sense. Its scope is systems having embedded knowledge and intelligence, which may be applied to the solution of world problems in industry, the environment and the community. It also focusses on the knowledge-transfer methodologies and innovation strategies employed to make this happen effectively. The combination of intelligent systems tools and a broad range of applications introduces a need for a synergy of disciplines from science, technology, business and the humanities. The series will include conference proceedings, edited collections, monographs, handbooks, reference books, and other relevant types of book in areas of science and technology where smart systems and technologies can offer innovative solutions. High quality content is an essential feature for all book proposals accepted for the series. It is expected that editors of all accepted volumes will ensure that contributions are subjected to an appropriate level of reviewing process and adhere to KES quality principles. ** Indexing: The books of this series are submitted to ISI Proceedings, EI-Compendex, SCOPUS, Google Scholar and Springerlink **

More information about this series at http://www.springer.com/series/8767

A. N. R. Reddy Deepak Marla Milan Simic Margarita N. Favorskaya Suresh Chandra Satapathy •







Editors

Intelligent Manufacturing and Energy Sustainability Proceedings of ICIMES 2019

123

Editors A. N. R. Reddy Department of Mechanical Engineering Malla Reddy College of Engineering & Technology Secunderabad, Telangana, India Milan Simic School of Aerospace, Mechatronics and Manufacturing Engineering RMIT University, Bundoora East Campus Bundoora, Melbourne, VIC, Australia

Deepak Marla Department of Mechanical Engineering Indian Institute of Technology Bombay Mumbai, Maharashtra, India Margarita N. Favorskaya Institute of Informatics and Telecommunications Reshetnev Siberian State University of Science and Technology Krasnoyarsk, Russia

Suresh Chandra Satapathy School of Computer Engineering KIIT Deemed to be University Bhubaneswar, Odisha, India

ISSN 2190-3018 ISSN 2190-3026 (electronic) Smart Innovation, Systems and Technologies ISBN 978-981-15-1615-3 ISBN 978-981-15-1616-0 (eBook) https://doi.org/10.1007/978-981-15-1616-0 © Springer Nature Singapore Pte Ltd. 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Conference Committee

Chief Patron Sri. CH. Malla Reddy, Founder Chairman, MRGI

Patrons Sri. CH. Mahendar Reddy, Secretary, MRGI Sri. CH. Bhadra Reddy, President, MRGI

Conference Chair Dr. V. S. K. Reddy, Principal

Honorary Chair Dr. Milan Simic, RMIT University, Australia Dr. Margarita N. Favorskaya, Reshetnev Siberian State University of Science and Technology, Russia

Publication Chair Dr. Suresh Chandra Satapathy, Professor, KIIT, Bhubaneswar, India

v

vi

Conference Committee

Convener Dr. M. Murali Krishna, Dean Academics

Organizing Chair Dr. A. N. R. Reddy, HOD, Mechanical Engineering

Organizing Secretary Dr. Srikar Potnuru, Associate Professor, Mechanical Engineering

Coordinators Prof. Akhila Kakera, Assistant Professor, Mechanical Engineering

Session Chairs Dr. Shahrol Bin Mohamaddan, Senior Lecturer, Universiti Malaysia Sarawak Dr. U. Vidya Sagar, TCS, Hyderabad

Organizing Committee Dr. V. Madhusudhana Reddy, Professor Dr. T. Siva Kumar, Professor Dr. T. Lokeswara Rao, Professor Prof. D. Damodara Reddy, Associate Professor Prof. B. Rajeshwar Reddy, Administrative Officer

Web Developer Mr. Harish Makena, Assistant Professor

Conference Committee

Proceedings Committee Prof. A. Sridhar, Assistant Professor Mr. O. Y. V. Subba Reddy, Assistant Professor

Technical Program Committee Dr. G. Trisekhar Reddy, Professor Dr. L. Subramaniam, Professor Dr. B. Jain AR Tony, Professor

Publicity Committee Mr. Ch. Naveen Kumar, Assistant Professor Mr. D. V. Subba Rao, Assistant Professor

Registration Committee Mr. Srinivas Reddy, Associate Professor Ms. B. Sandhya Rani, Assistant Professor Ms. Anupama Sai Priya, Assistant Professor Mr. K. Mahantesh, Assistant Professor

Hospitality Committee Mr. D. Mangeelal, Associate Professor Mr. Soma Vivekanandha, Associate Professor

Sessions Committee Ms. M. L. R. Chaitanya Lahari, Assistant Professor Ms. Sruthi Srinivasan, Assistant Professor Ms. Settem Lakshmi Nitheesha, Assistant Professor

vii

viii

Conference Committee

Mr. Sridhar Akarapu, Assistant Professor Mr. S. Brahma Reddy, Assistant Professor

Certificate Committee Ms. Sandhya Rani, Assistant Professor Ms. Haseena Bee, Assistant Professor

Decoration Committee Ms. Prasuna Lily Florence, Assistant Professor Ms. Y. Indraja, Assistant Professor

Transportation Committee Mr. Y. Dilip, Associate Professor

International and National Advisory Committee Dr. Lakhmi C. Jain, University of South Australia, Australia Dr. Narayanan Kulathu Ramaiyer, Universiti Malaysia Sarawak, Malaysia Dr. Abu Saleh Ahmed, Universiti Malaysia Sarawak, Malaysia Dr. Jaesool Shim, Yeungnam University, South Korea Dr. Sinin Hamdan, Universiti Malaysia Sarawak, Malaysia Dr. Amiya Bhaumik, Lincoln University College, Malaysia Dr. Shahrol Mohamaddan, Universiti Malaysia Sarawak, Malaysia Dr. Bhaskar Kura, University of New Orleans, LA, USA Dr. T. V. M. Sreekanth, Yeungnam University, South Korea Dr. V. Vasudeva Rao, University of South Africa, South Africa Dr. S. V. Prabhakar, Yeungnam University, South Korea Dr. Raja V. Pulikollu, Electric Power Research Institute, North Carolina, USA Dr. Yequing Bao, University of Alabama, USA Dr. Angel Sanz Anderes, UPM, Madrid, Spain Dr. Sabastian Franchini, UPM, Madrid, Spain Dr. G. Balu, DOAD, DRDL, Telangana, India Dr. K. Vijay Kumar Reddy, JNTU Hyderabad, Telangana, India

Conference Committee

ix

Dr. P. K. Jain, Centre for Carbon Materials, ARCI, Hyderabad, Telangana, India Dr. Susanta Kumar Sahoo, NIT Rourkela, Odisha, India Dr. K. S. Reddy, Indian Institute of Technology Madras, Tamil Nadu, India Dr. G. Raghavendra, National Institute of Technology Warangal, Telangana, India Dr. T. Gangadhar, National Institute of Technology Tadepalligudem, Andhra Pradesh, India Dr. U. S. Paul Russel, Air India, India Dr. V. P. Chandra Mohan, National Institute of Technology Warangal, Telangana, India Dr. P. Narsimha Reddy, SNIST, Hyderabad, Telangana, India Dr. Swami Naidu, NIT Raipur, Chhattisgarh, India Dr. Vemuri Laxmi Narayana, RIET, Andhra Pradesh, India Dr. P. H. V. Sesha Talpa Sai, MRCET, Hyderabad, India

Preface

The International Conference on Intelligent Manufacturing and Energy Sustainability (ICIMES 2019) was successfully organized by Malla Reddy College of Engineering & Technology, an UGC autonomous institution, during June 21–22, 2019, at Hyderabad. The objective of this conference was to provide opportunities for the researchers, academicians, and industry persons to interact and exchange the ideas and experience, and gain expertise in the cutting-edge technologies pertaining to Industry 4.0. Research papers were received and subjected to a rigorous peer-review process with the help of editorial board, program committee, and external reviewers. The editorial committee has finally accepted 19.5% of the manuscripts for publication in a single volume with Springer SIST series. Our sincere thanks to Chief Guest Dr. Shahrol Bin Mohammad, who is an Outstanding Scientist and Senior Lecturer, at the Faculty of Engineering in Universiti Malaysia Sarawak (UNIMAS) and to Mr. Vidya Sagar Uddagiri for the keynote session—Technology Practice Leader and Enterprise Architect at TCS, Hyderabad, India. We are indebted to the editorial board, program committee, and external reviewers who have carried out critical reviews in a short time. We would like to express our special gratitude to publication chair Dr. Suresh Chandra Satapathy, Professor, KIIT, Bhubaneswar, for his valuable support and encouragement till the successful conclusion of the conference. We express our heartfelt thanks to our Chief Patron Sri. CH. Malla Reddy, Founder Chairman, MRGI, Patrons Sri. CH. Mahendar Reddy, Secretary, MRGI, Sri. CH. Bhadra Reddy, President, MRGI, Conference Chair Dr. V. S. K. Reddy, Convener Dr. M. Murali Krishna, Organizing Chair Professor Dr. A. N. R. Reddy, Organizing Secretary Dr. Srikar Potnuru, and Coordinator Ms. K. Akhila, for their

xi

xii

Preface

valuable contribution to the successful conduct of the conference. Last but not least, our special thanks to all the authors without whom the conference would not have taken place. Their technical contributions have made our proceedings rich and praiseworthy. Hyderabad, India Mumbai, India Melbourne, Australia Krasnoyarsk, Russia Bhubaneswar, India

Dr. A. N. R. Reddy Dr. Deepak Marla Dr. Milan Simic Dr. Margarita N. Favorskaya Dr. Suresh Chandra Satapathy

Contents

Performance Evaluation of Multi-layer Barriers for Machine-Induced Low-Frequency Noise Attenuation . . . . . . . . . . . . . . . . . . . . . . . . . . . Abid Hossain Khan, Muhammed Mahbubur Razzaque and Md. Shafiqul Islam 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Mathematical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experiment Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Planing Process on AISI S-1006, S-7, and S-4340, Based on Johnson–Cook Model Using Numerical Technique . . Abhinav, D. Prajwal and Punith Kumar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Modeling Techniques and Assumptions . . . . . . . 2 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . 3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

1

. . . . . .

. . . . . .

2 3 5 6 8 9

............

11

. . . . .

. . . . .

11 12 13 16 18

.

21

.

21

. . . .

23 28 29 31

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

An Analytical Study of Diametral Error in Simultaneous Turning Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sandeep Kumar, Kalidasan Rathinam, Vivek Sharma and VaitlaSai Kumar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Formulation of Diametral Error of Workpiece with Cutting Tools kept Opposite to Each Other . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Comparison of Analytical Model with Literature . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xiii

xiv

Contents

Improvement of Electric Heater Design for Household Cooking Application in Developing Countries . . . . . . . . . . . . . . . . . . . . . Angkush Kumar Ghosh, Abid Hossain Khan and A. N. M. Mizanur Rahman 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experiment Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

......

33

. . . . .

. . . . .

33 34 37 41 41

.....

43

. . . . . . . .

. . . . . . . .

44 45 47 48 48 50 51 52

.....

53

. . . . . .

. . . . . .

53 56 57 59 63 64

.......

67

. . . . . . . . .

67 68 70 71 71 75 75 76 76

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank . K. A. Alex Luke, J. Arun, R. Hemanth Prasanna and Ashish Selokar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 HHO Module Design . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Tank Design and Development . . . . . . . . . . . . . . . . . . . . 4 Water Flow Sensor Results and Discussions . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Automated Solar Photovoltaic Panel Cleaning/Cooling System Using Air–Water Mixture and Sustainable Solutions to Off-Grid Electrification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nithin Sha Najeeb, Prashant Kumar Soori and Iyad Al Madanat 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Proposed Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Automation of Solar Panel Cleaning and Cooling System . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design and Fabrication of Four-Way Multi-hacksaw Cutting Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anupoju Sai Vamsi, Chiranjeeva Rao Seela and Arnipalli Naveen 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Fabrication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Working . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Working Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Design Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Cost Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 SWOT Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . .

. . . . . . . . .

. . . . .

. . . . . . . .

. . . . . .

. . . . . . . . .

. . . . .

. . . . . . . .

. . . . . .

. . . . . . . . .

. . . . .

. . . . . . . .

. . . . . .

. . . . . . . . .

. . . . . . . . .

Contents

xv

.........

77

. . . . .

. . . . .

77 78 79 81 81

.............

85

. . . . . .

. . . . . .

86 87 87 88 91 91

..........

93

Adhesion Strength of Plasma Sprayed Coatings—A Review Abhinav, Harish Kumar Kustagi and Arun R. Shankar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Selection Criterion in Plasma Sprayed Coatings . . . . 2 Adhesion Strength of Plasma Sprayed Coatings . . . . . . . . 3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . An Experimental Studies on the Polymer Hybrid Composites—Effect of Fibers on Characterization . . . . M. Ashok Kumar, K. Mallikarjuna, P. V. Sanjeev Kumar and P. Hari Sankar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . 2.1 Fabrication of Composites . . . . . . . . . . . . . . . . 3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

Design Analysis and Pressure Loss Optimization of Automobile Muffler . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vikram Kumar, Naresh Prasad, M. K. Paswan, Pankaj Kumar and Sanjoy Biswas 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Backpressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Effects of Increased Back Pressure . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Meshing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Assumption and Boundary Conditions . . . . . . . . . . 4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Inlet Velocity: Case 1: 39 M/S . . . . . . . . . . . . . . . 4.2 Inlet Velocity: Case 2: 44 M/S . . . . . . . . . . . . . . . 4.3 Inlet Velocity: Case 3: 49 M/S . . . . . . . . . . . . . . . 5 Conclusion and Future Work . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

Different Modules for Car Parking System Demonstrated Using Hough Transform for Smart City Development . . . Janak D. Trivedi, M. Sarada Devi and Dhara H. Dave 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Technology Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Hough Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Find Circle Using Hough Transform . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. . . . .

. . . . . .

. . . . . . . . . . . . .

. 93 . 94 . 95 . 95 . 95 . 97 . 97 . 99 . 99 . 99 . 100 . 106 . 106

. . . . . . . . . . . 109 . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

109 110 111 112 114

xvi

5 6

IP Webcam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proposed Method . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Step-by-Step Procedure . . . . . . . . . . . . . . . . . 7 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . 7.1 Circle with Different Radius—Parallel Parking 7.2 Circle with the Same Radius, Angle Parking . 8 Conclusion and Future Work . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

Effect of Arrangement and Number of Water Mist Spray Nozzles on Air Humidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pravinth Balthazar, Mohd Azmi Ismail, Andyqa Abdul Wahab, Mohammad Nazmi Nasir, Muhammad Iftishah Ramdan and Hussin Bin Mamat 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Equipment Modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

Vibration Condition Monitoring of Spur Gear Using Feature Extraction of EMD and Hilbert–Huang Transform . . . . . . . . A. Krishnakumari, M. Saravanan, M. Ramakrishnan, Sai Manikanta Ponnuri and Reddy Srinadh 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Gear Vibration Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Time–Frequency Representations . . . . . . . . . . . . . . . . . . . . 3.1 HHT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 EMD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 HSA and FFT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

114 114 115 117 117 119 120 120

. . . . 123

. . . . .

. . . . .

. . . . .

124 125 127 131 131

. . . . . . . . 133

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Comparative Study of PWM Technique for Switching Loss Reduction and Acoustic Noise Reduction in VSI-Fed Drives . . . . . . . . . . . . . . . . Tarang Kalaria, Tapankumar Trivedi, Vinod Patel, Rajendrasinh Jadeja and Chandresh Patel 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Advanced PWM Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Switching Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Acoustic Noise Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Simulation Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

134 135 135 136 138 138 140 144 144

. . 147

. . . . . . .

. . . . . . .

147 148 150 151 152 154 154

Contents

An Improved Genetic Algorithm for Production Planning and Scheduling Optimization Problem . . . . . . . . . . . . . . . . Aditya Kunapareddy and Gopichand Allaka 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Research Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Problem Modelling . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Chromosome Encoding . . . . . . . . . . . . . . . . . . . . . 2.3 Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Crossover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 Mutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Experimental Results and Discussion . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xvii

. . . . . . . . . . 157 . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

Automatic Calibration for Residential Water Meters by Using Artificial Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Edwin Pruna, Carlos Bustamante, Miguel Escudero, Santiago Mullo, Ivón Escobar and José Bucheli 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Equipment Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Instrument Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

Hardware-in-the-Loop of a Flow Plant Embedded in FPGA, for Process Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Edwin Pruna, Icler Jimenez and Ivón Escobar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Development of the HIL . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 The Flow Plant Design . . . . . . . . . . . . . . . . . . . . . . . 2.2 Implementation of the Flow Plant . . . . . . . . . . . . . . . 3 Controller Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Design of the Continuous PID Controller . . . . . . . . . . 3.2 Design of the Discrete PID Controller . . . . . . . . . . . . 4 Tests and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

157 158 158 159 159 159 162 162 163 163 163 169 170

. . . . . 173

. . . . . .

. . . . . .

. . . . . .

. . . . . .

173 175 177 179 180 180

. . . . . . . . 181 . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

181 183 183 183 184 184 185 185 185 186 188 188

xviii

Contents

Error Diagnosis in Space Navigation Integration Using Wavelet Multi-Resolution Analysis with General Regression Neural Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ramanan Gopalakrishnan, Diju Samuel Gnanadhas and Madhu Kiran Reddy Muli 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Inertial Navigation Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Prediction of Errors in GPS . . . . . . . . . . . . . . . . . . . . . . 2.2 Parameters Obtained . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Wavelet Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . 191

. . . . . . . . .

Hydromagnetic Squeeze Film Performance of Two Conducting Longitudinally Rough Elliptical Plates . . . . . . . . . . . . . . . . . . . . J. V. Adeshara, M. B. Prajapati, G. M. Deheri and R. M. Patel 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

Performance and Emission Characteristics of Biodiesel from Rapeseed and Soybean in CI Engine . . . . . . . . . . R. Udayakumar, Vivek J. Shah and Sai Vijay Venkatesh 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . 4.1 Brake Power . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Specific Fuel Consumption . . . . . . . . . . . . . . . 4.3 Brake Thermal Efficiency . . . . . . . . . . . . . . . . 4.4 NOx Emission . . . . . . . . . . . . . . . . . . . . . . . . . 4.5 CO2 Emission . . . . . . . . . . . . . . . . . . . . . . . . . 4.6 Particulate Matter Emission . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

191 192 193 193 193 194 195 198 198

. . . . . . 199 . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

200 202 204 207 208

. . . . . . . . . . . . . 209 . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

A Practical Approach to Camera Calibration for Part Alignment for Hybrid Additive Manufacturing Using Computer Vision . . . . . Pallavi Kulkarni, Atul Magikar and Tejas Pendse 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Problem Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Camera Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

209 210 211 212 213 214 215 216 217 218 218 219

. . . . 221 . . . . 221 . . . . 222 . . . . 223

Contents

2 3

Literature Review . Experimentation . . 3.1 Setup . . . . . . 3.2 Results . . . . . 3.3 Result Table . 4 Conclusion . . . . . . 5 Future Work . . . . . References . . . . . . . . . .

xix

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

Development of a Colour and Orientation Detection System for Small Part Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aditya Sachdeva, Rohan Kapoor, Awwal Singh and Pradeep Khanna 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Colour Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Infrared Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Servo-Based Actuator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Microcontroller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Description of Algorithm Used in Code . . . . . . . . . . . . . . . . . . . 3.1 Mean of a Set of 10 Readings of Colour Sensor to Reduce the Error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Identification of the Colour from the Mean RGB Values . . . 3.3 Taking the Mean of 10 Infrared Sensor Values for More Accuracy and Sensing the Orientation of the Parts . . . . . . . 3.4 Condition for Rejection of the Undesirable Part by the Actuator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Working . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

Comparative Analysis on Battery Used in Solar Refrigerated E-Rickshaw in India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Surender Kumar and Rabinder Singh Bharj 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Smart Mobility System for Indian Megacities . . . . . . . . . . . . . . 3 BMS Duty for Li-Ion Battery Use in E-Rickshaw and SHRERs 4 The Relation Between State of Function with SOC and SOH . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

223 225 225 226 227 228 228 228

. . . . 231 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

231 233 233 234 234 234 234

. . . . 234 . . . . 235 . . . . 236 . . . .

. . . .

. . . .

236 237 237 237

. . . . . 239 . . . . . .

. . . . . .

. . . . . .

. . . . . .

239 241 242 243 244 248

Application of Value Stream Mapping (VSM) in a Sewing Line for Improving Overall Equipment Effectiveness (OEE): A Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Shawkat Imam Shakil and Mahmud Parvez 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 1.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

xx

1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Case Studies . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Conceptual Work . . . . . . . . . . . . . . . . . . . . . 2.3 Modeling Work . . . . . . . . . . . . . . . . . . . . . . 2.4 Survey Articles . . . . . . . . . . . . . . . . . . . . . . . 3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Current State VSM . . . . . . . . . . . . . . . . . . . . 3.2 Rough Set Theory . . . . . . . . . . . . . . . . . . . . . 3.3 Future State VSM . . . . . . . . . . . . . . . . . . . . . 4 OEE and Other Performance Parameters . . . . . . . . 4.1 Formula Used . . . . . . . . . . . . . . . . . . . . . . . . 4.2 OEE Calculation . . . . . . . . . . . . . . . . . . . . . . 4.3 Existing Scenario . . . . . . . . . . . . . . . . . . . . . 4.4 Proposed Scenario . . . . . . . . . . . . . . . . . . . . 4.5 Calculation of Other Performance Parameters . 5 Result and Conclusion . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Contents

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

CFD Analysis to Increase Heat Transfer in Pipe with Combined Effect of Circular Perforated Ring and Twisted Tape Insert . . . . . . . . Prashant Negi, Vikash Kumar Gupta, Anil K. Prasad and Inzamam Ahmad 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Numerical Model Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Geometrical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Governing Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

Study on Solar Parabolic Trough Collector with Different Copper Absorber Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Hari Haran and P. Venkataramaiah 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Absorber Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 WASPAS Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

250 250 250 251 251 251 252 252 252 255 255 255 257 257 257 258 259 259

. 261

261 262 262 264 265 267 269

. . . . 271 . . . . . . . .

. . . . . . . .

. . . . . . . .

271 272 272 274 274 277 277 278

Contents

xxi

Harmonic Analysis for Bidirectional Grid-Connected Converter for Electrical Vehicle During Charging and Discharging Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chitrang Vyas, Amit Ved, Tapankumar Trivedi and Rajendrasinh Jadeja 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Grid-Connected Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . 279 . . . . .

. . . . .

280 281 283 286 287

Effect of Tribo-layer on the Sliding Wear Behavior of Detonation Sprayed Alumina–Titania Coatings . . . . . . . . . . . . . . . . . . . . . . . . P. Uday Chandra Rao, P. Suresh Babu, D. Srinivasa Rao, S. V. Gopala Krishna and K. Venkateswara Rao 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Sample Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Powder Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Process Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Coating Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Surface Roughness, Microstructure and Porosity Values of Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 XRD Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 MicroHardness Values . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Sliding Wear Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Coefficient of Friction Values . . . . . . . . . . . . . . . . . . . . . . 3.6 Wear Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7 EDS Analysis of Worn Surfaces . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

289 291 291 291 291 291 292

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

292 293 293 294 295 295 296 297 298

Development of Sustainable Cementitious Binder Utilizing Silicomanganese Fumes . . . . . . . . . . . . . . . . . . . . . . . . . . . Syed Khaja Najamuddin, Megat Azmi Megat Johari, Mohammed Maslehuddin and Moruf Olalekan Yusuf 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology of Research . . . . . . . . . . . . . . . . . . . . . . . 2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Experimental Program . . . . . . . . . . . . . . . . . . . . . . 2.3 Evaluation Methods . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Setting Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Compressive Strength . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . 289

. . . . . . . . . . 299

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

300 301 301 302 303 304 304 305 305

xxii

Contents

4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308 Review on Sliding Wear of Ti–6Al–4V Alloy Concerning and Sliding Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . K. Suneel, N. Nagmohan Rao, R. Balaji, N. Srikanth, Gnanadurai Ravikumar Solomon and Ashish Selokar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Microstructures . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Coatings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Heat and Surface Treatment . . . . . . . . . . . . . . . 2.4 Ti–6Al–4V Against Same and Different Alloys . 3 Discussion and Result . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion and Future Scope of Work . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Counterface . . . . . . . . . . . . 309

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Hardness and Microstructure Studies on the Effect of Solution-Treated Age-Hardened Al-6082 Alloy . . . . . . . . . . . . C. B. Shashikumar, K. R. Harish and Raghavendra P. Nilugal 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Material Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Heat Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Hardness Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Hardness and Microstructure Results of Experiment No. 1 5.2 Hardness and Microstructure Results of Experiment No. 2 5.3 Hardness and Microstructure Results of Experiment No. 3 5.4 Hardness and Microstructure Results of Experiment No. 4 5.5 Hardness and Microstructure Results of Experiment No. 5 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Floating Photovoltaic Thin Film Technology—A Review R. Nagananthini, R. Nagavinothini and P. Balamurugan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Floating photovoltaic (FPV) Technology . . . . . . . . . . 2.1 Design Requirements . . . . . . . . . . . . . . . . . . . . 2.2 Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Thin Film Technology . . . . . . . . . . . . . . . . . . . . . . . 3.1 Commercial Thin Film Technologies in Practice . 3.2 Future Thin Film Technologies . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

310 310 310 311 311 312 314 317 317

. . . . . 319 . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

319 320 320 320 320 322 322 324 325 326 327 327

. . . . . . . . . . . . 329 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

329 331 331 333 334 335 337 337 337

Contents

xxiii

Effect of Ethanol Fumigation on Performance and Combustion Characteristics of Compression Ignition Engine Fuelled with Used Cooking Oil Methyl Ester in Dual-Fuel Mode . . . . . . . . . Reddy Srinadh, Velmurugan Ramanathan, Mayakrishnan Jaikumar, Raja Selvakumar, V. A. Shridhar, E. Sangeethkumar and N. Sasikumar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Present Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Trans-esterification of UCO . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experimental Set-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Details of Ethanol Fumigation Rate . . . . . . . . . . . . . . . . . . . 3.2 Details of Energy Sharing of Ethanol . . . . . . . . . . . . . . . . . . 4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Effect of Ethanol on Engine Performance . . . . . . . . . . . . . . . 4.2 Effect of Ethanol on Combustion Analysis . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Estimation and Moderation of Harmonics in Distribution Systems J. Viswanatha Rao and G. Lakshminarayana 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Estimation of Harmonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Mitigation of Harmonics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Harmonic Reduction Using Pulse Converters . . . . . . . . . . . 4 Proposed Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Result and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Theoretical Evaluation of Energy Performance of a Vapour Compression Refrigeration System Using Sustainable Refrigerants Sharmas Vali Shaik and T. P. Ashok Babu 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Development of Sustainable Alternative Refrigerants . . . . . 3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Thermodynamic Analysis of Alternative Refrigerants . . . . . 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Energy Efficiency Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Discharge Temperature of Compressor . . . . . . . . . . . . . . . . 4.3 Power Spent Per Ton of Refrigeration . . . . . . . . . . . . . . . . 4.4 Volumetric Refrigeration Capacity . . . . . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . 339

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

340 341 342 342 342 342 343 344 346 346 348 350 351

. . . . 353 . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

353 354 355 355 355 356 359 360

. . . . 361 . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

361 362 362 362 362 366 366 366 367 368 369 370

xxiv

Effect of Autofrettage on the Properties of the Aluminium Cylinders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neelkant Patil, Shivarudraiah, D. Amaresh Kumar and Kalmeshwar Ullegaddi 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

. . . . . . . . . . 371

. . . . .

. . . . .

Influence of Graphene and Tungsten Carbide Reinforcement on Tensile and Flexural Strength of Glass Fibre Epoxy Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kalmeshwar Ullegaddi, C. R. Mahesha and Shivarudraiah 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Experimental Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Tensile Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Flexural Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Effect of Fibre Loading on Tensile Properties of Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Effect of Fibre Loading on Flexural Properties of Composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

371 372 373 376 377

. . . . . . . . 379 . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

379 380 381 382 382 383 384 384

. . . . . . . . 384 . . . . . . . . 386 . . . . . . . . 387 . . . . . . . . 388

Investigation of Partial Discharge Due to Copper Spherical Particle in Power Transformer Under Various Oil Flow Models Using CFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . N. Vasantha Gowri 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Analysis Under Non-directed Oil Flow System . . . . . . . . . . . 2.2 Analysis of Directed Oil Flow System . . . . . . . . . . . . . . . . . 3 Discussions and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . 391 . . . . . .

. . . . . .

. . . . . .

391 392 393 396 399 399

A High-Power High-Frequency Isolated DC Power Supply for Electric Vehicle Charging Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401 Shivam Joshi, Rajesh Patel, Vinod Patel, Tapankumar Trivedi and Pavak Mistry 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

Contents

2

Proposed Topology of EV Power Supply . 2.1 T-type Converter Topology . . . . . . . 3 Charge Controller . . . . . . . . . . . . . . . . . . 4 Efficiency Calculation . . . . . . . . . . . . . . . 5 Simulation Results and Discussions . . . . . 5.1 Dynamic Load Test . . . . . . . . . . . . 5.2 Charging of Battery Pack . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . .

xxv

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Effect of Slicing Thickness and Increment on the Design of Patient Specific Implant for Total Knee Replacement (TKR) Using Magnetic Resonance Imaging (MRI)—A Case Study . . . . . . . . . . . . . . . . . . . . . Y. Sandeep Kumar, K. V. S. Rajeswara Rao, Sunil R. Yalamalle, S. M. Venugopal and Sandeep Krishna 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Creating a 3D Model of Knee . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Optimization of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

403 403 404 405 406 407 408 409 410

. . 411

. . . . . . .

Microstructure and Hardness Behaviour Study of Carbon Nanotube in Aluminium Nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prashant S. Hatti, K. Narasimha Murthy and Anupama B. Somanakatti 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Synthesizing of Al–Cnt Metal Matrix Composites . . . . . . . . . . . . . 4 Microstructural Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Hardness Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

Comparison of Electric Fields with and Without Corona Ring for 66 kV Line Insulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Prapti Jethva, Krishna Patel and Dinesh Kumar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Numerical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Calculation of Electric Field . . . . . . . . . . . . . . . . . . . . 3 Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Properties of Different Materials . . . . . . . . . . . . . . . . . 3.2 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

411 412 412 414 415 415 418

. . 421 . . . . . . . .

421 423 424 424 426 426 427 428

. . . . . . . 429 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

429 431 431 431 432 432 433

xxvi

4

Result and Discussion . . . . 4.1 Without Corona Ring 4.2 With Corona Ring . . 5 Conclusion . . . . . . . . . . . . References . . . . . . . . . . . . . . . .

Contents

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

Experimental and FEA Simulation of Thermal-Fluid Interaction Between TIN Coated Tungsten Carbide Tool and Inconel-825 Workpiece . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M. Sivaramakrishnaiah, P. Nandakumar and G. Rangajanardhana 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Design of Experiment and Experimentation . . . . . . . . . . . . . . . 3 Multiple Regression Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Geometry and Mesh Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Dry and Wet Condition . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Dry and Wet Cutting Condition . . . . . . . . . . . . . . . . . . . . 6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

The Effect of Recrystallization on Electrical Resistivity of Stir Casted SiCP/AA6061 Composite After Shot Peening . . . . . . . . Venumurali Jagannati and Bhanodaya Reddy Gaddam 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Microstructural Examination . . . . . . . . . . . . . . . . . . . . 3.2 XRD Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Electrical Resistivity . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

Design and Modeling of Wye Piece . . . . . . . . . . . . . . . . Syed Sha Khalid, G. Vijaya and M. S. Rajagopal 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Modelling Using CATIA and Hypermesh . . . . . . . . . . 3 Meshing and Boundary Conditions Using Hypermesh . 4 Analysis Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Future Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . .

. . . . .

. . . . .

. . . . .

434 434 435 438 439

. . . . . 441 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

442 443 444 446 446 447 447 449 451

. . . . . . . 453 . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

453 454 455 455 456 457 459 460

. . . . . . . . . . . . 461 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

461 462 463 464 470 470 470

Contents

Comparative Investigation on Gas Tungsten Arc Welding and Friction Stir Welding of Electrolytic Tough Pitch Copper Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . K. A. Alex Luke, V. Balaji Reddy and Ashish Selokar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion and Further Scope . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xxvii

. . . . . . . . . . 473 . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

Synthesis and Characterization of Functionally Graded Ceramic Material for Aerospace Applications . . . . . . . . . . . . . . . . . . . . . . M. Jeshrun Shalem, A. Devaraju and K. Karthik 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Slip Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Dielectric Constant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Compressive Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Scanning Electron Microscope (SEM) Analysis . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

473 474 474 477 480 480

. . . . . 483 . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

CI Engine Characteristic Investigation by Application of Metal-Based Additives with Biodiesel Blends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . L. Bharath and D. K. Ramesha 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Production of Biodiesel from Waste Cooking Oil (WCO) . . . . 3.2 Determination of Fuel Properties . . . . . . . . . . . . . . . . . . . . . . 3.3 Preparation of Nanofluids . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Performance Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Combustion Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Emission Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Oxides of Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Carbon Monoxide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Unburnt Hydrocarbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Smoke Opacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

483 484 484 484 484 486 486 486 487 487

. . 489 . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

489 490 491 491 491 492 492 492 493 493 493 495 495 497 497 498

xxviii

Simulation of Underground Cable Defects with the Detection of Partial Discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hardikkumar Delvadiya, Dinesh Kumar and Krishna Patel 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Cable Defects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 What Is Cable Defects? . . . . . . . . . . . . . . . . . . . . . . . 2.2 Cause of Cable Defects . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Void . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 External Cut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Termination and Left Out Completely . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

. . . . . . . . 499 . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

Comparison of C.I Engine Performance Parameters and Emissions by Varying Designs of Intake Manifolds . . . . . . . . . . . . . . . . . . . . . N. Balaji Ganesh and P. V. Srihari 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Set-Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Engine Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Designs of Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Normal Manifold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Nozzle Inside the Inlet Manifold . . . . . . . . . . . . . . . . . . . . . 3.3 Internal Threaded Manifold . . . . . . . . . . . . . . . . . . . . . . . . . 4 Tabular Columns for Experimental Results . . . . . . . . . . . . . . . . . . 5 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

Variation of Time Lag, Decrement Factor and Inside Surface Temperature with Solar Optical Properties of Building Envelope in Different Climatic Zones of India . . . . . . . . . . . . . . . . . . . . . . . Debasish Mahapatra and T. P. Ashok Babu 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Time Lag and Decrement Factor . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Sole Air Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Climatic Zones of India . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Time Lag and Decrement Factor Calculation . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Time Lag and Decrement Factor Calculation . . . . . . . . . . 3.2 Electricity Saving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

499 501 501 501 501 502 503 504 506 507

. . . 509 . . . . . . . . . . .

. . . . . . . . . . .

509 510 511 511 512 512 512 513 513 520 522

. . . . . 523 . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

523 524 525 525 526 526 527 527 530 531 532

Contents

xxix

Study on Tensile and Hardness Properties of Aluminium 7075 Alloy Reinforced with Graphite, Mica and E-Glass . . . . . . . . . . . T. G. Gangadhar, D. P. Girish, A. C. Prapul Chandra, Gangadhar Angadi and K. V. Karthik Raj 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Hardness Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Tensile Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Result and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Tensile Strength and Hardness . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Signal-to-Noise Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Analysis of Variance (ANOVA) . . . . . . . . . . . . . . . . . . . . . . . 3.4 Microstructure Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

ANN-Based Fault Classification and Section Identification Technique Using Superimposed Currents for Three-Terminal Transmission Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shivani Vaghela, Nishant Kothari and Dinesh Kumar 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 System Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Feature Selection for Section Identification and Fault Classification 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . .

An Analytical Investigation for Combined Pressure-Driven and Electroosmotic Flow Without the Debye–Huckel Approximation . Avisankha Dutta and Sudip Simlandi 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Description of the Physical Problem . . . . . . . . . . . . . . . . . . . . 3 Mathematical Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Poisson–Boltzmann Equation . . . . . . . . . . . . . . . . . . . . . . 3.2 Momentum Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Energy Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Solution of the Governing Equations . . . . . . . . . . . . . . . . . . . . 4.1 Electric Potential Distribution . . . . . . . . . . . . . . . . . . . . . 4.2 Velocity Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Temperature Distribution . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

. 533

534 534 534 535 536 536 537 537 538 539 540 541 541

. . . 543 . . . . . .

543 544 546 547 548 552

. . . . . 553 . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

553 555 555 556 556 556 557 557 559 560 561 563 563

xxx

Contents

Addressing the Green Tribology Advancement, Future Development, and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Surender Kumar and Rabinder Singh Bharj 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Global Sustainable Development . . . . . . . . . . . . . . . . . . . . . . 2 Green Tribology Fields in Engineering . . . . . . . . . . . . . . . . . . . . . . 3 Utility of Green Tribology in Real Life . . . . . . . . . . . . . . . . . . . . . 4 Green Tribology Development Direction . . . . . . . . . . . . . . . . . . . . 5 Current Status of Energy Loss Due to Friction and Wear Worldwide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . 565 . . . . .

. . . . .

565 567 567 569 569

. . . .

. . . .

570 570 572 573

Effect of Drying Temperature to the Thin Layer Drying Model of Sago Starch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maswida Mustafa Kamal, Rubiyah Baini, Lim Soh Fong, Mohd Hasnain Md Hussain and Shahrol Mohamaddan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Sample Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Drying Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Data Analysis and Mathematical Modeling . . . . . . . . . . . . 3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Effect of Drying Temperature to the Drying Curve of Traditionally Processed Sago Starch and Drying Characteristic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Effect of Drying Temperature to the Fitting Drying Model of Traditionally Processed Sago Starch . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . 579 . . . . . 579 . . . . . 582

Evaluating Mechanical Properties of Egg Shell, and Coco Peat Reinforced Epoxy Composite . . . . . . . . . . . . . . . . . . . . . . . . . . . Vijay Kumar Girisala, D. Mangeelal and Sunkara Jaya Kishore 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Preparation of Composite Material . . . . . . . . . . . . . . . . . 2.2 Testing of Composite Material . . . . . . . . . . . . . . . . . . . . 3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Tensile Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Compressive Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Hardness Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . 575

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

575 576 576 576 576 577

. . . . . 577

. . . . . . 583 . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

584 585 585 587 588 588 589 590 590 591

Contents

Design Modification of Rear Axle Housing by Fatigue Failure Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Barada P. Baisakh and Anil K. Prasad 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 CAD and FE Model Description . . . . . . . . . . . . . . . 3 Axle Housing Material . . . . . . . . . . . . . . . . . . . . . . 4 Loading Condition . . . . . . . . . . . . . . . . . . . . . . . . . 5 Finite Element Analysis of Housing and Results . . . . 6 Theoretical Validation . . . . . . . . . . . . . . . . . . . . . . . 7 Design Optimization . . . . . . . . . . . . . . . . . . . . . . . . 8 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . 9 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xxxi

. . . . . . . . . . . . . 593 . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

Mixture Design Using Low-Cost Adsorbent Materials for Decolourisation of Biomethanated Distillery Spent Wash in Continuous Packed Bed Column . . . . . . . . . . . . . . . . . . . . Ishwar Chandra, Anima Upadhyay and N. Ramesh 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Collection and Bio-chemical Analysis of BDSW . . . . 2.2 Preparation and Characterisation of Biosorbent Carrier Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Batch Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Continuous Column Flow Studies . . . . . . . . . . . . . . . 2.5 Decolourisation Assay . . . . . . . . . . . . . . . . . . . . . . . . 2.6 Nutritional Analysis of Spent Modified Soil . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Characteristics of BDSW . . . . . . . . . . . . . . . . . . . . . . 3.2 Properties of Biosorbent . . . . . . . . . . . . . . . . . . . . . . 3.3 Colour Degradation . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Nutritional Quality of Spent Modified Soil . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

Prediction of the WPPO Biodiesel-Fuelled HCCI Engine Using Artificial Neural Networks . . . . . . . . . . . . . . . . . . . Ramavathu Jyothu Naik and Kota Thirupathi Reddy 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Modelling with ANN . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

593 594 595 597 597 599 600 602 603 605

. . . . . . . . 607 . . . . . . . . 608 . . . . . . . . 610 . . . . . . . . 610 . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

610 611 611 613 613 614 614 614 616 617 617 620

. . . . . . . . . . . 623 . . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

623 624 625 629 631 631

xxxii

Contents

Dimensional Analysis of Form Drilling Parameters by Buckingham Pi Theorem and Optimization of Heat Generation in Form Drilling Process by Taguchi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Y. Bhargavi and V. Diwakar Reddy 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Pi Terms Evaluated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Process Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Analysis of Variance for Heat Generation . . . . . . . . . . . . . . . 5.2 Analysis of Variance for Petal Height . . . . . . . . . . . . . . . . . . 5.3 Taguchi Analysis for Heat Generation . . . . . . . . . . . . . . . . . . 5.4 Taguchi Analysis for Petal Height . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

Comparison of Ductile, Flexural, Impact and Hardness Attributes of Sisal Fiber-Reinforced Polyester Composites . . . . . . . . . . . . . . . . S. Sathees Kumar, V. Mugesh Raja, B. Sridhar Babu and K. Tirupathi 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Fabrication of Specimen . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Mechanical Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Tensile Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Flexural Strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Impact Strength and Hardness . . . . . . . . . . . . . . . . . . . . . . . 3 Scanning Electron Microscope (SEM) . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Ductile Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Flexural Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Impact Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Hardness Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Analysis of SEM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

. . 633 . . . . . . . . . . . .

633 634 635 636 636 637 639 639 640 640 641 643

. . . 645 . . . . . . . . . . . . . . . .

Optimization of EDM Process Parameters Using Standard Deviation and Multi-objective Optimization on the Basis of Simple Ratio Analysis (MOOSRA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . J. Anitha and Raja Das 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Optimization Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . .

645 646 647 649 649 649 649 649 650 650 650 651 652 652 653 654

. . 655 . . . .

. . . .

655 656 657 657

Contents

xxxiii

3.2 Standard Deviation Concept for Weight Calculation 3.3 MOOSRA Methodology . . . . . . . . . . . . . . . . . . . . 4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Allocation of Weights . . . . . . . . . . . . . . . . . . . . . . 4.2 Result Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

Modeling and Simulation of Trans Z-Source Inverter with Maximum Constant Boost PWM Control for Solar PV System . . . . . . . . . . . . . K Chitra 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 T-Source Inverter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Equivalent Circuit and Operating States of TSI . . . . . . . . . . . . 3 Modeling of TSI-Based Solar PV System . . . . . . . . . . . . . . . . . . . . 4 Maximum Constant Boost PWM Control (MCB PWM) . . . . . . . . . 5 Simulation Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Simulation Results of TSI for Solar PV System Using MCB PWM Algorithm . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Driver Drowsiness Monitoring System . . . . . . J. V. V. S. N. Raju, P. Rakesh and N. Neelima 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Survey . . . . . . . . . . . . . . . . . . . 3 Proposed System . . . . . . . . . . . . . . . . . . . 3.1 Eye Blink Detection . . . . . . . . . . . . . 3.2 Yawn Detection . . . . . . . . . . . . . . . . 3.3 Head Location Analysis . . . . . . . . . . 4 Result . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

657 658 659 659 660 661 662

. . 663 . . . . . .

. . . . . .

663 664 664 666 668 669

. . 670 . . 672 . . 673

. . . . . . . . . . . . . . . . . . . . 675 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Detection and Control of Water Leakage in Pipelines and Taps Using Arduino Nano Microcontroller . . . . . . . . . . . . . . . . . . . . . M. Pravin Kumar, R. Velmurugan and P. Balakrishnan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Arduino Nano . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Water Flow Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Water Drop Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Liquid Crystal Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Relay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Solenoid Valve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Experimental Set-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

675 676 677 679 681 682 682 683 683

. . . . . . 685 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

685 686 687 688 688 689 690 690 691

xxxiv

Contents

10 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 691 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 692 Smart Automated Processes for Bottle-Filling Industry Using PLC-SCADA System . . . . . . . . . . . . . . . . . . . . . . . Nikhita Nadgauda, Senthil Arumugam Muthukumaraswamy and S. U. Prabha 1 Introduction and Background Theory . . . . . . . . . . . . . . 2 Proposed Methodology . . . . . . . . . . . . . . . . . . . . . . . . 2.1 SCADA Implementation of Bottle-Filling Industry 2.2 Warehouse Management Systems . . . . . . . . . . . . 2.3 Energy-Efficient Lighting Solutions . . . . . . . . . . . 3 Results and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 SCADA Implementation of Bottle-Filling Industry 3.2 Warehouse Management Systems . . . . . . . . . . . . 3.3 Energy-Efficient Lighting Solutions . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . 693

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

Effect of Stirring Speed During Casting on Mechanical Properties of Al–Si Based MMCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ajit Kumar Senapati, Sasank Shekhar Panda, Bibhash Kumar Dutta and Shubham Mishra 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Experimental Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Impact Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Tensile Strength Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Compression Strength Test . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Hardness Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

Exploration of Pillars of Industry 4.0 Using Latent Semantic Analysis Technique . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Aniruddha Anil Wagire, Ajay Pal Singh Rathore and Rakesh Jain 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Latent Semantic Analysis (LSA) . . . . . . . . . . . . . . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Data Selection and Analysis . . . . . . . . . . . . . . . . . . . . 3.2 Interpretation of Results . . . . . . . . . . . . . . . . . . . . . . . 3.3 Pillars of Industry 4.0 . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusions and Implications . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

693 695 695 696 697 697 697 698 700 701 702

. . . . 703

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

703 704 704 705 706 707 707 707 709 710

. . . . . . . 711 . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

711 713 713 713 714 714 715 718 718

Contents

xxxv

An Overview of Different Topologies of DC/DC Bidirectional Converter for Different Applications . . . . . . . . . . . . . . . . . . . S. Sathishkumar, G. Preethi and V. Kamatchi Kannan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Overview of Different Topologies . . . . . . . . . . . . . . . . . . . 3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . 721 . . . .

. . . .

. . . .

. . . .

An Approach to Detect and Classify Defects in Cantilever Beams Using Dynamic Mode Decomposition and Machine Learning . . . Kailash Nagarajan, J Ananthu, Vijay Krishna Menon, K. P. Soman, E. A. Gopalakrishnan and Ajith Ramesh 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Simulation, Data Acquisition, and Analysis . . . . . . . . . . . 2.2 Dynamic Mode Decomposition . . . . . . . . . . . . . . . . . . . . 2.3 Support Vector Machines . . . . . . . . . . . . . . . . . . . . . . . . 3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

Musculoskeletal Simulation and Analysis of Upper Limb Rehabilitation Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shahrol Mohamaddan, Aliff Rahman, Annisa Jamali, Helmy Hazmi and Michelle Maya Limbai 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Upper Limb Rehabilitation Device Selection and Setting . 2.2 Musculoskeletal Analysis using AnyBody Software . . . . 3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

Fuzzy Regenerative Braking Strategy . . . . . . . Pranvat Singh Dang and Rudraksh Raajesh Haran 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 Vehicle Dynamics and Braking . . . . . . . . . . 3 Mathematical Model of Energy Regeneration 4 Braking Strategy . . . . . . . . . . . . . . . . . . . . . 5 Design of Fuzzy Logic Controller . . . . . . . . 6 Simulation Results and Discussions . . . . . . . 7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

. . . .

. . . .

. . . .

721 722 730 730

. . . . . 731

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

732 732 732 733 735 735 736 737

. . . . . . 739

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

739 740 740 742 743 746 748

. . . . . . . . . . . . . . . . . . . 749 . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

749 750 751 752 753 755 756 757

xxxvi

Contents

. . . . . . . . 759

A New Framework for Secure Outsourcing of Medical Data . J. Hyma, Moulica Sudamalla, Dharma Teja Vanaparthi, Koushik Vinnakota and Vamsi Krishna Choppavarapu 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Paper Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Proposed System Architecture . . . . . . . . . . . . . . . . . . . . . . 5 Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Least Significant Bit . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 BlowFish Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 SHA-256 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Security Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Conclusion and Future Work . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

A Comprehensive View for Providing the Decision on Medicare Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . P. Naga Jyothi, D. Rajya Lakshmi and K. V. S. N. Rama Rao 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Recommended System for Providing the Decision on Medicare Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Sources of Healthcare Claims Data . . . . . . . . . . . . . 3.2 Analysis of Medicare Data . . . . . . . . . . . . . . . . . . . 3.3 Development Phase . . . . . . . . . . . . . . . . . . . . . . . . . 4 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion and Future Work . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

760 760 760 761 762 762 763 764 764 764 766 766

. . . . . . . . . 769 . . . . . . . . . 769 . . . . . . . . . 770 . . . . . . .

Predicting Cardiac Arrhythmia Using QRS Detection and Multilayer Perceptron . . . . . . . . . . . . . . . . . . . . . . . . . . . Harika Gundala, Mayank Sethia, Mehul Sethia, Shreyas Gonjari, Akshay Gugale and Rajeshkannan Regunathan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Prediction System Architecture . . . . . . . . . . . . . . . . . . . . . 3.1 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Multilayer Perceptron . . . . . . . . . . . . . . . . . . . . . . . . 3.3 QRS Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Pan–Tompkins Algorithm . . . . . . . . . . . . . . . . . . . . . 4.2 Multilayer Perceptron . . . . . . . . . . . . . . . . . . . . . . . . 5 Dataset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

772 772 772 773 775 778 779

. . . . . . . . 781

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

781 782 783 783 783 784 785 785 786 786 786

Contents

xxxvii

7 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 788 8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 788 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 789 Full Length Driver Drowsiness Detection Model—Utilising Driver Specific Judging Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sree Pradeep Kumar Relangi, Mutyam Nilesh, Kintali Pavan Kumar and Anantapalli Naveen 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Facial Landmarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Eye Aspect Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Mouth Aspect Ratio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Drowsiness Predication . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Enterprise Reporting Solution on Integrating Business Intelligence for Operational and Financial Data . . . . . . . . . . . . . . . . . . . . . . . . Anuraag Mazumdar and Rajeshkannan Regunathan 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Survey of Existing Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Architecture of Analytical Reporting Using Business Intelligence 3.1 Detailed Design, ER (Entity Relationship) Diagram . . . . . . 4 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Survey on Accelerating the Classifier Training Using Various Boosting Schemes Within Cascades of Boosted Ensembles . . . . . . . A. S. Venkata Praneel, T. Srinivasa Rao and M. RamaKrishna Murty 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Boosting Chain Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Booster Cascade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Cascade Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Tuning the Boosting Cascade . . . . . . . . . . . . . . . . . . . . . . 2.4 Learning the Naive Booster Chain Framework . . . . . . . . . . 2.5 Using Bootstrap for Boosting Chain Learning . . . . . . . . . . 3 Optimization of Boosting Chain . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Booster Optimization Based on a Linear Model . . . . . . . . . 3.2 Adjusting the Classifier . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Removal of Redundancy in Boosting . . . . . . . . . . . . . . . . .

. . . . 791

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

791 792 793 793 793 795 796 796 797 798

. . . . 799 . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

799 800 801 803 804 805 806

. . . . 809 . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

810 811 812 812 813 814 815 817 818 819 820

xxxviii

Contents

4 Performance Comparisons of the Three Detectors . 5 Observations of Empirical Outcomes . . . . . . . . . . 6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

Service Layer Security Architecture for IOT Using Authentication and Cryptography Technique . . . . . Santosh Kumar Sharma and Bonomali Khuntia 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 System Architecture . . . . . . . . . . . . . . . . . 2.2 Algorithm (Honey Encryption) . . . . . . . . . 3 AVISPA Code . . . . . . . . . . . . . . . . . . . . . . . . . 4 Result Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 5 Conclusion and Future Work . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

821 821 824 824

Biometric . . . . . . . . . . . . . . . . 827 . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

827 828 829 830 832 834 835 835

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 839

About the Editors

Dr. A. N. R. Reddy Professor of Mechanical Engineering and Head of the Department, obtained his B.Tech. and M.Tech. from JNT University, Hyderabad, India and Ph.D. from Universiti Malaysia Sarawak, Malaysia. Dr. A. N. R. has received many prestigious research innovation and academic excellence awards that include Malaysian Government International Student Award ‘Malaysian International Scholarship’; Zamalah Graduate Scholarship from UNIMAS; Two Silver medals and one best Paper Presentation Award for in State-of-the-art for research innovations in biofuels and nano catalysts. Dr. Reddy’s research interests include Bioenergy, Pyrolysis of Biomass, Synthesis of Nano materials, Engine Performance & Exhaust gas Analystics, Applied & Fluid mechanics, Spectrophotometry, Modelling, Optimization and Design of experiments and TRIZ. Dr. Reddy, as a principle investigator, has successfully completed one AICTE, Govt. of India sponsored research project entitled “Multi Objective Optimization of Production Process Parameters using Evolutionary Algorithms.” Dr. A. N. R. is a life member of renowned professional associations such as Operational Research Society of India (ORSI), The Indian Society of Theoretical and Applied Mechanics (ISTAM), Indian Association for Computational Mechanics (IndACM), Indian Society for Technical Education (ISTE), Engineers Without Borders (EWB), and Member of Society of Automotive Engineers India (SAE India), International Society for Structural and Multidisciplinary Optimization (ISSMO). Over 19 years of service in both academic and research domains, Dr. ANR has published 24 research works in ISI and Scopus indexed journals, to name a few ACS Energy Fuels; AIP J. Renew. Sustain. Energy; Journal of chemistry etc. as well as international conferences and guided many PG and UG students’ projects. Further, Dr Reddy is very actively involved in organizing in Seminars/Conferences, Faculty development Programmes and Workshops of National and International repute for the benefit of academia. Dr. Deepak Marla is currently working as an Assistant Professor in the Department of Mechanical Engineering at the Indian Institute of Technology Bombay (IIT Bombay). He has obtained Ph.D from IIT Bombay and had done his post-doctoral xxxix

xl

About the Editors

work from the Technical University of Denmark & University of Illinois at Urbana-Champaign. His work is in the domain of micro/nano-manufacturing using advanced techniques that involve lasers, electric discharges, electrochemical reactions, plasmas, and micro-tools. His research focuses on gaining a fundamental insight into these processes through a synergetic use of multi-physics modeling and simulation, and experiments with an eye on addressing critical challenges at the process level. Dr. Milan Simic has PhD, Master and Bachelor in Electronic Engineering, as well as Graduate diploma in Education. He has comprehensive experience from industry (Honeywell Information Systems), Research institute and academia, from overseas and Australia. Dr Simic is conducting research and teaching in Mechatronics that is a comprehensive engineering discipline, which includes the following scientific areas in his work physical systems modelling, autonomous systems, biomedical engineering, robotics, intelligent transportation systems, green energy, and information theory, wireless transfer of energy, engineering management, and education. Dr. Simic is currently the General Editor for Knowledge Engineering Systems (KES) International Journal, Program Manager for Master of Engineering (Management), Associate Director of Australia-India Research Centre for Automation Software Engineering, Associate Editor for Intelligent Decision Technologies (IDT) International Journal, and few more. Dr. Margarita N. Favorskaya received her engineering diploma from Rybinsk State Aviation Technological University, Russia, in 1980 and was awarded a Ph.D. by S.-Petersburg State University of Aerospace Instrumentation, S.-Petersburg, in 1985. Since 1986 she is working in the Siberian State Aerospace University, Krasnoyarsk, in which she is responsible for the Digital Image and Videos Processing Laboratory. Dr. Margarita is a Full Professor and the Head of Department of Informatics and Computer Techniques, Siberian State University of Science & Technology. Her main research interests are in the areas of digital image and videos processing, pattern recognition, fractal image processing, artificial intelligence, information technologies, and remote sensing. She is the authored/ co-authored more than 130 publications. Margarita Favorskaya is a member of KES International organization and the IPC member of a number of National and International Conferences. She is on the editorial board of Int. J. Computer and Information Science and International Journal of Intelligent Decision Technology. She has a number of awards from the Ministry of Education and Science of the Russian Federation for significant contribution in educating and training a number of highly qualified specialists over a number of years. Dr. Suresh Chandra Satapathy has varied research interest, currently working at KIIT University, Bhubaneswar as a senior professor. He has obtained his Ph.D in Computer Science and Engineering from JNTU Hyderabad and M.Tech in CSE from NIT, Rourkela, Odisha, India. He has 26 years of teaching experience. His

About the Editors

xli

research interests include machine intelligence to intelligent system design, swarm intelligence & artificial intelligence. He has acted acted as a program chair for many international conferences and edited 8 volumes of proceedings from Springer SIST, LNCS and AISC series. He is currently guiding 8 scholars for Ph.D. Dr. Satapathy is also a senior Member of IEEE. His publication are total of above 120 publications and around many papers which are currently in review are not accounted. His publication are total of above 120 publications and around many papers which are currently in review are not accounted. Most of my publications are indexed in SCOPUS and Journal papers having SCI Impact factor. Details of my publications etc. can be found from DBLP and Google scholar citation link.

Performance Evaluation of Multi-layer Barriers for Machine-Induced Low-Frequency Noise Attenuation Abid Hossain Khan, Muhammed Mahbubur Razzaque and Md. Shafiqul Islam

Abstract In this work, the performance of different multi-layer barrier constructions in attenuating noise has been studied experimentally. Machine-induced lowfrequency noise within the frequency range of 100–500 Hz is focused in this work. Six different multi-layer barrier constructions have been employed for this purpose. Wood has been used as the rigid, reflective layer while glass wool and PE foam has been used as the soft, absorbing layers. An enclosure with five fixed walls and one flexible wall containing the noise barrier has been constructed to perform the experiments. A CASELLA CEL-62X Sound Pressure Level meter has been used to measure sound pressure at different frequencies. Results indicate that the transmission losses are not higher than 18 dB for the frequency range of interest. Results also reveal that triple-layer wooden barrier has superior performance in attenuating low-frequency noise although sandwich barriers are more suitable for higher frequencies.

Nomenclature Symbol τ ρ ρs

Parameter Transmittance Density of medium Mass per unit area

A. H. Khan (B) Department of Industrial and Production Engineering, Jashore University of Science and Technology, 7408 Jashore, Bangladesh e-mail: [email protected] M. M. Razzaque Department of Mechanical Engineering, Bangladesh University of Engineering and Technology, 1000 Dhaka, Bangladesh e-mail: [email protected] Md. S. Islam Department of Nuclear Engineering, University of Dhaka, 1000 Dhaka, Bangladesh e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_1

1

2

c η k ω θ TL I

A. H. Khan et al.

Speed of sound Structural loss factor Stiffness per unit area Frequency of incident sound Angle of incident sound Transmission loss Sound intensity

1 Introduction In case of a single-layer barrier of homogeneous material, a thick layer is required to restrict high-level noise. But in many cases, size is a limiting issue. For this reason, barriers employing new absorber materials as well as multi-layer composite barriers have been studied for better performance of the barriers with lesser space requirements. There are different constructions for multi-layer barriers. Among these construction types, double-layer construction and sandwich construction are two common types. As absorber material, the use of natural fiber has been expected by many researchers due to their high porosity. Larbig et al. [1] worked on natural fiber reinforced foam for noise reduction in automotive interiors. Khedari et al. [2] developed new insulating barriers with Durian peels and coconut coir fibers used as the raw material. The cost of the barriers was later reduced [3]. Rozli and Zulkarnain also worked on the performance of coconut fiber as noise absorber layer [4]. Other researches on natural fibers have also been conducted to identify their effectiveness in double-layer barriers [5, 6]. Over the years, a great deal of research has been carried out in identifying the physical and transmission loss characteristics of different sandwich panel constructions. Bending deformation in a sandwich panel construction was studied by Ross [7]. The description of bending rigidity has been utilized by Holmer [8] for developing a coincidence barrier design. Manning [9] found a way to optimize the performance of the coincidence wall and developed expressions for the effective damping in panels with multiple layers. The first description of the effects of thickness deformation on panel transmission loss is observed in the work of Lord [10]. Smolenski et al. [11] applied the same method in studying the effects of core compliance on panel transmission loss. Most of the machineries and vehicles generate low-frequency noises. These lowfrequency noises are very hard to attenuate due to their longer wavelengths. Lee developed compact sound absorbers for low frequencies [12]. Yang et al. studied meta-material panels for low-frequency noise absorption [13]. Mei et al. used similar study with dark meta-material panels [14]. From the literature study, it may be observed that the effect of multiple layers on transmission characteristics of low-frequency noise is yet to be realized fully.

Performance Evaluation of Multi-layer Barriers …

3

This work focuses on attenuation of machine-induced low-frequency noise within the frequency range of 100–500 Hz. Different barrier types have been experimentally studied to identify the one with superior performance. Glass wool and PE foam have been used in the absorber layer while wood has been used as the rigid layer.

2 Mathematical Model When sound is incident on the separating surface between two mediums of different densities and speeds of sound, a portion of the sound is absorbed in the second medium, a portion is reflected back to the previous medium, and the rest of the sound is transmitted through the second medium. If I i is the intensity of the incident sound, I a is the intensity of the absorbed sound, I r is the intensity of the reflected sound and I t is the intensity of the transmitted sound, we may write, Ii = Ia + Ir + It

(1)

In order to indicate the ease of sound passing through a medium, the transmittance is used. The transmittance of a medium is given by the ratio of the intensity of sound transmitted to the intensity of the sound incident on a surface, as shown in Eq. (2), τ=

It Ii

(2)

This transmittance is a function of the densities and speed of sound in two mediums and is given by Eq. (3), τ=

4(ρc)1 (ρc)2 {(ρc)1 + (ρc)2 }2

(3)

The characteristics of a bounded homogeneous barrier are shown schematically in Fig. 1. There are five regions of interest: stiffness-controlled region, resonanceFig. 1 Transmission loss in a homogenous barrier [15]

4

A. H. Khan et al.

controlled region, mass-controlled region, coincidence-controlled region, and damping controlled region [15]. From Fig. 1, it may be observed that at low frequencies, the transmission loss characteristics should be dominated either by stiffness or mass. The coincidence region should not be encountered for machine-induced low-frequency noises. At frequencies below the first natural frequency, the transmission loss depends on the bending stiffness of the material, thus called stiffness-controlled region. Since this region is seen at frequencies below 20 Hz, this region is not expected to be observed during this study. Followed by the stiffness-controlled region, a number of oscillations are observed in the curve due to the natural frequencies of the barrier. At natural frequencies, the incident sound waves excite the structural materials, leading to resonance within the barrier. As a result, the transmission loss increases or decreases significantly for these frequencies. The frequencies depend on the size and geometry of the barrier, which may be determined by Eq. (4) [15],

f i,n

π = 2



  k i2 n2 + 2 ρs a 2 b

(4)

where k is the bending stiffness, ρ s is mass per unit area, a and b are barrier dimensions, and i and n are integers. At frequencies above the first few natural frequencies, the response is mass-controlled. In that region, the equation of transmission loss is given by Eq. (5) [15],  TL = 10 log10

ω2 ρs2  4(ρa c)2 cos2 θ

 (5)

where θ is the angle of incidence, ρ a is the density of air, c is the speed of sound, and ω is the angular frequency. It may be shown mathematically that there is a 6 dB increase in transmission loss per octave increase in frequency. There is also a 6 dB increase in transmission loss if the mass is doubled. In case of multi-layer barriers, it is hard to obtain a universal characteristic equation. There are a few reasons for difficulties such as • With the addition of each layer, an additional surface interface is introduced, causing reflection of a portion of the transmitted sound wave. • If an absorbing material is added with a rigid material to form a composite barrier, the equivalent absorption and reflection coefficient is altered significantly for different frequencies. • Also, if the barrier is backed by a reflective material, the reflection of sound is increased, increasing transmission loss.

Performance Evaluation of Multi-layer Barriers …

5

Due to these reasons, the transmission losses differ from barrier to barrier. The non-homogeneity of the barrier material makes it even harder to derive a generalized equation since each barrier has a different composition.

3 Experiment Methodology In this experimental study, a noise source (drilling machine) is placed inside an enclosure which has fixed walls on five sides. On the remaining side, different types of barriers can be mounted. The inner surfaces of the fixed walls were covered with foam as a noise absorbing material. The replaceable wall has 6.35 mm wooden board support that can accommodate different material layers of the barriers. The dimension of the enclosure is 0.61 m × 0.61 m × 0.61 m. A sound level meter, CASELLA CEL-62X, is placed 1 m away from the barrier to measure the sound pressure level of the noise transmitting through the multi-layer barrier. Figures 2 and 3 show the noise enclosure and the replaceable wall that is used in the experiment. In order to collect base-line data, the front wall is kept open for initial measurement, i.e., without placing a barrier. The sound pressure level (SPL) meter is set to 1/3 octave bands. A-weighted frequency setting has been used since it is the most used one. After that, the data has been collected from the same distance with the same SPL meter settings placing different barriers in the front wall. The experiments were performed at the laboratory facility of the Department of Mechanical Engineering, Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh. The time of data collection was 2:00 pm to 3:00 pm local time since the area around the facility had minimum vehicle rush within that time frame. An average of 15 measured data was taken for each barrier at each frequency level so that the possibility of random error is minimized. Fig. 2 The noise enclosure

6

A. H. Khan et al.

Fig. 3 The replaceable wall

Fig. 4 Different barrier layer arrangements

In this study, six different barrier constructions have been studied, using different material arrangements, as shown in Fig. 4. The six constructions are as follows: (i) (ii) (iii) (iv) (v) (vi)

6.35 mm wooden board followed by 19.05 mm layer of wooden board. 6.35 mm wooden board followed by 12.7 and 6.35 mm layers of wood. 6.35 mm wooden board followed by 19.05 mm layer of glass wool. 6.35 mm wooden board followed by 19.05 mm layer of PE foam. 12.7 mm glass wool layer between two 6.35 mm wooden boards. 12.7 mm PE foam layer between two 6.35 mm wooden boards.

4 Results and Discussion In order to validate the accuracy of the experimental data, it is necessary to compare the data with that of available literature. The transmission loss characteristics for a barrier made of plywood has been presented in the work of Rudder, which is shown

Performance Evaluation of Multi-layer Barriers …

7

Fig. 5 Transmission loss characteristics for a heterogeneous material (plywood) [16]

in Fig. 5 [16]. Here, plywood has been taken as a reference since it is a heterogeneous material, similar to wood that has been used in this study as the rigid barrier layer. From Fig. 5, it may be observed that the transmission loss characteristics of plywood do not strictly follow the mass low, showing fluctuations in the lower frequency range due to resonance up to around 2000 Hz. After that, there is a gradual rise in transmission loss with increased noise frequency. However, doubling the frequency does not lead to 6 dB increase in transmission loss, rather there is less than 4 dB increase in transmission loss per octave increase in frequency. Figures 6, 7, and 8 show the average transmission loss measured for different Fig. 6 Transmission loss characteristics for double and triple layer wooden barrier

Fig. 7 Transmission loss characteristics for barriers made of wood and glass wool

8

A. H. Khan et al.

Fig. 8 Transmission loss characteristics for barriers made of wood and PE foam

barrier constructions. In this study, transmission losses for each barrier have been measured for noises in the range of 20–20,000 Hz. However, the focus is given on the frequency range 100–500 Hz since the machine-induced low-frequency noise is within this range. From 20 to 2000 Hz, a fluctuation of transmission loss occurs resembling the resonance-controlled region, similar to Fig. 5. Thus, only resonancecontrolled region is observed in the range 100–500 Hz. At frequencies higher than 2000 Hz, the mass-controlled region has appeared in almost all the barriers, though the attenuation is lower than that predicted by mass law, similar to that observed in Fig. 5. For double- and triple-layer wooden barriers, the transmission loss increase per octave is nearly 3.2 dB. In case of barriers containing Glass wool and PE foam, the transmission loss increase per octave could not be measured due to the presence of the coincidence region, observable in Figs. 7 and 8. From Figs. 6, 7, and 8, it may also be observed triple-layer wooden barrier has shown the best performance in attenuating noise in the frequency range 100–500 Hz compared to other barrier constructions. The triple-layer wooden barrier has better performance than doublelayer wooden barrier, suggesting that increasing layers may increase transmission loss. Apart from triple-layer barrier, sandwich barriers have shown good attenuation characteristics, compared to double-layer barriers made of Glass wool and PE foam. Sandwich barrier with Glass wool has greater attenuation than that made of PE foam. However, none of the barriers have shown attenuation greater than 18 dB. Also, Wood-glass wool sandwich barrier has good transmission loss characteristics than that of triple-layer wooden barrier at frequencies greater than 2000 Hz. So, Wood-glass wool sandwich barrier should be preferred when attenuation of both low and high-frequency noises are required.

5 Conclusion This work attempts to identify the performance of different multi-layer barriers in attenuating low-frequency noises. The focus is given on machine-induced noise with a frequency range of 100–500 Hz. Both double layer and sandwich barriers have been studied experimentally. Wood has been used as the rigid reflecting material

Performance Evaluation of Multi-layer Barriers …

9

while glass wool and PE foam has been used as the soft absorbing material for constructing the barriers. Six barrier constructions have been studied to analyze the effect of material construction on barrier performance. A 0.61 m × 0.61 m × 0.61 m enclosure with five fixed sides and one replaceable side is used to enclose the noise source. The replaceable wall can accommodate different constructed barriers. A CASELLA CEL-62X noise level meter has been used to record noise data for 15 days in order to evaluate the average values of transmission losses for each type of barrier. A comparative study between available literature and obtained results indicate that the transmission loss characteristics of the barrier constructions follow the pattern of that of heterogeneous material such as plywood. It is also observed that the transmission loss characteristics for all the barrier constructions have shown a trait of resonance region in the range 100–500 Hz, causing massive fluctuations in transmission losses with change in frequencies. The transmission loss is observed to be not higher than 18 dB for low-frequency noises. Among the six barriers, triple-layer wooden barrier has the most consistent performance within the frequency range of interest showing lesser degree of fluctuation and higher transmission loss, followed by Wood-Glass wool sandwich barrier. However, at higher frequencies, the use of wooden barriers only does not give satisfactory results, and both the sandwich barriers have the upper hand. Among the two sandwich barriers, Wood-Glass wool sandwich barrier has better noise attenuation properties than Wood-PE foam sandwich barriers. In this study, experiments have been conducted for only a few material combinations. The study may be extended to explore the effect of other rigid or absorber materials on the characteristics of multi-layer barriers. The use of homogeneous rigid material may also be explored. Again, more than three layers may be introduced in order to understand the effect of layering on TL characteristics of a barrier. Finally, the effect of the air gap between the layers may be evaluated.

References 1. Dahlke, B., Larbig, H., Scherzer, H.D., Poltrock, R.: Natural fiber reinforced foams based on renewable resources for automotive interior applications. J. Cell. Plast. 34(4), 361–379 (1998) 2. Khedari, J., Charoenvai, S., Hirunlabh, J.: New insulating particleboards from durian peel and coconut coir. Build. Env. 38(3), 435–441 (2003) 3. Khedari, J., Nankongnab, N., Hirunlabh, J., Teekasap, S.: New low-cost insulation particleboards from mixture of durian peel and coconut coir. Build. Env. 39(1), 59–65 (2004) 4. Rozli, Z., Zulkarnain, Z.: Noise control using coconut coir fiber sound absorber with porous layer backing and perforated panel. Amer. J. Appl. Sci. 7(2), 260–264 (2010) 5. Ono, T., Miyakoshi, S., Watanabe, U.: Acoustic characteristics of unidirectionally fiberreinforced polyurethane foam composites for musical instrument soundboards. Acoust. Sci. Tech. 23(3), 135–142 (2002) 6. Yang, H.S., Kim, D.J., Kim, H.J.: Rice straw–wood particle composite for sound absorbing wooden construction materials. Biores. Tech. 86(2), 117–121 (2003) 7. Ross, D.: Damping of plate flexural vibrations by means of viscoelastic laminae. Struct. Damp. 49–97 (1959)

10

A. H. Khan et al.

8. Holmer, C.I.: The coincidence wall: a new design for high transmission loss or high structural damping. J. Acoust. Soc. Am. 46(1A), 91 (1969) 9. Manning, J.E.: Development of the coincidence wall as a high TL panel. Cambridge Collaborative Report No. 1 (1971) 10. Ford, R.D., Lord, P., Walker, A.W.: Sound transmission through sandwich constructions. J. Sound Vibr. 5(1), 9–21 (1967) 11. Smolenski, C.P., Krokosky, E.M., Ewers, G.: Dilatational-mode sound transmission in sandwich panels. J. Acoust. Soc. Am. 45(1), 297–298 (1969) 12. Lee, J.: Compact sound absorbers for low frequencies. Noise Contr. Eng. J. 38, 109–117 (1992) 13. Yang, Z., Dai, H.M., Chan, N.H., Ma, G.C., Sheng, P.: Acoustic metamaterial panels for sound attenuation in the 50–1000 Hz regime. Appl. Phy. Lett. 96(4), 041906 (2010) 14. Mei, J., Ma, G., Yang, M., Yang, Z., Wen, W., Sheng, P.: Dark acoustic metamaterials as super absorbers for low-frequency sound. Nat Comm. 3, 756 (2012) 15. Cowan, A.J.: Sound transmission loss of composite sandwich panels (2013). https://ir. canterbury.ac.nz/handle/10092/7879 16. Rudder, Jr., F.F.: Airborne sound transmission loss characteristics of wood-frame construction (No. FSGTR-FPL-43). Forest Products Lab Madison WI (1985)

Planing Process on AISI S-1006, S-7, and S-4340, Based on Johnson–Cook Model Using Numerical Technique Abhinav, D. Prajwal and Punith Kumar

Abstract A two-dimensional numerical technique has been adopted to study the temperature rise, chip morphology, plastic stress/strain, strain hardening, and softening effect on the selected steel grades, namely S-1006, S-7, and S-4340 in the planing process. The material properties and constants are instituted on the popular Johnson–Cook (JC) constitutive model. A comparative analysis has shown, the maximum temperature rises in the case of S-7 (641.32 °C) followed by S-4340 (478.96 °C) S-1006 (287.19 °C). The temperature rise is mainly found in the secondary shear deformation zone and rise in shear stress found in the primary deformation zone. The plastic deformation is found facile in the case of S-1006 anticipated due to less dislocation within the crystal structure and work hardening effect found to be maximum in case of S-7 and the same is confirmed from the JC model material constants. It is believed that the simulation results may help in predicting machining uncertainties.

1 Introduction Earlier studies have shown, metal machining adds nearly 15% value in any manufacturing process [1]. Many works have been carried out in the area of metal machining either experimentally or using simulation techniques. A wide spectrum of critical issues are addressed viz. dimensional instability in the workpiece and tool, thermal fatigue, early tool wear, corrosion due to inappropriate selection of tool and cutting fluid, thermal softening and work hardening, microstructural transformation at the tool–chip interface, vibration often found to reduce the machining efficiency, etc. during the metal machining process [2–5]. Numerical techniques play a significant Abhinav (B) · D. Prajwal · P. Kumar Alliance College of Engineering and Design, Alliance University, Bangalore, India e-mail: [email protected] D. Prajwal e-mail: [email protected] P. Kumar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_2

11

12

Abhinav et al.

role in the comprehension of the machining process prior to the experimental one and save a great number of expenses. It has been also found that the reliability and accuracy of the metal machining process primarily depend on the material (ductile or brittle), cutting parameters viz. speed, depth of cut, type of coolant, etc. [6]. Many mathematical models are developed to examine the morphology of the chip formation, tool-chip interaction, and also, several attempts have been made to understand the mechanics of the machining process either analytically or by using simulation techniques. Since date to present many methamatical models viz. Baumann–Chiesa–Johnson (BCJ) law, Obikawa and Usui, Rhim and Oh model, and Johnson–Cook (JC) material the model has been applied and shown significant results when used with the numerical technique [7–10]. The advantage of integrating the JC model in FEA simulation necessitates less material constants compared to the other models [11]. Most of the previous works have been done on the tool materials viz. high-speed steel, brazed tools, and carbide tools and found a mixed outcome. Alumina ceramic with partially stabilized zirconia being very popular in recent tooling materials [12]. However, very limited work has been reported on the machining of the steel grades S-1006, S-7000, and S-4340. Due to the above research gap, an attempt is made to investigate the material machining behavior using simulation technique. The steel grades, namely S-1006 find applications in automobiles and where severe bending and welding were frequently required. S-7000 or S-7 grades are popular and extensively used as inserts used in metal turning operations Also, steel grade, S-4340 are used in the automobile transmission system and aircraft landing gears. In this work, an attempt is made to simulate the material behavior under planing process using simulation technique and also, the endeavor is made to showcase the technical challenges that may encounter during the machining process. Numerical technique has been adopted to simulate the machining behavior of the above-said materials.

1.1 Modeling Techniques and Assumptions A two-dimensional model developed in commercial software (Ansys Version 18 academic License). Three materials, namely S-1006, S-7, and S-4340 steel grades were selected for the study of the planing process. The motive behind the selection of these materials is based on frequent applications in the engineering and technology sectors. Two-dimensional model developed and meshed in the Ansys software. Structured quadrilateral mesh developed on the specimen body and unstructured mesh assign to the tool body. The element size for the material is assigned to be 0.001 mm. The tool geometry considered to be the rigid body and perfectly insulated at the tool-chip interface. The meshed model along with the boundary conditions (y- and z-direction of the work piece is constrained and all degree of freedom (dof) is assigned zero, tool rake angle (α): 10°, cutting velocity: 20 m/s, depth of cut: 0.2 mm and tool-chip interface friction: 0.1) applied on the model is shown in Fig. 1. The boundary conditions applied in the present case are based on the requirement raised during the

Planing Process on AISI S-1006, S-7, and S-4340 …

13

Fig. 1 Schematic of 2 D meshed model along with boundary conditions

machining of ductile materials viz. high cutting speed, small feed rate and depth of cut, sharp cutting edge, and low tool-chip friction [13]. A few important material properties and constants required for simulation purpose has been presented in Table 1. The material constants for the above-said materials are based on the Johnson–Cook (JC) model and taken from the material library database of the software. The JC model is popularly known as visco-plastic constitutive model principally suited for the materials which exhibit high plastic deformation or high strain rate [14, 15]. Mathematically this model is expressed as follows: where, ε, T room , and T melt are flow stress, plastic strain, effective strain rate, reference strain rate (1 s−1 ), room temperature, and melting temperature, respectively. A, B (in MPa) and n represents the yield stress of the material at room temperature, strain hardening also popularly known as hardening modulus and work- hardening exponent, respectively. The constants C and m represent the strain rate hardening and thermal softening coefficient, respectively. The JC model expression can be given as shown below:      ˙  T − Troom m ε¯ 1− σ¯ = [A + B ε¯ n ] 1 + C ln ε¯ o Tmelt − Troom

2 Results and Discussion Results obtained in the earlier experimental studies revealed that tool wear increases with the rise in temperature at the tool-chip interface, which is [16] also acknowledge in the present analysis. In the two-dimensional numerical studies of chip-tool engagement, it has been found that temperature at the tool-chip interface reaches up to the maximum value of 641.32 °C in the case of S-7 grade of steel followed by S-4340 (478.96 °C) and S-1006 (287.19 °C). The rise in temperature mainly found at the chip-tool interface. No temperature gradient found on the tool rake due to rigid body assumptions. A significant change in temperature gradient observed on the edge of the chip refer Fig. 3a–c. A gradual temperature drop found across the thickness of the chip in all the materials also, least temperature found at the tip of the

14 Fig. 2 Comparative analysis of the three materials. a Temperature rise, b equivalent plastic strain, c equivalent plastic stress

Abhinav et al.

Planing Process on AISI S-1006, S-7, and S-4340 … Table 1 Material properties and constants of Johnson–Cook model

Parameters Density,

kg/m3

Specific heat, J/kg/°C

15 S-1006

S-7

S-4340

7896

7750

7830

452

477

477

Shear modulus, Pa

8.18e10

8.18e10

8.18e10

Melting temperature, °C

1537.9

1489.9

1519.9

Johnson–Cook material constant A, Pa

3.5e8

1.539e9

7.92e8

B, Pa

2.75e8

4.77e8

5.1e8

n

0.36

0.18

0.26

c

0.022

0.012

0.014

m

1

1

1.03

chip. Not much significant temperature rise found in the primary deformation zone compared to the secondary shear deformation zone. The probable reason attributed to the work hardening effect. The work hardening anticipated being happened due to a repulsive interaction between the crystal and within the crystal structure. As the localized dislocation densities increase the bulk crystals block the translation of grains in the direction of the planing process. The work hardening in case of S-7 may be anticipated to cause early damage to the tool and found to be the main reason behind an increase in temperature compared to other materials. Therefore, judicious selection of tool material and design is an important index in the planing/orthogonal metal removal process. The dissipation of heat/temperature found gradually decreasing with the increase in the length of the continuous chip and found minimum at the extreme tip end of the chip. It has been also noticed that there is a significant amount of temperature gradient existing across the thickness of the chip. The relative comparison of the temperature rise of the three materials is shown in Fig. 2. It has been noticed that relatively faster shear happens in the case of S-1006 followed by S-4340 and S-7. It can be understood that the uncut layer of the work material (S-1006) ahead to the cutting tool offers less resistance/compression at the tool cutting edge as a result faster shear acknowledge. However, maximum compression/resistance at the edge of cutting tool can be understood in case of S-7 and confirm evident from the results obtained, the equivalent plastic strain (1.191) which is found comparatively less compared to S-1006 (1.248) reason attributed to greater yield stress constant of the work material and Strain hardening constant (A and B) refer to Table 1 and Fig. 4. From Fig. 2b it has been noticed that plastic strain in case of S-7 and S-4340 runs very closely, however, there is a significant difference in the values of equivalent plastic stress refer Figs. 2c and 5a, b.

16

Abhinav et al.

Fig. 3 Schematic of change in temperature in the work piece materials. a S-7, b S-4340, c S-1006

3 Conclusions At a given rake angle (α) 10°, continuous chips found to be formed irrespective of material types. Among all the materials S-1006, S-7, and S-4340, S-7 grade of steel exhibited maximum rise in temperature, plastic strain and plastic stress reason attributed to the greater anticipated relatively higher amount of strain hardening in the material and also confirmed from the Johnson–Cook model hardening constant (A, B and n) values. The change in temperature initiated from the top side of the chiptool interface and gradually found to be decreased across the thickness in all the

Planing Process on AISI S-1006, S-7, and S-4340 …

17

Fig. 4 Schematic of change in equivalent plastic strain in the work piece materials. a S-7, b S-4340, c S-1006

workpiece materials the obvious reason found due to insulated cutting tool treated as a rigid body. Friction found in the primary shear zone has a negligible influence on the temperature rise reason attributed to less area of contact made by the tool geometry (nose tip) with the work piece in the primary shear zone and also due to viscoplastic characteristic of materials. Due to more slack slips, high shear deformation is acknowledged in the primary shear zone.

18

Abhinav et al.

Fig. 5 Schematic of change in equivalent plastic stress in the work piece materials. a S-7, b S-4340, c S-1006

References 1. Calamaz, Madalina, et al.: A new material model for 2D numerical simulation of serrated chip formation when machining titanium alloy Ti–6Al–4V. Int. J. Mach. Tools Manuf. 48, 275–288 (2008) 2. Shaw, C., et al.: Machining Titanium. Massachusetts Institute of Technology (1954) 3. Komanduri, R., et al.: New observations on the mechanism of chip formation when machining titanium alloys. Wear 69, 179–188 (1981) 4. Vyas, A., et al.: Mechanics of saw-tooth chip formation in metal cutting. J. Manuf. Sci. Eng. 121, 163–172 (1999)

Planing Process on AISI S-1006, S-7, and S-4340 …

19

5. Hua, J., et al.: Prediction of chip morphology and segmentation during the machining of titanium alloys. J. Mater. Process. Technol. 150, 124–133 (2004) 6. Generalized model of chip formation (Chap. 1). In: Astakhov, V.P., (ed.) Tribology and Interface Engineering Series, vol. 52, pp. 1–67 (2006). https://doi.org/10.1016/S0167-8922(06)80003-9. https://www.sciencedirect.com/science/article/abs/pii/S0167892206800039?via%3Dihub 7. Guo, Y.B., et al.: Dynamic material behavior modelling using internal state variable plasticity and its application in hard machining simulations. J. Manuf. Sci. Eng. 128, 749–756 (2006) 8. Rhim, S.H., et al.: Prediction of serrated chip formation in metal cutting process with new flow stress model for AISI 1045 steel. J. Mater. Process. Technol. 171, 417–422 (2006) 9. Barge, M., et al.: Numerical modelling of orthogonal cutting: influence of numerical parameters. J. Mater. Process. Technol. 164(165), 1148–1153 (2005) 10. Melzi, N., et al.: Applying a numerical model to obtain the temperature distribution while machining. Acta Physica Polonica A 131 (2017) 11. Yi-ben, Zhang, et al.: A modified Johnson-Cook model for 7N01 aluminum alloy under dynamic condition. J. Cent. South Univ. 24, 2550–2555 (2017) 12. Smuk, B., Zutkowska, M.S., et al.: Alumina ceramics with partially stabilized zirconia for cutting tools. J. Mater. Process. Technol 133(1–2), 195–198 (2003) 13. Yan, J., Syoji, K., et al.: Ductile regime turning at large tool feed. J. Mater. Process. Technol. 121(2–3), 363–372 (2002) 14. Ning, Jinqiang, et al.: Model-driven determination of Johnson-Cook material constants using temperature and force measurements. Int. J. Adv. Manuf. Technol. 97, 1053–1060 (2018) 15. Ning, Jinqiang, et al.: Inverse determination of Johnson-Cook model constants of ultra-finegrained titanium based on chip formation model and iterative gradient search. Int. J. Adv. Manuf. Technol. 99, 1131–1140 (2018) 16. Jin, Du, et al.: Heat partition and rake face temperature in the machining of H13 steel with coated cutting tools. Int. J. Adv. Manuf. Technol. 94, 3691–3702 (2018)

An Analytical Study of Diametral Error in Simultaneous Turning Process Sandeep Kumar, Kalidasan Rathinam, Vivek Sharma and VaitlaSai Kumar

Abstract Diametral error plays a significant role in determining the quality of the machined component. It becomes more important for long-slendered workpiece with slenderness ratio greater than six. In the present work, an attempt is made to estimate the diametral error analytically during simultaneous turning process. The cutting tools are kept opposite to one another, so that the cutting forces act opposite to each of them. Euler–Bernoulli beam theory was applied to determine the cutting tool deflection. The workpiece was assumed as a propped cantilever beam. The diametral error was determined for various slenderness ratios of the workpiece. It was revealed that the diametral accuracy increased when the slenderness ratio decreased. This is due to the fact that lesser length-to-diameter ratio contributed to increased rigidity of the workpiece, resulting in the reduction of diametral error. The maximum and minimum diametral error occurred for the workpiece slenderness ratio of 8 and 4, respectively. Further, the results of the developed analytical model were compared with the published literature, and a good agreement was found.

1 Introduction In turning process, the diametral error is caused due to cutting force, compliance of machine tool, misalignment of assembled parts and cutting conditions such as cutting speed, feed and depth of cut. One method of mitigating the diametral error is S. Kumar · K. Rathinam (B) · V. Sharma · V. Kumar School of Mechanical Engineering, Lovely Professional University, Phagwara, Punjab, India e-mail: [email protected] S. Kumar e-mail: [email protected] V. Sharma e-mail: [email protected] V. Kumar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_3

21

22

S. Kumar et al.

to use two cutting tools kept opposite to each other to machine the workpiece simultaneously. The concept of turning with multiple tools was started almost six decades back. McCullough [1], way back in 1963, reported multi-tool lathe operations which are carried out for maximizing the production rate and tool life. Simultaneous turning was used to improve the quality of the machined component as well as to increase the productivity. In recent times, attention was given on stability aspects of simultaneous turning process. Gouskov et al. [2] made an analytical investigation on the stability of multi-cutter turning process by a trace. It was revealed that the stability of simultaneous turning process depends upon the rotational speed of the workpiece and it does not depend upon the angular position of the cutting tools. The same authors [3] in another work mathematically modelled the stability of steady-state double-tool turning process. It was noted that the instability zone extended with the increase in cutting speed. Additionally, the cutting force and tool deflection remained unaffected by the variation in cutting speed. Kalidasan and Sandeep [4] performed a numerical study of double-tool turning process. It was observed that the cutting forces were uninfluenced with the change in distances between the two cutting tools. The same phenomena were observed for cutting temperature also. Reith et al. [5] made a theoretical investigation to increase the stability regions during parallel turning by tuning the dynamic properties of the cutting tools. The overhang length of a cutting tool was varied. It was concluded that cutting performance can be improved significantly by appropriately detuning the cutting tool. The same authors [6] predicted the stability properties of double-cutter turning operation. It was concluded that the stable machining region can be extended for detuned two-cutter system by suitably altering the dynamic properties of the cutting process. Yadav [7] studied the effect of cutting conditions on the surface roughness during duplex turning process. It was reported that a better surface finish was obtained when both feed and secondary depth of cut were lower. On the other hand, surface finish was unaffected by primary depth of cut. The same author [8] optimized the duplex turning process by using Taguchi and response surface methodology. It was reported that the optimized cutting conditions obtained by hybrid approach yielded a good surface finish as compared with other methods. An experimental investigation was performed by Kalidasan et al. [9] on the machining accuracy of two-tool simultaneous turning process. It was observed that the workpiece diametral error was higher along the tailstock side compared to headstock side. The effect of depth of cut and distances between the cutting tools on the diametral error was also studied. Additionally, it was found that the diametral error increased at higher depths of cut. This was due to increased cutting forces at larger depths of cut. In another research work, Kalidasan et al. [10] analytically estimated the diametral error considering the headstock and tailstock as flexible supports for conventional turning process. It was found that the maximum diametral error occurred at the mid-length of the workpiece. Chen and Tian [11] predicted the diametral error during batch manufacturing process. The error correction was applied by a modified CNC part program. It was noted that the diametral accuracy improved significantly by this method.

An Analytical Study of Diametral Error in Simultaneous …

23

Fig. 1 a Euler–Bernoulli beam model and b force model of two cutting tools kept on opposite sides

It is observed from the literature survey that certain amount of research work was done in simultaneous turning process towards the improvement of process stability and few on numerical and analytical prediction of process parameters and machining accuracy. By positioning the two cutting tools opposite to each other, the radial force component of the cutting tools can be effectively utilized to reduce the diametral error without the need of any additional resources. Therefore, an effective analytical model considering the cutting forces to predict the diametral error in simultaneous turning becomes indispensible. The present research work aims to fill that research gap. The main aim of the present research work is to develop an analytical model to estimate the diametral error of workpiece for various slenderness ratios and compare with the published literature.

2 Formulation of Diametral Error of Workpiece with Cutting Tools kept Opposite to Each Other Schematic diagram of Euler–Bernoulli beam model and the force model of simultaneous turning process with two cutting tools positioned on opposite sides is shown in Fig. 1a, b. Cutting forces play a major role in workpiece deflection and diametral

24

S. Kumar et al.

error of the machined component. In this study, tangential cutting force component was considered in predicting the diametral error. The work and tool materials are grey cast iron and coated tungsten carbide. The workpiece was modelled as a propped cantilever beam. Euler–Bernoulli beam theory was used. In order to include the machine tool deflection, flexible support was considered at the ends. The compliances of the machine tool and tool holders were adopted from the research work of Kops et al. [12]. In this formulation, the plane x-z is the plane in which tangential force is acting and depth of cut is provided in the x-direction. The force and moment equilibrium equations are written as: F1 − F2 = Rs + Rt

(1)

Ms = F1 z 1 − F2 z 2 − Rt L

(2)

where F 1 and F 2 are the tangential force components. Rs and M s are the reaction and moment at head stock, Rt is the reaction at the tail stock, L is the workpiece length, and z1 and z2 are the distance between headstock and line of action of the tangential force component of the first and second cutting tools, respectively. Considering section Z a − Z a (0 < x < z1 ), moment equilibrium in terms of reaction force is expressed as: E I 1 v  = Ms − Rs x

(3)

where x is the distance of section Z a − Z a from the head stock. I 1 and E are the initial moment of inertia and elastic modulus of the workpiece, respectively. Substituting the values of Rs and M s from Eqs. (1) and (2) in Eq. (3), the moment is given as: v  =

(x − z 2 )F2 + (−L + x)Rt + F1 (−x + z 1 ) EI1

(4)

Integrating Eq. (4), the slope and deflection in section Z a − Z a are given by: x((−2L + x)Rt − F1 (x − 2z 1 ) + F2 (x − 2z 2 )) + C1 2E I 1   x 6EC1 I1 − 3L x Rt + x 2 Rt − x F1 (x − 3z 1 ) + x F2 (x − 3z 2 ) v= + C2 6E I 1 v =

(5) (6)

where C 1 and C 2 are the constants of integration. At x = 0, v  = 0 and v = Rs Cs = (F1 − F2 − Rt )Cs . Applying the boundary condition, the values of constant of integration are determined as C1 = 0 and C2 = Cs (F1 − F2 − Rt ). On substituting in Eq. (6), slope and deflection at section Z a − Z a are expressed as: v =

x(−x F1 + x F2 − 2L Rt + x Rt + 2F1 z 1 − 2F2 z 2 ) 2E I 1

(7)

An Analytical Study of Diametral Error in Simultaneous …

v = Cs (F1 − F2 − Rt ) +

25

x 2 (−x F1 + x F2 − 3L Rt + x Rt + 3F1 z 1 − 3F2 z 2 ) (8) 6E I 1

Now taking the section Z a − Z a in between F 1 and F 2 (z1 < x < z2 ), the moment equilibrium equation can be written as: E I 2 v  = Ms − Rs x + F1 (x − z 1 ) v  =

(−L + x)Rt + F2 (x − z 2 ) EI2

(9) (10)

where I 2 is the moment of inertia of the workpiece after turning by the first cutting tool. Integrating the Eq. (10), the slope and deflection are obtained as: 2EC3 I2 + x(x F2 − 2L Rt + x Rt − 2F2 z 2 ) 2E I 2

(11)

6EC4 I2 + x(6EC3 I2 + x(x F2 − 3L Rt + x Rt − 3F2 z 2 )) 6E I 2

(12)

v  = v=

where C 3 and C 4 are the constants of integration. At x = z 1 , the deflection and slope of this section and the previous section become equal, hence equating Eqs. (7) and (11) for slope and Eqs. (8) and (12) for deflection. The values of constant are obtained as x(x F2 I1 + x F1 I2 − x F2 I2 − 2L I1 Rt + x I1 Rt +2L I2 Rt − x I2 Rt − 2F1 I2 z 1 − 2F2 I1 z 2 + 2F2 I2 z 2 ) C3 = − 2E I 1 I2

(13)

6ECs F1 I1 I2 − 6ECs F2 I1 I2 − 6ECs I1 I2 Rt − 3L I1 Rt Z 12 + 3L I2 Rt Z 12 +2F2 I1 Z 13 − F1 I2 Z 13 − 2F2 I2 Z 13 + 2I1 Rt Z 13 −2I2 Rt Z 13 − 3F2 I1 Z 12 Z 2 + 3F2 I2 Z 12 Z 2 C4 = 6I1 I2 (14) Substituting the values of constant in Eqs. (11) and (12), the slope and deflection of section Z a − Z a in between F 1 and F 2 are given by: I1 (x − z 1 )(−2L Rt + Rt (x + z 1 ) + F2 (x + z 1 − 2z 2 )) +I2 z 1 (−2L Rt + F1 z 1 + F2 z 1 + Rt z 1 − 2F2 z 2 ) v = 2E I1 I2

(15)

26

S. Kumar et al.

(6ECs I1 I2 (F1 − F2 − Rt ) + I1 (x − z 1 )2 (−3L Rt + Rt (x + 2z 1 ) + F2 (x + 2z 1 − 3z 2 )) +I2 z 1 (3x F2 z 1 + 3x Rt z 1 + F1 (3x − z 1 )z 1 − 2F2 z 12 −2Rt z 12 + 3L Rt (z 1 − 2x) − 6x F2 z 2 + 3F2 z 1 z 2 )) v= 6E I1 I2

(16)

Similarly, for the section Z a − Z a in between F 2 and Rt (z2 < x < L), the moment equilibrium equation is expressed as: E I 2 v  = Ms − Rs x + F1 (x − z 1 ) − F2 (x − z 2 ) v  =

(−L + x)Rt EI2

(17) (18)

Integrating the Eq. (18), the slope and deflection are given by: v  = C5 + v=

x(−2L + x)Rt 2E I 3

6EC6 I3 + x(6EC5 I3 + x(−3L + x)Rt ) 6E I 3

(19) (20)

where I 3 is the moment of inertia of the workpiece after turning by the second cutting tool. C 5 and C 6 are the constants of integration. At x = z 2 , the deflection and slope of this section and the previous section become equal, hence equating Eqs. (15) and (19) for slope and Eqs. (16) and (20) for deflection. The values of constant are obtained as: 1  2L I1 I3 Rt z − 2L I2 I3 Rt z 1 − F2 I1 I3 z 12 + F1 I2 I3 z 12 2E I 2 I3 + F2 I2 I3 z 12 − I1 I3 Rt z 12 + I2 I3 Rt z 12 + 2L I1 I2 Rt Z 2

C5 =

− 2L I1 I3 Rt Z 2 + 2F2 I1 I3 Z 1 Z 2 − 2F2 I2 I3 Z 1 Z 2  −F2 I1 I3 Z 22 − I1 I2 Rt Z 22 + I1 I3 Rt Z 22 C6 =

(21)

1 (6ECS F1 I1 I2 I3 − 6ECS F2 I1 I2 I3 − 6E Rt CS I1 I2 I3 6E I 1 I1 I3 − 3L I1 I3 Rt Z 12 + 3L I2 I3 Rt z 12 + 2F2 I1 I3 z 13 − F1 I2 I3 z 13 − 2F2 I2 I3 z 13 + 2I1 I3 Rt z 13 − 2I2 I3 Rt z 13 − 3F2 I1 I3 z 12 z 2 + 3F1 I2 I3 z 12 z 2 − 3L I1 I2 Rt z 22  (22) +3L I1 I3 Rt z 22 + F2 I1 I3 z 23 + 2I1 I2 Rt Z 23 − 2I1 I3 Rt z 23

An Analytical Study of Diametral Error in Simultaneous …

27

Substituting the values of constant in Eqs. (19) and (20), the slope and deflection of section Z a − Z a in between F 2 and Rt can be written as: I2 I3 z 1 (−2L Rt + F1 z 1 + F2 z 1 + Rt z 1 − 2F2 z 2 ) − I1 (I2 Rt (2L − x − z 2 )(x − z 2 ) + I3 (z 1 − z 2 )(−2L Rt + F2 (z 1 − z 2 ) + Rt (z 1 + z 2 ))) v = 2E I1 I2 I3 

(23)

6E I1 I2 I3 Cs (F1 + F2 − Rt ) + I2 I3 z 1 (−3x F2 z 1 + 3x Rt z 1 + F1 (3x − z 1 )z 1  +2F2 z 12 v= 6E I1 I2 I3   1 2 + (I2 I3 z 1 ) −2Rt z 1 + 3L Rt (−2x + z 1 ) + 6x F2 z 2 − 3F2 z 1 z 2 6E I1 I2 I3   1 + I1 −I2 Rt (3L − x − 2z 2 )(x − z 2 )2 + I1 I3 (z 1 − z 2 )(3L Rt (2x − z 1 − z 2 )) 6E I1 I2 I3 1 + I1 I3 (z 1 − z 2 )(F2 (3x − 2z 1 − z 2 )(z 1 − z 2 ) 6E I1 I2 I3    (24) +Rt −3x(z 1 + z 2 ) + 2 z 12 + z 1 z 2 + z 22

At x = L, the deflection is equal to Rt Ct . Substituting the value of x, the value of Rt is determined. Substituting the value of Rt in Eq. (16) and putting x = z1 , the deflection at first cutting tool edge is obtained as: 6ECs I1 (−F2 (6I2 I3 (L − z 1 )z 1 (L − z 2 ) + I1 (3I3 (L − z 2 )(2L − z 1 − z 2 )    (z 2 − z 1 ) + 2I2 L 3 + 3ECt I3 − 3L 2 z 2 + 3Lz 22 − z 23 + 2F1 3I2 I3 (L − z 1 )2 z 1   3   2 2 3 2 +I1 I2 L + 3ECt I3 − 3L z 2 + 3Lz 2 − z 2 − I3 (z 1 − z 2 ) 3L + z 12 + z 1 z 2   + z 12 (F1 z 1 (3I2 I3 (L − z 1 )2 z 1 + 4I1 I2 (L 3 + 3ECt I3 +z 22 − 3L(z 1 + z 2 )     −3L 2 z 2 + 3Lz 22 − z 23 − I3 (z 1 − z 2 ) 3L 2 + z 12 + z 1 z 2 + z 22 − 3L(z 1 + z 2 )  +F2 −3I2 I3 (L − z 1 )z 12 (L − z 2 ) + I1 (2I2 (z 1 − 3z 2 )  3  L + 3ECt I3 − 3L 2 z 2 + 3Lz 22 − z 23 +3I3 (z 1 − z 2 )(L 2 (z 1 + 3z 2 ) + z 2 (z 12 + z 1 z 2 + 2z 22 )   −L z 12 + 2z 1 z 2 + 5z 22 V1 =   2  12E I1 (I2 I3 z 1 3L − 3Lz 1 + z 12  3  +I1 (I2 L + 3ECs I3 + 3ECt I3 − 3L 2 z 2 + 3Lz 22 − z 23  2  −I3 (z 1 − z 2 ) 3L + z 12 + z 1 z 2 + z 22 − 3L(z 1 + z 2 )

(25) Substituting the value of Rt in the Eq. (24) and putting x = z2 , the deflection at second cutting tool edge is obtained as:

28

S. Kumar et al.

 2 F2 −3I22 I3 z 14 (L − z 2 )2 + I12 (z 1 − z 2 )3 (3I  3 (L − z 22 ) (z 2 − z 1 ) 2 3 3 2 +4I2 L + 3ECt I3 − 3L z 2 + 3Lz 2− z 2 − F1 I2 z 1 (3I2 I3 (L − z 1 ) z 12 (L − z2 ) + I1 (2I2 (z 1 − 3z 2 ) −L 3 − 3ECtI3 + 3L 2 z 2  2 2 2 −3L z2 + z 23 ) − 3I3 (z 1 − z 2 ) L 2 (z 1 + 3z 2 ) + z 2 z 1 + z 1 z 2 + 2z 2  2 + 6ECs I1 I2 −2F2 (3I2 I3 z 1 (L z 2 )2 −L z 12 +  2z 1 z 2 + 5z22  − 2 3 +I1 3I3 (L − z 2 ) + I1 (3I3 (L  − z 2 ) (z 2 − z 1 ) + I2 L 2 3 2 + 3EC I − 3L z + 3L − z ) + F1 (6I2 I3 (L − z 1 )z 1 (L − z 2 ) 2 2 2 t 3 +I1 3I3 (L − z 2 )(2L − z 1 − z 2 )(z 2 − z 1 ) + 2I2 L 3 + 3ECt I3 2 3 2 −3L z 2 + 3Lz 2 − z 2 V2 =    2  z 12 + I1 (I2   3 12E I1 I2 I3 z 1 3L − 3Lz 1 + L + 3ECs I3 + 3ECt I3 − 3L 2 z 2 + 3Lz 22 − z 23  − I3 (z 1 − z 2 ) 3L 2 + z 12 + z 1 z 2 + z 22 − 3L(z 1 + z 2 ) (26) In case of simultaneous turning with tools kept on opposite sides, the cutting-edge deflection of the first and second cutting tools is given by Eqs. (25) and (26). This gives the radial error, and twice of it gives the diametral error of the workpiece. It can be observed from Eqs. (25) and (26) that the diametral error depends on many parameters such as cutting forces, work material property, moment of inertia of the workpiece before machining and after machining by first and second cutting tools, rigidity of headstock, and rigidity of tailstock. The analytically determined diametral error is compared with the published literature in the subsequent section.

3 Comparison of Analytical Model with Literature Figure 2 shows the diametral error of the workpiece obtained from the developed analytical model for the slenderness ratios of 4, 6, 7 and 8. It can be observed that the maximum diametral error of 0.052 mm occurred for a workpiece slenderness ratio of 8 with x/L ratio of 0.55. On the other hand, a minimum diametral error 0.008 mm occurred for a workpiece slenderness ratio of 4 with x/L ratio of 0.43. Experimental work of Kalidasan et al. [9] reveals that the diametral error was 0.075 mm for x/L ratio of 0.55, while turning grey cast iron with a carbide tool. The deviation between the analytical and experimental values was found to be in the range of 30–35%. Table 1 shows the comparison of proposed analytical model for two slenderness ratios with the published experimental results of Kalidasan et al. [9]. On similar lines for the slenderness ratio of 6, the diametral error was found to be 0.02 mm for the x/L ratio of 0.6. For the slenderness ratio of 7, almost at the midsection with x/L ratio of 0.55, the diametral error was found to be 0.03 mm. Hence, in between the region having x/L ratio of 0.55 and 0.6, the diametral error decreases by 33% when the slenderness ratio is decreased from 7 to 6. Apart from the reduction in the diametral error, the material removal rate is increased due to the usage of

An Analytical Study of Diametral Error in Simultaneous …

29

Fig. 2 Workpiece diametral error for different slenderness ratios

two cutting tools. It is well known that in traditional turning process, a follower rest was used to prevent the workpiece deflection thereby reducing the diametral error. Jianliang and Rongdi [13] reported that the diametral error is reduced by keeping the workpiece length as constant and increasing the diameter of the workpiece. Another way is to keep the diameter of the workpiece as constant and decrease its length. For all the slenderness ratios, the diametral error of the workpiece at the tailstock is more than the headstock due to the lesser rigidity of the tailstock. It is observed that at the headstock end, the diametral error is 0.014 mm for the lowest slenderness ratio. For the highest slenderness ratio, the diametral error is 0.023 mm. Thus, the diametral error increases by 1% at the headstock end when the slenderness ratio increases from 4 to 8. In similar way, at the tailstock end, the diametral error was 0.019 mm for the lowest slenderness ratio. For the highest slenderness ratio, the diametral error was 0.04 mm. Here, the diametral error increases by 2% when the slenderness ratio increases from 4 to 8. Murthy [14] reported that form error caused due to the compliance of the workpiece is decreased when the rigidity of the machine tool components is increased.

4 Conclusion The diametral error for the slenderness ratios of 4, 6, 7 and 8 is analytically evaluated. Further, it is compared with the published literature. The following conclusions are made. • The maximum diametral error of 0.052 mm and the minimum diametral error of 0.008 mm occur for the slenderness ratios of 8 and 4, respectively. The corresponding x/L ratios are 0.55 and 0.43.

0.016

0.027

6

0.2

0.027

0.001

0.4

Distance (x/L)

0.025

0.001

0.6

Experimental diametral error (mm)

4

Slenderness ratio

0.025

0.027

0.8 0.046

0.026

1.0 0.02

0.012

0.2

0.020

0.001

0.4

0.020

0.001

0.6

Analytical diametral error (mm)

Table 1 Comparison of diametral error between developed analytical model and experimental results

0.020

0.020

0.8

0.034

0.020

1.0

30 S. Kumar et al.

An Analytical Study of Diametral Error in Simultaneous …

31

• The diametral error decreases by 33% when the workpiece slenderness ratio is decreased from 7 to 6. The decrease in diametral error leads to an increase in diametral accuracy. • The diametral error at the headstock end of the workpiece is lesser than the tailstock end for all slenderness ratios. • When the slenderness ratio increases from lowest value of 4 to highest value of 8, the diametral error increases by 1 and 2% at the headstock end and tailstock end, respectively.

References 1. McCullough, E.M.: ASME J. Eng. Ind. 402–404 (1963) 2. Gouskov, A.M., Guskov, M.A., Tung, D.D., Panovko, G.Y.: J. Machin. Manuf. Reliab. 47(4), 317–323 (2018) 3. Gouskov, A.M., Guskov, M.A., Tung, D.D., Panovko, G.Y.: Vibroeng. Procedia 17, 124–129 (2018) 4. Kalidasan, R., Sandeep, K.: Matec Web of Conferences, vol. 192, pp. 01001–01004 (2018) 5. Reith, M.J., Bachrathy, D., Stepan, G.: J. Dyn. Syst. Measur. Control 139, 014503-1–0145037(2017) 6. Reith, Marta J., Bachrathy, Daniel, Stepan, Gabor: Mach. Sci. Technol. 20(3), 440–459 (2016) 7. Yadav, R.N.: Adv. Manuf. 5(2), 149–157 (2017) 8. Yadav, R.N.: Measurement, 100, 131–138 (2017) 9. Kalidasan, R., Senthilvelan, S., Dixit, U.S., Vaibhav, J.: Int. J. Precis. Technol. 6(2), 142–158 (2016) 10. Kalidasan, R., Sandeep, K., Vivek, S.: Proceedings of 2nd International Symposium on Mechanical Design, Manufacture and Automation. Khalifa University, Abu Dhabi (2018) 11. Chen, T., Tian, X.: Int. J. Adv. Manuf. Technol. 77(1–4), 281–288 (2015) 12. Kops, L., Gould, M., Mizrach, M.: J. Eng. Ind. 115, 253–257 (1993) 13. Jianliang, G., Rongdi, H.: Int. J. Mach. Tools Manuf. 46, 1002–1012 (2006) 14. Murthy: Int. J. Mach. Tool Des. Res. 10(2), 317–325 (1970)

Improvement of Electric Heater Design for Household Cooking Application in Developing Countries Angkush Kumar Ghosh, Abid Hossain Khan and A. N. M. Mizanur Rahman

Abstract In this experimental study, the possibility of improvement of conventional electric heater to reduce energy loss has been investigated. In order to do so, a modified electric heater construction has been employed, one covered with glass wool insulation. The temperature profile inside the heater has been measured both in vertical and radial direction with the help of K-type thermocouples. Three different food items, namely, Water, Minicate Rice, and Red Lentil, have been used to test the performance of the modified insulated heater compared to the conventional one. Results indicate that it is possible to save energy up to 43.9%, 32%, and 23.08% for the above-mentioned three food items, respectively, by using the insulated heater. However, inconsistency in the amount of energy-saving is observed. It has also been observed that the energy-saving for boiling water increases with increasing amount of water, suggesting that the insulated heater is more efficient for higher cooking volume compared to the conventional one.

1 Introduction Energy crisis is becoming more and more visible in recent times due to the diminishing inventory of fossil fuels. A group of scientists and researchers are trying to find the reasons behind the crisis and identify solutions for the problem [1] while others are looking for sustainable sources of energy [2]. Though there have been A. K. Ghosh (B) · A. N. M. M. Rahman Department of Mechanical Engineering, Khulna University of Engineering and Technology, Khulna 9203, Bangladesh e-mail: [email protected] A. N. M. M. Rahman e-mail: [email protected] A. H. Khan Department of Industrial and Production Engineering, Jashore University of Science and Technology, Jashore 7408, Bangladesh e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_4

33

34

A. K. Ghosh et al.

suggestions for focusing on alternative sources of energy such as renewable energy, the results are far from encouraging. The availability of renewable energy is highly dependent on the geographic location of a country [3, 4]. As a result, the best possible outcome may be achieved from the efficient utilization of available energy. Although the whole world is concerned about energy crisis, third world countries are the ones who will suffer the most from this. This is because of their ever-growing need for energy for development [5, 6]. Without sustainable energy, their development process is bound to face a roadblock. Since per capita energy consumption itself is a measure of development of a country, it is likely that energy consumption will rise with increase in income of the people. As a result, developing countries are experiencing rapid rise in energy demand, especially in Asia [7]. One of the major areas of energy consumption in third world countries, especially the ones that are developing, is residential usage for cooking [8, 9]. Although most of the rural areas are dependent on biomass [10, 11] for cooking, the use of a conventional electric heater is not negligible [12]. The structure of a conventional electric heater is such that appreciable amount of heat is wasted during cooking. A greater portion of the bottom area of the heater is open to the atmosphere which causes heat loss by radiation. Again, through the inside wall of the heater which is mainly mortar, appreciable amount of heat is lost by conduction. Moreover, the height of the heater body is not to any standard and there is air gap between utensil’s bottom and the coil plate. This causes less heat to receive by the utensil as air acts as thermal insulator which ultimately is a loss of energy. Due to these losses, the heater is somewhat less efficient (around 66%) compared to other advanced electric cooking devices like hot plate (around 90%) [13]. The aim of this work is to conserve energy by means of improving the design of the conventional electric heater by reducing the energy loss and, thus, reducing the cooking time. To do so, a modified heater construction has been proposed. In this construction, the heater is covered with insulation material. Among various thermal insulators, glass wool has been considered suitable for the purpose because of its very low thermal conductivity and temperature resistance [14]. Both the modified insulated heater and the conventional heater have been tested to measure the achievable energy-saving.

2 Experiment Methodology In this experimental study, the possible improvement of the construction of a conventional electric heater has been investigated. To do so, as shown in Fig. 1a, a modified construction of the conventional heater has been studied in which the heater is covered with glass wool insulation. In order to make a glass wool insulated electric heater, a box made of mild-steel sheet has been used to support the insulation panels placed around the heater. As shown in Fig. 1b, c, the bottom and sides of the heater have also been insulated by glass wool, respectively. In addition, as shown in Fig. 1d,

Improvement of Electric Heater Design for Household Cooking …

35

Fig. 1 Construction of the insulated heater; a photographic view, b glass wool insulation at the bottom of the heater, c insulation surrounding the heater, and d reflective cement coating on inside wall

on the inside wall of the heater, a mixture of white cement, chalk powder, and water have been brushed so that the mixture acts as a reflective coating. Once construction is completed, Fig. 2 illustrates the experimental setup to determine the temperature profiles for the modified electric heater (denoted as Insulated Heater), as well as the conventional one (denoted as Bare Heater). This is done to

Fig. 2 Experiment setup for determining temperature profiles at radial and vertical direction for a insulated heater, and b conventional bare heater

36

A. K. Ghosh et al.

obtain and understand the optimum height condition between the coil surface and the bottom surface of the utensil. Temperature profiles for both of the heaters have been determined along radial and vertical direction from the coil plate surface to a height of 4.5 cm. As shown in Fig. 2, three thermocouples and three temperature recorders have been used for this purpose. A wooden frame is also constructed and used to hold the thermocouples along the radial direction and to move the thermocouples vertically easily at different heights. Along radial direction, three thermocouples at 0 cm, 5.75 cm, and −5.75 cm distance from the center of the coil plate are placed to record the temperatures. Along the vertical direction, temperature is determined at six heights from coil plate to the utensil’s bottom surface. However, Fig. 3 illustrates the experimental setup to test the performance of the modified heater and the conventional one. Three different food items, namely, Water, Minicate Rice, and Red Lentil have been used for this purpose. As shown in Fig. 3, cooking time, initial water temperature, final water temperature, and energy consumption have been measured for both of the heaters with the help of stopwatch, thermocouples, temperature recorders, and energy meters. The variation in percent energy save with amount of cooking and hence maximum percentage of energy save is also determined. The energy consumption in the heater is calculated using Eq. (1), E = V It

(1)

Here, E is the energy consumption in the heater in kilowatt-hour (kW h), V is the voltage difference across the heater in kilovolts (kV), I is the current flow in amperes (A) and t is the cooking/boiling time in hours (h).

Fig. 3 Experiment setup for the performance test

Improvement of Electric Heater Design for Household Cooking …

37

3 Results and Discussion The temperature profiles along radial direction at three different points (i.e., − 5.75 cm, 0 cm, and 5.75 cm from the center of the coil plate) are shown in Figs. 4 and 5 for both the insulated and conventional electric heaters, respectively. From Figs. 4 and 5, it is evident that the temperature at various points inside an insulated heater is more than that of the conventional one at the same point. For an example, in case of insulated heater (see Fig. 4), at 0 cm vertical distance from the coil plate, the average temperatures at radial positions (i.e., −5.75 cm, 0 cm, and 5.75 cm from the center of the coil plate) are 333.67 °C, 409 °C, and 342.33 °C, respectively, whereas in case of conventional one (see Fig. 5), the average temperatures are 309.33 °C, 380 °C, and 314.33 °C, respectively. This is due to the fact that some of the heat lost in the conventional heater is recovered by providing thermal insulation. Figures 4 and 5 also show that the temperature at the mid-point is always higher than that at Fig. 4 Temperature profile inside insulated electric heater

Fig. 5 Temperature profile inside conventional electric heater

38

A. K. Ghosh et al.

the sides for both types of heaters. This may be due to the higher coil density at the mid-section of the heater. It may further be observed that at points closer to the coil plate, the temperature is significantly more than that at distant points for both types of heaters. For example, as shown in Fig. 4, at 0 cm vertical distance from the coil plate, the average temperatures at radial positions (i.e., −5.75 cm, 0 cm, and 5.75 cm from the center of the coil plate) are 333.67 °C, 409 °C, and 342.33 °C, respectively, whereas at 4.5 cm vertical distance, at the same radial positions, the average temperatures are 269.33 °C, 313.67 °C, and 271 °C, respectively. This is due to the air gap that has a very low thermal conductivity (0.024 W/m K) and it acts as an insulator between utensils’ bottom surface and coil plate. Nevertheless, by observing these temperature profiles, it may be stated that heat energy and electrical energy can be saved if an electric heater is insulated and coated. Another observation is that the utensils will receive more heat if the distance between utensils’ bottom surface and coil plate is reduced. As a result, the cooking time will be less and energy will be saved. It is noteworthy that care should be taken so that the utensils’ bottom must not touch the coil surface to avoid accident. For the sake of performance tests and comparison between the conventional and insulated heater, three different food items, namely, Water, Minicate Rice, and Red Lentil have been used for several days. In each case, the percentage energy save is calculated from the observed information during cooking. As shown in Fig. 6, in case of water, it is observed that the insulated heater performs better compared to the conventional bare heater for raising the temperature of water to its boiling point. For each observation, 800 ml water is used as sample. By determining the energy consumption by both heaters, the percent energy save has been calculated. The insulated heater reduced heat losses, saved cooking time and, thereby, appreciable amount of energy is saved. Hence, from Fig. 6, it is observed that maximum energy save is 43.9% while the minimum energy save is around 15%.

Fig. 6 Energy consumption for boiling 800 ml of water

Improvement of Electric Heater Design for Household Cooking …

39

Similarly, performance tests have also been conducted for cooking 50 g Minicate rice along with 500 ml water (see Fig. 7). As shown in Fig. 7, maximum energy save is 32% and minimum energy save is 25% for cooking Minicate rice. Again for cooking 50 g red lentil along with 375 ml water, as shown in Fig. 8, maximum energy save is 23.07% and minimum energy save is 16%. In addition, Fig. 9 shows the energy consumption with the change in amount of boiled water, i.e., cooking time. Figure 10 shows the change in energy saving for the same case. It is evident from Fig. 9 that when amount of water is less, the difference between the energy consumptions by the conventional and insulated heater is less. With the increase in amount of cooking, i.e., amount of water (ml), this difference increases. As a result, when the amount of cooking is less, save in energy (%) is also less; when the amount increases, percent energy save also increases (as seen from

Fig. 7 Energy consumption for cooking 50 g Minicate rice with 500 ml of water

Fig. 8 Energy consumption for cooking 50 g red lentil with 375 ml of water

40

A. K. Ghosh et al.

Conventional Bare Heater

Fig. 9 Energy consumption with increasing amount of water (ml)

Insulated Heater

0.5 33.33%

Energy Consumption (kW-hr)

0.45

32.43%

0.4

31.25%

0.35 0.3

20.83% 19.05%

0.25

0.2 12.5% 0.15 7.69% 0.1 0.05 0

200

400

600

800 1000 1200 1400

Amount of Water (ml)

40

Fig. 10 % energy saved by insulated heater with increasing amount of water (ml)

ΔE/E (%) = -9eM^2 + 0.036M - 0.144

% Energy Saving

35 30 25 20 15 10 5 0

0

200

400

600

800 1000 1200 1400 1600

Amount of Water (ml)

Fig. 10). For example, while boiling 200 ml water, save in energy is 7.69% whereas while boiling 1400 ml water, save in energy is 33.33% (see Figs. 9 and 10). This is due to the fact that when cooking amount is less, then less cooking time is needed. As a result, heat entrapment inside the insulated electric heater is less for small period of time and, hence, save in energy (%) is not so significant. However, with the increase in the amount of cooking, the time period increases. As a result, heat entrapment occurs inside the heater for longer time and, hence, significant portion of energy is saved. From the results, it may thus be stated that the modified insulated electric heater is more efficient than conventional heater when the cooking time is high. However, the amount of energy saved is less when the volume of cooking is small. Nevertheless, the modified electric heater has great potential in solving energy crisis in developing countries.

Improvement of Electric Heater Design for Household Cooking …

41

4 Conclusions In this work, an attempt has been made to improve the energy utilization of a conventional electric heater. For that, a modified construction has been employed with glass wool insulation and a reflective coating of white cement. The temperature profile inside the heater refers to the significant amount of heat loss reduction. Three different food items, namely, Water, Minicate Rice, and Red Lentil, have been used to test the performance as well. Results indicate that maximum energy saving of 43.9% is possible for boiling 800 ml water, a maximum of 32% energy may be saved for cooking 50 g Minicate rice with 500 ml water, and a maximum of 23.07% energy may be saved for cooking 50 g red lentil with 375 ml water. Thus, the modified design has lower heat loss compared to the conventional design. The implementation of this modified design might be a simple and creative choice for energy conservation in household cooking applications in developing countries as well. In this study, glass wool insulation and reflective coating of white cement are used. As a future extension of this study, the effect of using only reflective material might also be studied since the construction will be much simpler and less costly.

References 1. Calhoun, K.: The Global Energy Crisis (2017) 2. Chu, S., Cui, Y., Liu, N.: The path towards sustainable energy. Nat. Mater. 16(1), 16 (2017) 3. Luthra, S., Kumar, S., Garg, D., Haleem, A.: Barriers to renewable/sustainable energy technologies adoption: Indian perspective. Renew. Sust. Energ. Rev. 41, 762–776 (2015) 4. Huber, M.: Theorizing energy geographies. Geo. Comp. 9(6), 327–338 (2015) 5. Kaygusuz, K.: Energy for sustainable development: a case of developing countries. Renew. Sust. Energ. Rev. 16(2), 1116–1126 (2012) 6. Wolfram, C., Shelef, O., Gertler, P.: How will energy demand develop in the developing world? J. Econ. Pers. 26(1), 119–138 (2012) 7. Asafu-Adjaye, J.: The relationship between energy consumption, energy prices and economic growth: time series evidence from Asian developing countries. Energ. Econ. 22(6), 615–625 (2000) 8. Daioglou, V., Van Ruijven, B.J., Van Vuuren, D.P.: Model projections for household energy use in developing countries. Energy 37(1), 601–615 (2012) 9. Leach, G.A.: Residential energy in the Third World. Annu. Rev. Energ. 13(1), 47–65 (1988) 10. Hou, B.D., Tang, X., Ma, C., Liu, L., Wei, Y.M., Liao, H.: Cooking fuel choice in rural China: Results from microdata. J. Clean. Prod. 142, 538–547 (2017) 11. Tamire, M., Addissie, A., Skovbjerg, S., Andersson, R., Lärstad, M.: Socio-cultural reasons and community perceptions regarding indoor cooking using biomass fuel and traditional stoves in rural Ethiopia: a qualitative study. Int. J. Env. Res. Pub. Healt. 15(9), 2035 (2018) 12. Pokharel, S.: Energy economics of cooking in households in Nepal. Energy 29(4), 547–559 (2004) 13. Anozie, A.N., Bakare, A.R., Sonibare, J.A., Oyebisi, T.O.: Evaluation of cooking energy cost, efficiency, impact on air pollution and policy in Nigeria. Energy 32(7), 1283–1290 (2007) 14. Cengel, Y.: Heat and mass transfer: fundamentals and applications. McGraw-Hill Higher Education (2014)

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank K. A. Alex Luke, J. Arun, R. Hemanth Prasanna and Ashish Selokar

Abstract Advancements in fuel curtailment have created a new perspective towards finding out new solutions to enhance the vehicle’s fuel consumption and emission diagnostics. The objective of sorting out new alternatives has led to the introduction of new formats of saving fuel. Brown’s gas, which involves the production of oxy-hydrogen gas (HHO) on demand by virtue of an electrolyzer, is one such alternative. In two-wheeler vehicles, the HHO module can be installed anywhere within the confines of the vehicle body; however, this pales out the design profile of the vehicle as it produces hindrance regarding space availability, visual appearance and physical tampering. After analyzing the body of a conventional two-wheeler, a novel design is presented where the fuel tank entails an auxiliary HHO generator in it without compromising any of the above-mentioned factors. An electrolyzer reactor with minimum possible volume factor was designed and fabricated. The unit was successfully installed within the left forward corner of the fuel tank and connected to the engine assembly. The test used to evaluate fuel economy here was an equivalent to the World Motorcycle Test Cycle (WMTC), where the drive cycle was performed with the complete motorcycle operated on a chassis dynamometer. The vehicle’s performance with and without the HHO generator was tested for increased fuel efficiency.

K. A. Alex Luke (B) Centre for Simulation and Engineering Design (SIMENDES), Hindustan Institute of Technology and Science, Chennai 603103, India e-mail: [email protected] J. Arun · R. Hemanth Prasanna · A. Selokar Department of Mechanical Engineering, Hindustan Institute of Technology and Science, Padur, Chennai 603103, India e-mail: [email protected] R. Hemanth Prasanna e-mail: [email protected] A. Selokar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_5

43

44

K. A. Alex Luke et al.

1 Introduction The knowledge and publications on oxy-hydrogen or popularly known as Brown’s gas (named after its inventor Yull Brown) are on the rise, due to its proved potential to run an internal combustion engine with higher efficiency and with least or no pollution to the environment. Numerous literatures from past three decades’ state its production methodologies, storage, merits and its limitations. HHO, a stoichiometrically formed gas, has the potential to augment the engine functionality and emission diagnostics by upgrading the combustion dynamics. It does so by injecting tiny atoms (relative to carbon-chained macromolecules) of HHO to the engine, completely surrounding the air/fuel mixture and enhancing the nature of combustion. More energy can be extracted out of the combustion process in terms of power which can propel the engine piston with greater frequency [1]. The engine torque is increased (than before) as more powerful combustion is now driving the piston thus increasing engine torque and brake thermal efficiency. This improves the energy factor being derived from the combustion of fuel which gets converted to mechanical terms more efficiently, thereby improving the overall engine power. Greater engine power without any doubt transcends the engine functionality. Specific fuel consumption value is decreased as improved combustion phenomenon utilizes lesser fuel for the same injection due to complete burn (very minimum unburned hydrocarbon of fuel) and combustion. The emission index regarding all the carbonyl compounds, NOx , HC, particulate matter, etc., also gets improved [2]. To add to it, a part of the product formed by the implosion of HHO gas is water which obviously gets exhausted out to the atmosphere for better climate balance. The fluctuating production of energy from solar or wind power devices has more losses in handling comparing to HHO energy, as there is no technology more efficient for matching its supply on demand criteria. Electrolysis is a known common method for the production of this gas, where a stoichiometric mixture of monatomic hydrogen (2/3rd volume) and oxygen (1/3rd volume); 55.55 mol of H2 and 27.775 mol of O2 formed after careful pulsed electrolysis process of water, KOH (potassium hydroxide) and NaOH (sodium hydroxide) is the best convenient catalyst to be used in this electrolysis process [3, 4]. Four-wheeled vehicles usually have more space for setting up of this generator without causing any hindrance to the vehicle aesthetics also to maintain it from tampering from outside forces. Considering the use of this technology on all internal combustion vehicles in the near future, a novel design criterion is laid out specially for two-wheeler to accommodate the HHO setup. The problem with a two-wheel-drive vehicle is that it does not entertain space availability for installing any auxiliary units (in this case HHO generator) for purposes related to increased vehicle performance. The vehicle is primarily covered with the chassis connectors all along its body and the only free space available for installing any extra unit is the area beneath the seating area and under the fuel tank which is almost occupied by the engine assembly (Fig. 1). Restrictions exist regarding finding out enough space for placing the generator within the two-wheeler vehicle structure without compromising the

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank

45

Fig. 1 Available space in a two-wheel motorbike for installing the HHO generator

aesthetic profile and open visibility to curtail damage to it. A novel HHO module was designed, fabricated, mounted to the vehicle and tested for fuel efficiency and emissions.

2 Literature Survey In 2006, White et al. demonstrated that the most progressive nature of H2 –O2 mixture used as a catalytic aid for gasoline run engines is that it has the ability to augment its performance and the regime of emission diagnostics [4]. Karagoz et al. [5] experimented with a gasoline engine injected with hydroxygen (H2 –O2 ) mixture and water to the intake manifold in order to study the performance and emission diagnostics. Three different combinations of mixtures 0, 3.75 and 7.5% of hydroxygen gas by volume were used for all the spectrum of speed range (1500–5000 rpm). It was found that brake power, brake thermal efficiency and oxides of nitrogen increased up to 11.7, 5.9 and 141.1% with HHO injection. Improvements showed up with coefficient of variation (COV), brake specific energy consumption (BSEC), total hydrocarbons (HC) and carbon monoxide (CO) by 15.2%, 5.6%. 74.5% and 59.5%, respectively, for hydroxygen addition. Yilmaz et al., in 2010, studied the dynamic behaviour of HHO gas being injected into a four-cylinder, four-stroke, compression ignition engine in improving the engine performance and emission nature. Equipped with hydrogen electronic control unit (HECU) based on 555 timers, the experimental setup found out a steady increase in the brake thermal efficiency above 1750 rpm engine speed. The CO and HC emissions were found to be reduced by an average of 13.5% and 5%, respectively.

46

K. A. Alex Luke et al.

Also, a considerable decrease of the specific fuel consumption (SFC) was found due to the uniform mixing of the HHO gas with air/fuel mixture [6]. Bari et al. in the year 2008 evaluated the performance enhancement of a conventional diesel engine through the addition of H2 /O2 mixture generated through water electrolysis. Experimental results indicated that hydroxygen addition acquired an average of 15.063% of fuel savings. The HC emissions dropped down by an average of 97.65%, and the CO emissions reduced by an average of 0.240% for three different readings. However, the NOx emissions were found to be increased by an average of 68 ppm for three different load readings [7]. A heavy-duty diesel engine emits a wide spectrum of harmful carbonyl compounds which not only lowers the effectiveness of diesel run engine but also promotes noxious environmental hazards. They are believed to be sources of ozone and peroxy nitrates precursors which have adverse negative effects on human health. Wang et al. investigated the emissions of carbonyl compounds from a diesel engine at low steady condition. Agilent HP 1100 HLDC/UV analyzer was used to identify the variety of carbonyl compounds from diesel engine. Formaldehyde was found to be the major one among all. With 10–40 L/min of H2 –O2 mixture addition, the emission of carbonyl compound formaldehyde was found to be decreased by 5.1– 31.7%. In addition to that, with H2 –O2 injection to the engine the brake specific fuel consumption (BSFC) was found to be decreased by an average of 7.95% for three different HHO injection rates [8]. HHO generation statistics suggest that with the current supply of 25 A, hydrogen-oxygen mixture can be generated by an amount of 1.88 L/min. A total of 1 L of water can produce 1860 L of HHO gas. This creates a total vehicle engine loss of 0.34 kW for powering the generator at 13.8 V [9]. Engine running on enriched hydrogen-aided gasoline also has quite influential effects on parameters such as flame speed and ignition delay. John. F. Cassidy in the year 1977 presented a NASA technical paper for the Society of Automotive Engineers (SAE) in which he put forward his experimental work of a multi-cylinder reciprocating engine using the combined mixture of gasoline and hydrogen in order to extend the lean operating range of the fuel. Hydrogen was produced using a methanol (CH3 OH) reformer and it was taken as an on-board source. With constant hydrogen flow rate of 0.64 kg/h, the apparent combustion flame speed was found to be 61% faster than normal. A sizeable reduction in ignition delay time was found with the addition of hydrogen for all equivalence ratio [10]. Hydrogen along with oxygen forms up a stoichiometric mixture which can be produced by variety of processes available. One of the most commonly used techniques is the carefully pulsed process of electrolysis of water. Alkaline water electrolysis shapes up as an easy technique for hydrogen production, having the advantage of simplicity. It has the potential to reduce energy consumption, cost and maintenance work which other processes involved. Also, it can increase reliability, durability and safety to the entire hydrogen generation process due to its simplicity [11]. Zoulias et al. presented a paper that discusses various ways by which hydrogen and oxygen mixture can be generated using electrolysis as the governing principle. A critical voltage of about 1.2 V is required to split water into H2 and O2 . Out of various modes of electrolysis such as alkaline electrolysis, PEM electrolysis, steam

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank

47

electrolysis, wind electrolysis, solar electrolysis and geothermal electrolysis, alkaline electrolysis was considered to be the most effective way of generating H2 –O2 mixture due to its incredibly simple, reliable and cost-effective construction [12].

3 Materials and Methods The method of generating HHO or Brown’s gas or oxy-hydrogen gas in an onboard vehicle involves using controlled current (done with the help of a modulator with varied frequency) from vehicles alternator-battery unit and discharging it to the electrolyzer unit to produce the oxy-hydrogen gas in the required quantity. The electrolyzer unit consists of a set of electrodes, arranged with proper configuration, distance and number (of plates), a container with distilled water and a suitable electrolyte, to which the controlled current is discharged, dissociating water into a stoichiometric mixture of oxygen and hydrogen (Fig. 2). This, when added to conventional fuel as a catalyst becomes powerful enough than gasoline to augment the combustion process, consequently engine torque and power. Given below is an equation of the conversion process: 2H2 O → 2H2 (g) + O2 (g); H2 O ↔ H+ (aq) + OH− (aq) [General Equation] 4OH− ↔ 2H2 O + O2 + 4e− [Anode, oxidation] [Cathode, reduction] 2H+ (aq) + 2e− ↔ H2 (g)

Fig. 2 Stack dimensioning of electrode plates

48

K. A. Alex Luke et al.

Fig. 3 Schematic diagram of HHO generator placed inside fuel tank

3.1 HHO Module Design Stack development is a detailed experimentation performed to position the electrodes with respect to each other to extract the maximum gas production out of them. For the proposed model, SS 316 electrode plates were used. Considering the fuel tanks available space, the dimensions of each electrode plates were 90 mm × 30 mm × 1 mm, a total of 12 plates (4 positive; 4 negative; 4 neutral) were placed inside the generator body spaced by 6 nos. neoprene gaskets of 5 mm between each other (Fig. 2b). The volume of the electrodes stack is calculated to be 135 cm3 , and the volume of total space available in the empty reactor is calculated to be 424 cm3 . Hence, the total space for the electrolyte is calculated to be ±290 mL inside the reactor. It was noted that a total volume of more than 195 mL of water-electrolyte mixture can be added to the generator for HHO gas production in this module. Previous researches state 1 L of water can produce about 1860 L of HHO gas during its operation of 3000 km without the need to replace the water for unwanted slag formation. Thus, if one litre of distilled water can produce 1860 L of HHO [9], hence, 0.195 L of injected water can support the primary fuel (gasoline) up to 362.7 L of HHO under this proposed model of Honda CB bike fuel tank. The generator was designed with minimum possible volume faction, where the volume of the generator/reactor was calculated to be 509 cm3 and volume of the bubbler tank was 168 cm3 . Total volume occupied by the generator and bubbler = (509 + 168) cm3 = 677 ≈ 680 cm3 . Figure 3 represents hydroxy gas production mechanism.

3.2 Tank Design and Development After carefully analyzing the body of a conventional two-wheeler, it was found out that the vehicle’s fuel tank (Fig. 4) entails enough space for placing an auxiliary unit

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank

49

Fig. 4 Closed installation of HHO module inside tank

like the HHO module which lays up an innovative solution for installation criteria in relation to aesthetics and open visibility. But this solution of covered installation shapes up a problem of tank’s volume reduction as its original capacity of storing fuel gets reduced. This creates a problem regarding the need to refuel with increased frequency. This improved combustion reduces the specific fuel consumption (SFC) factor of the vehicle by some reasonable degree. So even if the reduced volume of the tank demands frequents refuelling, the reduced SFC phenomenon balances that reduction. The problem of reduce tank capacity gets evenly balanced by this potent gas’s unique combustion behaviour. The test fuel tank used for this project was of HONDA CB unicorn bike with a fuel capacity of 13 and 1.3 L as reserve. The objective is to install a fabricated HHO generator inside the fuel tank for which it was required to cut it into two halves (Figs. 5a, b). One part was used completely to store fuel and a quantified amount of space within the second part for installing the generator unit. After successfully fabricating the generator with minimum possible volume factor, it was fitted inside the front portion of the fuel tank using chopped E-glass fibre with epoxy resin mixed with uniform mixture of resin, catalyst and an oxidizer. This choice of using plastic moulding as a cover for the generator comes with a decision of laying up a strong

Fig. 5 a Chopped E-glass epoxy on HHO module; b closed installation of HHO module inside tank

50

K. A. Alex Luke et al.

protective sheet of hardened surface over the body of the generator in order to arrest the various regimes of mechanical shocks, vibrations, road bumps and also avoids contact with gasoline inside the fuel tank. The plastic moulding laid up over the surface of the generator enables it to be chemically stable with the gasoline fuel within its confines. Plastic moulds have the characteristic property of being corrosion resistant, chemically stable and great soaking ability of shocks and vibrations, hence, used to cover the generator body provided great help owing to safety issues. After erecting the module inside the tank with fibre-reinforced plastic (FRP) mould the two cut parts of the tank were welded together by oxy-acetylene gas welding. Then the whole unit (tank and the installed generator) was taken for chassis dynamometer test trial for specific fuel consumption.

4 Water Flow Sensor Results and Discussions In this study, the Hero Honda brand Splendor bike was used for testing for fuel efficiency in chassis dynamometer by replacing the vehicles regular fuel tank with the new hybrid fuel tank setup, the method of testing implemented was identical to WMTC since no separate methodology of testing was available for two-wheeler. Cycle-1 of the WMTC—low speed testing a representative for urban traffic was considered for this test [13]. The objective was to test the mileage efficiency that the installed generator could provide for the vehicle. The generator’s HHO outlet tube was connected to the engine’s air filter inlet and the necessary electrical connections were made. Important among them was the connection of the pulsed-width modulator (PWM) used to vary and monitor the gas flow rate accordingly. Vehicle’s 12 V battery was used to power the generator. Critical study indicated that around 1.2 V of electrical input is required to split water into hydroxygen mixture. Approximate current given to the reactor was about 4 A. After powering on the unit, it was found that for every 65 mL of water contained within the reactor, approximately 0.92 L of HHO gas was produced. It was made sure with the help of the pulse dozer that no HHO gas should be produced below speed 1500 rpm as it blocks the intake manifold of the engine due to its over-availability. Only at speeds above 1500–1600 rpm the pulsed modem with its 555 timer most arrangement allows current from the battery to the generator. The combined unit was used for real-time odometer testing without and with HHO module and results were tabulated (Tables 1 and 2). From the above list of tables, it was calculated that without injected HHO test gave an average covered distance of 7.05 km considering all the five readings and the second trial of injected HHO test (Fig. 6) gave an average reading of 8.655 km. • Without injected HHO, average distance travelled = 7.05 km. • With injected HHO, average distance travelled = 8.655 km. × 100 = 18.544%. • Efficiency (%) = 8.655−7.05 8.655

Embodiment of an Efficient Brown’s Gas Compound Fuel Tank

51

Table 1 Tabulation of results obtained without injected HHO Without injected HHO S. no

Amount of fuel consumed (ml)

Speed (km/hr)

Odometer readings (km)

Distance covered (km)

1

50

0–60

0–2.1

2.1

2

100

0–60

2.1–6.3

4.2

3

150

0–60

6.3–12.1

5.8

4

200

0–60

12.1–20.3

8.2

5

250

0–60

20.3–30.6

10.3

6

300

0–60

30.6–42.3

11.7

Table 2 Tabulation of results obtained with injected HHO With injected HHO S. no.

Amount of fuel consumed (ml)

Speed (km/hr)

Odometer readings (km)

Distance covered (km)

1

50

0–60

0–3.2

3.2

2

100

0–60

3.2–8.8

5.6

3

150

0–60

8.8–16

7.21

4

200

0–60

16–26.1

10.12

5

250

0–60

26.1–38.7

12.6

6

300

0–60

38.7–51.9

13.2

Fig. 6 Comparison analysis of the HHO injection

5 Conclusions Based on the discussion following conclusions are made from the present work: • This paper suggests that Brown’s gas can be used as an additive or a catalyst to augment the engine functionality by performance.

52

K. A. Alex Luke et al.

• It can be safely mixed with conventional fuel devoid of any catastrophic imbalance and accident as it is a production on demand technology. • The electrolyzer unit can be fabricated with minimum possible volume factor, small enough to mount it inside the vehicle’s fuel tank without much hindrance. • The HHO unit was successfully installed within the left forward corner of the tank and it was successfully tested for increased fuel efficiency. • Future researchers can, however, try to fabricate micro/mini-sized HHO reactors and contribute towards minimum volume consumption within the body of the fuel tank. • Also, the challenge remains in a question, whether this HHO gas can be used as a complete fuel having the capability to propel an engine entirely on its own. If in future, researchers can explore the extended quest of this gas to run an engine solely with it, then without any doubt this technology will be a great leap forward in taking transportation, climate balance and the great oil crisis to a new untouched level.

References 1. Sánchez Jalet, J.C., Casanova Treto, P., Solís Ramírez, K.: J. Engg. 6(6), 54–58 (2016) 2. De Silva, T.S., Senevirathne, L., Warnasooriya, T.D.: Euro. J. Adv. Engg. Tech. 2(4) 1–7 (2015) 3. Ymamoto, H.: Explanation of Anomalous Combustion of Brown’s Gas using Dr. Mills’ Hydrino Theory, (No. 1999-01-3325), SAE Technical Paper, pp. 1–6 (1999) 4. White, C.M., Steeper, R.R., Lutz, A.E.: Inter. J. Hydrogen Energy 31(10), 129–1305 (2006) 5. Karagöz, Y., Yüksek, L., Sandalcı, T., Dalkılıc, A.S.: Inter. J. Hydrogen Energy 40(1), 692–702 (2015) 6. Yilmaz, A.C., Uludamar, E., Aydin, K.: Inter. J. Hydrogen Energy 35(20), 11366–11372 (2010) 7. Bari, S., Esmaeil, M.M.: Fuel 89(2), 378–383 (2010) 8. Wang, H.K., Chen, K.S., Lin, Y.C.: Aerosol Air. Qual. Res. 13, 1790–1795 (2013) 9. Keršys, A., Kalisinskas, D., Pukalskas, S., Vilkauskas, A., Keršys, R., Makara’s, R.: Eksploatacja i Niezawodno´sc´ 15(4) 384–389 (2013) 10. Cassidy, J.F.: Emissions and Total Energy Consumption of a Multicylinder Piston Engine Running on Gasoline and a Hydrogen-Gasoline Mixture, NASA TN D-8487, Washington D.C. (1977) 11. Zeng, K., Zhang, D.: Prog. Energy Combu. Sci. 36(3), 307–326 (2010) 12. Zoulias, E., Varkaraki, E., Lymberopoulos, N., Christodoulou, C.N., Karagiorgis, G.N.: Tcjst 4(2), 41–71 (2004) 13. Steven, H.: Worldwide harmonised motorcycle emissions certification procedure. Inst. Veh. Technol. (2002)

Automated Solar Photovoltaic Panel Cleaning/Cooling System Using Air–Water Mixture and Sustainable Solutions to Off-Grid Electrification Nithin Sha Najeeb, Prashant Kumar Soori and Iyad Al Madanat

Abstract Solar energy has enormous potential to fulfil the energy requirements of the world and can be extracted using solar cells. However, the solar cells are affected by poor efficiency and further affected by wind speed, orientation of the panel, temperature and dust deposition. There are different cleaning technologies devised by many industry experts to clean the solar panels. However, they are facing drawbacks when deployed in the solar farms. An efficient cleaning system, along with an added cooling system, must be devised so that the solar panels must be cleaned and cooled to maximize the energy production. This paper presents a low-cost, fully automated, smart, innovative dust cleaning and cooling system for photovoltaic (PV) panels. The system is designed, fabricated, fully automated using programmable logic controller (PLC) and tested successfully. The automation procedure is explained in detail. A battery-charging kiosk, capable of charging two, 24 V lead–acid batteries embedded within this prototype, shall provide clean energy in a sustainable manner to the rural communities of the developing nations. The user can check the status of the battery such as battery voltage, battery temperature and state of charge on the human–machine interface (HMI) panel while charging the batteries.

1 Introduction The energy consumption in Middle East has increased because of the growth in the energy demand [1]. Solar energy is an abundant energy resource available on the earth. The electricity generation using solar panel is the best and promising N. S. Najeeb · P. K. Soori (B) Heriot Watt University, Dubai, UAE e-mail: [email protected] N. S. Najeeb e-mail: [email protected] I. Al Madanat Phoenix Contact Middle East, Dubai, UAE e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_6

53

54

N. S. Najeeb et al.

alternative in the Middle East. Though the solar energy is easily available, however, its utilization using solar panel is affected by many factors such as dust accumulation or soiling and increase in cell temperature. Dust accumulation or soiling is one of the prime factors which affects the performance of the solar panels in the Middle East and African regions [2]. Dust can be classified as particles which are having diameters less than 500 µm [3]. The reduction in the performance of the solar panel is not uniform, because the dust accumulation is not uniform, and it is highly dependent on the geographical site conditions [4]. Reports have been published, wherein it was showed that the loss of power production due to the soiling issues occurs mainly in the solar energy systems such as PV or concentrated solar power (CSP) plants [5, 6]. The dust deposited on the solar panels in the solar farm block the radiation from the sun reaching the solar panel. The researchers have reported that the dust accumulation depends on wind speed, atmospheric dust concentration, location and the cleaning frequency of the solar panels [7]. In general, more the dust accumulated on the solar panel, greater the reduction in the power output from the solar panel. The power output of the solar PV panel is directly proportional to the intensity of sunlight striking the panel’s surface. The harsh climatic conditions along with high temperature as well as humidity also play a vital role in the effective utilization of solar energy in these regions. The concern was raised for solar panels placed in dry semi-arid areas (gulf regions) because of the three environmental factors which resulted in the reduction in the power output from the solar panel: High ambient temperature, High humidity and Dust storms [8, 9]. The location of solar panels in areas which receive more sunlight is usually placed in arid or semi-arid areas, and this leads to the loss of 1% power reduction due to dust accumulation [10]. In dusty desert areas where the solar irradiation is highly available, dust deposition, hindrance of solar irradiation due to passing of clouds causes reduction in power production [11]. There are other factors which act a barrier to the performance of the solar panel. The factors are wind speed, energy conversion devices, temperature, humidity, regional characteristics such as plants, traffic and air pollution play an important role in dust deposition [12]. To improve the efficiency of the solar panel, the solar module must be cleaned with a very efficient cleaning technique. Desert storms along with light rains are a common phenomenon in desert environment which will lead to the formation of cemented particles on the top of the solar panel, thereby blocking the solar irradiance from reaching the solar cell [13]. In dry months, when there is no rainfall, the dust accumulation tends to increase and this leads to formation of cemented particles on the panel, thereby reducing the performance of the PV panels. Hence, for the solar panel to produce a stable output, the panel should be kept clean to absorb the incoming radiation. There are many methods available in the market for cleaning the PV panels. Existing methods are manual cleaning, use of water sprinklers, use of robotics for cleaning, brushing and wiping, anti-soiling coating and electrodynamic screens. Manual cleaning comes with the disadvantage that it is time-consuming and is labour-intensive method [14]. The chances of scratches on the panel are very high during the cleaning process. Sufficient quantity of water must be carried in a truck for cleaning the solar panel. Robotic cleaning has a drawback in which it takes more time to clean the

Automated Solar Photovoltaic Panel …

55

Table 1 Panel cleaning methods S. no

Methods

Drawbacks

1

Manual cleaning

Labour intensive Time consuming

2

Mechanical removal of dust by brushing and wiping, water sprinklers, water jets, ultrasonic driving

Surfaces of the panel may be damaged More water consumption Movements of mechanical parts on the surface of the panel

3

Use of robots

Expensive Needs periodic battery charging and replacements Time consuming

4

Electrostatic removal of dust

Not safe Not practicable with dust which are stuck to the panel

solar array. The reason being, these devices must move vertically and horizontally for cleaning the panels [13]. These devices operate using battery and need periodic battery replacement. Sprinkler system has a shortcoming in which it consumes more water for cleaning the solar panels. The solar panels which are usually in the desert regions have a problem with the availability of water, and hence, this is not a viable method. The electrodynamic screen uses high voltage to remove the dust from the solar panel. The high voltage is used in generating the electric field which produces a travelling wave, thereby pushing the dust away from the solar panels [15, 16]. Solar panel cleaning using detergents is time consuming, costly, and it can even corrode the frames of the solar panel [17]. The researchers have summarized the cleaning methods used in most of the PV power plants in Thailand and reported that manual cleaning method by washing with water is still employed [18]. Various cleaning methods proposed by various researchers are summarized in Table 1 [13, 14, 19, 20]. The researchers reported that increase in cell temperature results in lesser power production [21, 22]. The photovoltaic efficiency depends on the current and voltage of the solar panel which depends on the temperature. This leads to the linear reduction in performance as temperature increases [23]. The solar module when absorbs the radiation, the temperature increases which will result in decrease in efficiency. The researchers concluded that cell efficiency decreases by 0.45% for every degree rise in temperature [24]. The reduction of cell temperature is therefore necessary in maximizing the power output of the solar panel. The cooling of solar panel is only way to reduce the temperature rise of the solar panels. Additionally, the increase in temperature is also a threat to the panels as it leads to high densities of current and heat fluxes. Proper cooling technique is therefore a necessity in preventing the performance degradation of PV panels which shall bring irreversible damage. This paper explains a fully automated cleaning mechanism which uses pressurized air and water to clean the dusty solar panels as well as to cool down the panels in the

56

N. S. Najeeb et al.

event of high temperature. The prototype system is designed, fabricated and tested successfully at Heriot-Watt University, Dubai. The proposed method of cleaning and cooling the PV panel presented in this research work has no guard rails, less operator intervention, no need of any selfcleaning, no moving parts on the panel surface, and it can also withstand harsh climatic conditions (especially in the Middle East). The system is designed, fabricated, fully automated using PLC (ILC 171 ETH, Phoenix Contact) and tested successfully at Heriot-Watt University, Dubai. The system is expected to act as a dual-purpose technology: cleaning and cooling jobs.

2 Proposed Methodology In the proposed method, the mixture of air and water in a designed pressure is used to clean and cool the dusty PV panels along with the application of anti-soiling coating. The air compressor and water pump are used to maintain the designed pressure for cleaning. To automate the process of cleaning, the solenoid valves are connected to air compressor and water pump, respectively. During the process of pressure buildup in the air and water lines, the solenoid valves remain closed. At the time of cleaning/cooling, solenoid valves will open for the specified period of time to allow water and air to flow to the nozzle. The hydrophobic coating is applied on the panel for easy rolling of dirt along with water on the solar panel at the time of cleaning. In the initial phase of this research work, the solar panels are coated with a thin layer of anti-soiling coating before spraying the air–water mixture. The panel is cleaned using isopropyl alcohol, followed by coating applied in circular motions on the panel. The panel is left undisturbed for allowing the bonding of the coating to form on the solar panel. Later, using a soft cotton cloth, the panel is wiped out until no visible films are seen on the panel. The solar panel is installed in the farm. This hydrophobic coating is applied on the panel to prevent the dirt from sticking to the solar panel. During the process of cleaning and cooling the dusty panels, the air and water are fed into a common point where it can mix in a designed pressure before spraying on the dusty solar panel. The mixture is sprayed through a specially designed nozzle resulting in flat-fan spray which helps in more area coverage with less amount of water consumption. In this type of nozzle, fluid (in the form of droplets of water which are more or less consistent like sheet of waterfall) is moulded into a fan formed sheet of liquid. The angle of spraying ranges from 15° to 145°, thereby covering more area. The pressure for air and water is regulated using air and water pressure regulators. The flow rate of spraying the mixture can be adjusted using the nozzle. Depending on the flow rate adjustment, the spray mixture can be controlled, thereby cleaning the dusty solar panel. To the inlet of the nozzle, 6-mm-diameter hose is connected for both air and water outlets, and an area of 0.7 m2 is covered using less amount of water. The pictorial representation of the fabricated prototype is depicted in Fig. 1.

Automated Solar Photovoltaic Panel …

57

Fig. 1 Pictorial representation of the fabricated prototype

The system is integrated with automation procedures using PLC (ILC 171 ETH, Phoenix Contact) to perform the cleaning process. The duration of the whole process is programmed to be 240 s in which 60 s is where the actual cleaning process takes place. The cleaning time can be adjusted according to the site conditions. The entire process is programmed into PLC using ladder logic. A user-friendly design approach has been incorporated using HMI panel as an interface to control and monitor the progress of the entire cleaning/cooling operation. The air compressor and water pump are connected through a pressure regulator to regulate the pressure and then fed into the pneumatic nozzle which is placed on the top of the dusty solar panel. The solenoid valve is used for automation purpose where they remain closed at the time when the air line and water line are getting pressurized. The solenoid valves are used instead of already built-in manual valves. The control logic of the solar panel cleaning and the components used in the prototype working model is highlighted in Fig. 2.

2.1 Automation of Solar Panel Cleaning and Cooling System The prototype of solar panel cleaning and cooling system is built and automated using PLC (ILC 171 ETH). The user interface is incorporated using an HMI panel. The whole process runs on the timer basis, and the signal from the PLC to the field components is activated based on the time programmed into the PLC. In total, it has three commands: the first one for the air compressor, the second one for water pump and the third one for solenoid valves of both air and water lines. These commands are given by the PLC during the entire cleaning process. When the command is given by the user from the HMI panel, the PLC gives a signal to the air compressor, and the air compressor switches on (Fig. 3). The compressor tank builds pressure of the air up to 8 bar, and the compressor switches off due to the internal circuitry. The time taken by the air compressor to build pressure of the air up to 8 bar is designed for 120 s. When the air compressor has built the required pressure, the PLC is programmed to send a signal to turn on the water pump (Fig. 3). The water pump will then turn

58

N. S. Najeeb et al.

Fig. 2 Control logic of solar panel cleaning system

Fig. 3 Solenoid valve active status on the HMI panel

on, and the water line will be pressurized (4 Bar). The time taken for water line to get pressurized is programmed for 60 s. After 180 s, the signals are given to the solenoid valves simultaneously, to remain open for 60 s. The entire duration for cleaning process is 240 s (or 4 min). The status of components in operation displayed on the HMI panel is depicted in Fig. 3. Green colour in Fig. 3 indicates the system in operation. The time duration of cleaning process can be adjusted according to the site conditions. This is possible by changing the ladder logic program of the PLC. The ladder logic sequence for cooling process is same as that of the cleaning process of the solar panel. The PLC ladder logic program was developed using PC Worx (Phoenix Contact) software. The developed control logic is given in Fig. 4. The ILC 171 ETH PLC also comes with a real-time clock facility feature. The use of real-time clock facility feature is that the PLC on its own can initiate process without the user giving the command through HMI panel. Figure 5 shows the func-

Automated Solar Photovoltaic Panel …

59

Fig. 4 Developed ladder logic control schematics for solar panel cleaning/cooling system

Fig. 5 Functional block of the programmed real-time clock feature

tional block schematics implemented in the ladder logic program for the activation of real-time clock feature of ILC 171 ETH PLC module. The advantage of this feature is that, without an operator giving a command, the process of cleaning and cooling can take place at the programmed time to the PLC. Hence, operator does not need to be in the control room. In this research work, the real-time clock is programmed to be 12 o’clock noon time.

3 Results and Discussion Two identical test benches were set up in Heriot-Watt University, Dubai Campus, one with cleaned panel and another with uncleaned solar panel. Identical panels and loads were chosen for collecting the data. The chosen panel was a polycrystalline 60 W solar panel. The specification of the panel is depicted in Table 2.

60

N. S. Najeeb et al.

Table 2 Specification of the solar panel

Solar panel

Specification

Maximum power (Pmax )

60 ± 3% W

Maximum power current (Imp )

3.3 A

Maximum power voltage (Vmp )

18 V

Short-circuit current (Isc )

3.68 A

Open-circuit voltage (Voc )

21.2 V

Maximum system voltage

DC 1000 V

Maximum series current

10 A

Nominal operating temperature (NOCT)

−45–80 °C

According to the site conditions, the panel is kept at 50° facing south-east. The voltage (V) and current (A) readings were taken at every half an hour interval for cleaned and uncleaned panel of the same ratings. The two set-up benches were tested with identical load. Figure 6 depicts the connection diagram of the test rig up. The results of the tests are given in Tables 3 and 4. It is evident from the results presented in Tables 3 and 4 that there is a significant power loss when the solar panel is left uncleaned. This justifies the need for cleaning the panels periodically to capture the clean energy. The readings from the same identical panels showed that, with the same site conditions, the power produced is significantly more with the cleaned panel than that obtained with the uncleaned panel. At 12 noon, which is the peak time for power production, the power output of the cleaned panel is observed to be 32% more than the uncleaned panel. The test was conducted repeatedly for a period of six months to see the change in the variation of Fig. 6 Test set-up of two identical test benches

Table 3 Results with cleaned panel S. no

Time

Voltage (V)

Current (A)

Power (W)

1

11:00 am

12.71

2.08

26.43

2

11:30 am

15.17

2.31

35.04

3

12:00 pm

17.41

2.41

41.95

4

12:30 pm

15.50

2.75

35.65

5

01:00 pm

12.00

2.10

25.20

Automated Solar Photovoltaic Panel …

61

Table 4 Results with uncleaned panel S. no

Time

Current (A)

Power (W)

1

11:00 am

Voltage (V) 8.20

2.60

21.32

2

11:30 pm

12.10

2.00

24.20

3

12:00 pm

15.78

2.0

31.56

4

12:30 pm

13.0

2.1

27.31

5

01:00 pm

1.8

16.74

9.30

power produced before cleaning and after cleaning the solar panels. It was observed that test results were very similar to the one obtained earlier. To investigate the importance of cooling the panels, again the same two identical test benches were used. However, fabricated system using air–water mixture was deployed to clean and cool the panels. The results are tabulated in Tables 5 and 6. It can be clearly observed from Tables 5 and 6 that, the solar panels when placed outside during the peak hours, the power production from the solar panel reduces due to rise in temperature. This is also one of the contributing factors in the reduction of power production in addition to soiling and other related issues. The air–water mixture, when sprayed on the panel using the prototype, resulted increased power output from the panel as well as reduction in temperature. Hence, it is evident from the results that the air–water mixture is also acting as a coolant for the solar panels. The cooling timings can be set using PLC timer. It is observed from Tables 5 and 6 that the periodic cooling of the panel shall reduce the cell temperature and enhance the performance output. Table 5 Results without cooling S. no

Time

Voltage (V)

Current (A)

Power (W)

Temperature (°C)

1

2:00 pm

7.1

1.4

9.9

31.9

2

2:15 pm

8.0

1.5

12.0

45.0

3

2:30 pm

7.5

1.4

11.2

47.0

4

2:45 pm

7.0

1.4

9.8

45.8

5

3:00 pm

6.4

1.4

8.9

47.1

Table 6 Results after cooling using the fabricated prototype (spraying air–water mixture using the prototype) S. no

Time

Voltage (V)

Current (A)

Power (W)

Temperature (°C)

1

2:00 pm

7.9

1.5

12

31.9

2

2:15 pm

8.4

1.6

13.4

29.0

3

2:30 pm

8.3

1.6

13.2

30.0

4

2:45 pm

7.6

1.5

11.4

33.1

5

3:00 pm

6.8

1.4

9.6

32.2

62

N. S. Najeeb et al.

The energy consumption of the various auxiliaries used in the fabricated prototype is presented in Table 7. The total energy consumed for the entire cleaning and cooling process per cycle is observed to be 0.0197 kWh. The water consumption for a period of 60 s is approximately 500 mL per cycle. The water consumption will change according to duration of the cleaning process programmed. The energy consumption may vary according to the different components and according to the usage depending upon the site conditions in the solar farm. The fabricated prototype is shown in Fig. 7. The entire mechanical parts such as compressor, water pump, solenoid valves are placed in the bottom portion of the fabricated system. PLC and power supply unit are in the middle of the assembled unit. The cleaning process of the panel was tested by collecting sand from the surrounding areas and manually spraying the dust on the panel for dust deposition. After manually dusting the panel, the air–water mixture was sprayed on the top of the panel to see the effectiveness of the designed cleaning process of the panels. The prototype is also designed to charge two 24 V lead–acid batteries in addition to its cooling and cleaning job. The user-friendly design approach is used in designing battery-charging kiosk. Using our battery-charging kiosk, the user can check the status of the battery (battery voltage, battery temperature, state of charge) on HMI Table 7 Energy consumption of the components used Equipment

Power rating (W)

Operating time (minutes)

Energy consumed (kWh)

Air compressor

550

2

0.018

Water pump

80

1

0.0013

Solenoid valve

12

1

0.0002

Solenoid valve

12

1

0.0002

Total consumption

Fig. 7 Fabricated prototype

0.0197

Automated Solar Photovoltaic Panel …

63

Fig. 8 Battery-charging status displayed on the HMI panel

panel while charging the batteries. In the fabricated prototype, the batteries are monitored using PLC via an intelligent uninterruptible power supply (UPS) module. The UPS module can measure these three performance parameters of the battery placed for charging in the kiosk. The same is depicted in Fig. 8. Further validation of the result is done with practical measurement. The kiosk was tested by charging two lead–acid batteries from the solar panel and later measured the voltage of the charged batteries using digital multimeter. It was observed that the digital multimeter showed the same reading (24 V) as displayed on the HMI panel.

4 Conclusions The new innovative idea of solar panel cleaning and cooling system presented in this paper is a promising technology which can wash away all the industry facing issues. The energy consumption for this new technology is relatively less 0.0197 kWh per cycle because there are no moving parts associated with the solar panel cleaning compared to the present day installed technologies. The fabricated prototype has cleaned majority of the panel’s surface. The cleanliness depends on the position of the nozzle on the panel to be cleaned and as well as the orientation of the panel. However, the air–water mixture acting as a coolant was proven to be successful. With the present set-up, water consumption was observed to be 500 mL per cycle. Future work shall include the use of multiple nozzles placed on the panel or horizontal and vertical movement of a nozzle if single nozzle is used. The study will be conducted on the application of superhydrophobic coating on the panel. The cleaning process can be further extended to number of panels by mounting the nozzle on the solar panels. The compressor, pump and solenoid valves will be at a central control station and nozzles can be mounted on the panels. Either by programming using real-time clock feature of the PLC or by operator giving the command from the central station, the same methodology can be used to clean multiple panels.

64

N. S. Najeeb et al.

The cleaning and cooling method presented in this paper has numerous advantages: The operator intervention is less, there are no guard rails, no need of any periodic battery replacement, requires no self-cleaning, requires very less maintenance, and it can withstand harsh climatic condition like in the Middle East and African regions. The proposed methodology shall act as dual-purpose technology, wherein it is used to clean the solar panels, and at the same time, it can be used as a coolant for the solar panels which are situated in the hot, dry environments. In the fabricated prototype, two lead–acid batteries have also been configured for charging. This charging opens an eye for building charging station in the rural areas of the developing nations. In developing nations, community people still use traditional ways of lighting which brings lot of health-related problems to the people. By incorporating such charging stations, people can use light-emitting diode (LED) lamps for lighting purposes, thereby putting an end to the traditional methods of lighting in rural areas. The carbon emission from the rural places can be easily reduced, introduction of such kiosk will wave goodbye to all the health-related issues, fire hazards, and it will improve the quality of life of the people. Acknowledgements The authors would like to express sincere thanks to Expo-Live University Innovation Programme, Expo2020 Dubai Initiative for providing the funding for the development of this project. Special thanks to Ms. Fatma Ibrahim, Assistant Manager EXPO Live and her colleagues for their help and support.

References 1. Sadorsky, P.: Trade and energy consumption in the Middle East. Energy Econ. 33(5), 739–749 (2011) 2. Costa, C.S., Diniz, A.S.A, Kazmersk, L.L.: Dust and soiling issues and impacts relating to solar energy systems: literature review update for 2012–2015. Renew. Sustain. Energy Rev. 63, 33–61 (2016) 3. Mani, M., Pillai, R.: Impact of dust on solar photovoltaic (PV) performance: research status. Renew. Sustain. Energy Rev. 14(9), 3124–3131 (2010) 4. Walwil, H.M., Said, S.A.: Fundamental studies on dust fouling effects on PV module performance. Solar Energy 107, 328–337 (2014) 5. Suellen, A., Costa, C.: Dust and soiling issues and impacts relating to solar energy systems. Renew. Sustain. Energy Rev. 63, 33–61 (2016) 6. A comprehensive review of the impact of dust on the use of solar energy: history, investigations, results, literature and mitigation approaches. Renew. Sustain. Energy Rev. 22, 698–733 (2013) 7. Sayyah, A., Horenstein, M.N., Mazumder, M.K.: Energy yield loss caused by dust deposition on photovoltaic panels. Sol. Energy 107, 576–604 (2014) 8. Said, S.A., Walwil, H.M.: Fundamental studies on dust fouling effects on PV module performance. Solar Energy 328–337 (2014) 9. Alnaser, N.W., Othman, M., Dakhel, A.A., Bataresh, I., Lee, Jk, Najimali, S., Alothman, A., Alshowaikh, H., Alnaseer, W.E.: Comparison between performance of man-made and naturally cleaned PV panels in a middle of a desert. Renew. Sustain. Energy Rev. 82(1), 1048–1055 (2018) 10. Chesnutt, J.K.W., Ashkanani, H., Guo, B., Wu, C.Y.: Simulation of microscale particle interactions for optimization of an electrodynamic dust shield to clean desert dust from solar panel. Sol. Energy 155, 1197–1207 (2017)

Automated Solar Photovoltaic Panel …

65

11. Mazumder, M.K., Sharma, R., Biris, A.S., Horenstein, M.N., Zhang, J., Ishihara, H., Stark, J.W., Blumenthal, S., Sadder O.: Electrostatic removal of particles and its applications to selfcleaning solar panels and solar concentrators. Developments in Surface Contamination and Cleaning, pp. 149–199 (2011) 12. Arabatzis, I., Todorova, N., Fasaki, I., Tsesmeli, C., Peppas, A., Li, W.X., Zhao, Z.: Photocatalytic, self-cleaning, antireflective coating for photovoltaic panels: Characterization and monitoring in real conditions. Sol. Energy 159(1), 251–259 (2018) 13. Deb, D., Brahmbhatt, N.L.: Review of yield increase of solar panels through soiling prevention, and a proposed water-free automated cleaning solution. Renew. Sustain. Energy Rev. 82(3), 3306–3313 (2018) 14. Zhou, Z.L.C.: Review of self-cleaning method for solar cell array. In Procedia Engineering (2011) 15. Kawamoto, T.S.H.: Electrostatic cleaning system for removal of sand from solar panels (2013) 16. Vasiljev, P., Borodinas, S., Bareikis, R., Struckas, A.: Ultrasonic system for solar panel cleaning. Sens. Actuators, A 74–78, 200 (2013) 17. Syafiq, A., Pandey, A.K., Adzman, N.N., AbdRahim, N.: Advances in approaches and methods for self-cleaning of solar photovoltaic panels. Sol. Energy 162, 597–619 (2018) 18. Sakarapunthip, N., Chenvidhya, D., Chuangchote, S., Kirtikara, K., Chenvidhya, T., Onreabroy, W.: Effects of dust accumulation and module cleaning on performance ratio of solar rooftop system and solar power plants. Jpn. J. Appl. Phys. 56(82), 08ME02 (2017). Available: https:// doi.org/10.7567/jjap.56.08me02 19. Mishra, A., Sarathe, A.: Study of solar panel cleaning system to enhance the performance of solar system. Int. J. Emerg. Technol. Innov. Res. 4(9), 84–89 (2017) 20. Patil, P.A., Bagi, J.S., Wagh, M.M.: A review on cleaning mechanism of solar photovoltaic panel, presented at the 2017. In: International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS). Chennai, India, 17859170 (2017) 21. Al Hanai, T., Hashim, R.B., El Chaar, L., Lamont, L.A.: Environmental effects on a grid connected 900 W photovoltaic thin-film 2615–2622 (2011) 22. AlamImteaz, M., Ahsan, A.: Solar panels: real efficiencies, potential productions and payback periods for major Australian cities. Sustain. Energy Technol. Environ. 25, 119–125 (2018) 23. Skoplaki, E., Palyvos, J.A.: On the temperature dependent of photovoltaic module electrical performance: a review of effective/power correlations. Solar Energy 83, 613–624 (2009) 24. Wu, S.-Y., Chen, C., Xiao, L.: A review of solar photovoltaic panel cooling systems with special reference to ground coupled central panel cooling system (GC-CPCS). Renew. Energy 125, 936–946 (2018)

Design and Fabrication of Four-Way Multi-hacksaw Cutting Machine Anupoju Sai Vamsi, Chiranjeeva Rao Seela and Arnipalli Naveen

Abstract In the scenario, for cutting pipes, more human effort is required and the time consumption is more. So, in order to minimize the time and labour, in this work, efforts are made to design and fabricate a four-way multi-hacksaw cutting machine. The main objective of the work is to design an efficient four-way multi-hacksaw cutting machine which suits the industrial applications like cutting thin metal bars, wood, PVC pipes, etc. The machine is designed and fabricated using the principle of slider-crank mechanism, where the slider is replaced with hacksaw. Here, a disctype crank operates the hacksaw cutter to reciprocate in the guideway to ensure the cutting operation. The size and shape of the machine were finalized by considering the proper design and kinematic analysis. Also, the project is aimed to prepare a machine with less vibration, easy in operation and easy in mobility. The suggested machine with eight simultaneously cuts can improve the rate of cutting.

1 Introduction A mechanism in a device designed to transform input forces and movement into a desired set of output forces and movement. In this project, we are dealing with the single slider-crank mechanism that converts rotary motion into reciprocating motion where the slider is replaced with the hacksaw to get the required cutting. Here, manual operation is possible by using chain set mechanism. So that when the crank wheel in chain mechanism is rotated, then free wheel rotates Al pipe attached to it with high speed. Providing a lock between Al pipe and shaft in circular disc makes slider-crank mechanism. Here, the crank with shaft rotates and will make the hacksaw cutter to A. S. Vamsi · C. R. Seela (B) · A. Naveen Department of Mechanical Engineering, GMRIT, Rajam, Andhra Pradesh, India e-mail: [email protected] A. S. Vamsi e-mail: [email protected] A. Naveen e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_7

67

68

A. S. Vamsi et al.

reciprocate in the guideway where the cutting operation is done. The fabrication of the different components ensures the smooth working condition and future automation of different mechanical linkages [1]. A pedal operated hacksaw machine does not require any specific input. In this mechanism, chain drive is directly connected to the hacksaw cutter to get the required cuts of wooden blocks [2–4]. The human efforts can be reduced by a detailed theoretical analysis of the four-way hacksaw machine in which the rotary motion is converted to reciprocating motion of hacksaw cutter from where cutting operation will be done [5]. A detailed analysis of the power and torque is required to identify the type of failure mode [6, 7]. Modelling software like solid edge can be used for the kinematic analysis of the hacksaw cutter [8]. Fabricating the two-way pedal powered hacksaw machine and minimization of the power and the vibrations created by the motor are very essential [9, 10]. Also, a proper cutting fluid helps in reducing the heat generation [11].

2 Fabrication The equipment consists of a hacksaw blade, handle, frame and a wing nut. The hacksaw cutter consists of either solid or adjustable frame as shown in Fig. 1a, b, respectively. Solid-type frame can adopt only one particular length of blade, whereas in the adjustable frame different standard lengths of blades can be fitted. A hacksaw of 30 cm, a single-phase induction motor of 1440 rpm and 0.25 Hp and a circular disc of 16 cm were used in the fabrication. A shaft is attached at 5 cm from centre of a circular disc. The frame of dimensions 90 cm × 90 cm × 90 cm is used as base frame. Figure 2 shows the CAD model of base frame. The connecting rod (link 2) acts as a mechanical linkage between the crank and slider as in Fig. 3 and is used to convert the rotary motion of crank to reciprocating motion of the hacksaw. A guideway was made from a pipe of length 3.175 cm to facilitate the slider to reciprocate as in Fig. 5. The pipe holder made from iron rods of 30 × 10 cm (H × L) with a distance of 10 cm between them was attached to the outer frame as in Fig. 4. The AC motor of capacity 0.25 Hp @1440 rpm is placed vertically on the constructed outer frame and a hollow disc of diameter 16 cm is mounted on it with permanent welding. Then, the hacksaws which are arranged in the guideways were connected to the disc through the linkages as in Fig. 6.

Fig. 1 a Solid hacksaw frame and b adjustable hacksaw frame

Design and Fabrication of Four-Way Multi-hacksaw … Fig. 2 CAD model for base frame

Fig. 3 Connecting rod

Fig. 4 Pipe holders

69

70

A. S. Vamsi et al.

Fig. 5 Guideways

Fig. 6 Designed machine

The pipe holders as in Fig. 4 can hold the pipe of diameter up to 10 cm. Guideways in Fig. 5 have 0.5 inch clearance so that the hacksaw cutters will reciprocate to get the required cutting operation. Designed machine in Fig. 6 is the final assembly.

3 Working The working model was prepared as it will be operated both in automatic and manual mode. Figure 7a, b illustrates the mode of operation with block diagram. The detailed working of this machine is explained in Sect. 3.1.

Design and Fabrication of Four-Way Multi-hacksaw …

71

Fig. 7 a Automatic operation and b manual operation

Fig. 8 Slider-crank mechanism

3.1 Working Principle Here, the motion of the shaft attached to the disc and rotating with eccentricity will be converted into reciprocating motion of the hacksaw cutter by slider-crank mechanism as shown in Fig. 8. So, the four-way multi-hacksaw reciprocates in accordance with the shaft. A 220 v supply was given to the AC motor to rotate the disc. The eccentric rotation of the shaft causes the reciprocating motion of the hacksaw cutters along the guideways and ensures the required cutting of the material. Also, a provision for the manual operation was given on the top of the machine such that we can rotate the mechanism even in the absence of the power source. The rate of cutting of the material depends upon the speed and here, speed can be regulated by using variac.

4 Design Calculations The angle between the hacksaw should be 360 (where N = no of blades). A N 30 cm long blade of height of 10 cm was used. ii. Guideway size = slider size +0.25 inch (where clearance is 0.25 inch). So, guideway size is 1.25 inch iii. Design of connecting rod: i.

72

A. S. Vamsi et al.

Apply right angle law that l is connecting rod length and a is distance of slider from support rod and b is distance from shaft to slider a 2 + b2 = l 2 Distance between centre to guideway supporting rod is 40 cm and supporting rod to guideway is 10 cm. The shaft attached to circular disc has link of length 32 cm. The length of the connecting rod is: a 2 + b2 = l 2 152 + 502 = l 2 l = 52.2 cm As link attached to shaft is 32.2 cm then l = 52.2 − 32.2 = 20 cm Therefore, length of connecting rod is 20 cm iv. Design of drill holes size: The holes’ size was selected on the basis of strength of pipe material. The strength of galvanized steel pipes is 379.21 N/mm2 . Therefore, 8 mm holes are required of diameter v. Selection of motor: A pipe of diameter = 10 mm and required speed is 90 rpm Let τ be shear stress, T is Torque, J is polar moment of inertia and R be the radius. π × (D 4 − d 4 ) As we know that, TJ = Rτ . τ of pvc pipes is 104 N/mm2 and J = 32 and R = 5 mm π 32

T 104 =  4 4 5 × 10 − 7

T = 15517.42 N-mm = 5.52 N-m P=

2π (90)(15.52) 2π N T = = 146.27 watts = 0.2 hp 60 60

To cut eight pipes, required pipes = 8P = 1.6 hp. Considering a motor of 0.25 HP = 186 watts 186 =

2π (1440)(T ) 60

T = 1.23N/m As speed decreases, torque will increase so speed obtained after load is 90 rpm 186 =

2π (90)(T ) 60

T = 20.420 N/m

Design and Fabrication of Four-Way Multi-hacksaw …

73

vi. Design of guideway support rods: As hacksaw length is 30 cm and length of stroke is 10 cm, the guideways should be minimum of 40 cm but when the hacksaw moves 10 cm forward there may be a chance of harming. So, considering safety, we have taken 60 cm for guideway supporting rod. vii. Selection of material: The available materials are

Square pipes

Minimum tensile strength N/mm2

Cost (per kg)

Galvanized steel pipes

379.21

70

Steel pipes

289.57

50

Stainless steel pipes

558.47

250

As the corrosion resistance, strength and availability of galvanized steel is more, it was used for the base frame. Also, for the manual operation, the aluminium pipe was used because of its lighter weight. viii. Calculation of chain length: For a chain, stay length is 63.635 cm, biggest front gear number of teeth is 50 and biggest rear gear number of teeth is 28. Therefore, chain length is 71 links as shown in Fig. 9. ix. Pipe and its lock for manual operation: At the motor shaft, a pipe of length 35 cm and 1 inch diameter made from aluminium is placed vertically and connected to free wheel as in Fig. 10. As the length of the stroke is 10 cm, pipe is placed with 5 cm distance from centre of the circular disc. x. Kinematic analysis: The velocity diagram is given in Fig. 11 and the analysis is given in Table 1. As all hacksaws are equal, hacksaw will have a velocity of 2.481 m/s. xi. Design of pipe holder: Holding pipe is made from iron of diameter 1 cm (length = 33 cm). Fig. 9 Chain set-up

74

A. S. Vamsi et al.

Fig. 10 Pipe lock

Fig. 11 Velocity diagram

Table 1 Kinematic analysis S. no

Velocity vector

Magnitude

Direction

Sense

1

Vao = Va = oa

30 × 0.0942 = 2.826 m/s

Perpendicular to OA

Parallel to AB and towards A

2

Vba = ab



Perpendicular to AB



3

Vbg = Vb = gb



Parallel to motion of slider

Towards B

xii. Selection of circular disc: The machine is designed to cut the pipes up to the diameter 10 mm. So, the length of stroke should be 10 mm. Therefore, crank of 10 mm was selected.

Design and Fabrication of Four-Way Multi-hacksaw …

75

5 Cost Analysis Overall weight Total zinc pipes cost Motor 0.25 hp—1440 rpm Aluminium pipe Cycle cranks and free wheels and chains Wheels Paints and other materials Iron rods cost Disc Bolts and nuts Blades Lubricant Overall cost Time required to cut one pipe Therefore, time taken to cut two pipes of large and small simultaneously

50 kg 3250/1600/50/300/300/500/1000/150/170/80/100/7500/4 s for pvc pipe. 32 s

6 SWOT Analysis Strengths: • • • • • •

Reduce the work of labour High production rate Low initial cost and less weight Maintenance cost is low and skilled person is not required Simple construction and will work under all atmospheric conditions It is able to cut eight pieces simultaneously without any jerks and vibrations and hence requires less time. Weaknesses:

• Special device should be used to cut different materials and for the variation of speed. • Stress concentration in the threaded portions under variable load conditions in screwed joints will affect the life of the machine. Opportunities: • Construction industry • Mass production

76

A. S. Vamsi et al.

• Engineering industry • Workshops. Threats: • Occupies more space • Noise in operation.

7 Conclusions The following are the conclusions: After the completion of the project, we are able to cut eight numbers of pipes. So, this machine is more suggestible in mass production where continuous pipe cutting is done and the machine of low cost with simple design is fabricated. This machine reduces human effort and satisfied day-to-day household needs. It is used in industrial applications. The overall goal of this project is to design a machine where high productivity should be there. Therefore, four-way multi-hacksaw cutting machine using the slider-crank mechanism is fabricated and achieved the required stroke and velocity to cut pipes.

References 1. Patil, D., Raut, S., Jadhav, S., Kulkarni, G., Shinde.: Optimization and development of multiway hacksaw machine. Int. J. Res. Sci. Eng. 2(3), 799–804 (2016) 2. Subash, R., Meenakshi, C.M., Jayakaran, K.S., Venkateswaran, C., Sasidharan, R.: Fabrication of pedal powered hacksaw using dual chain drive. Int. J. Eng. Technol. 3(2), 220–223 (2014) 3. Vinod Babu, C., Chiranjeeva Rao, S., Vykunta Rao, M.: Modal and static analysis of automotive chassis frame by using FEA. Int. J. Appl. Eng. Res. 10(20), 19775–19777 (2015) 4. Vinod Babu, C., Chiranjeeva Rao, S., VykuntaRao, M.: Structural analysis of Eicher 11.10 chassis frame. Int. J. Eng. Trends Technol. 22(7), 315–319 (2015) 5. Rao, P.K.V.: Fabrication of hacksaw cutter using slider cranks mechanism. Int. J. Eng. Trends Technol. 6(3), 10–19 (2016) 6. Biswasa, A., Salunkeb, A., Patilc, A., Patil, D.: A review on multiway hacksaw machine. Int. J. Innov. Emerg. Res. Engineering. 4(3), 92–96 (2017) 7. Josh, T.: Automated double hacksaw cutter. Int. J. Eng. Res. Technol. 7(7), 49–56 (2018) 8. Mates: Design and fabrication of fourway hacksaw machine. Int. J. Civ., Mech. Energy Sci. 1, 5–8 (2017) 9. Satishkumar, S.S., Senapati, A.K., Pal, S.K., Mohanty, S.: Fabrication of two-way pedal powered hacksaw machine. Int. J. Innov. Sci. Eng. Technol. 3(3), 683–687 (2016) 10. Vykunta Rao, M., Chiranjeeva Rao, S., Vinod Babu, C., Sekhar Babu, M.V.: Influence of reinforced particles on the mechanical properties of aluminium based metal matrix composite— a review. Chem. Sci. Rev. Lett. 4(13), 335–341 (2015) 11. VenkataRao, D., Prasad Rao, K., Chiranjeeva Rao, S., Umamaheswara Rao, R.: Design and fabrication of power generation system using speed breaker. Int. J. Curr. Eng. Technol. 4(4), 2697–2702 (2014)

Adhesion Strength of Plasma Sprayed Coatings—A Review Abhinav, Harish Kumar Kustagi and Arun R. Shankar

Abstract In the present study, a critical review of the adhesion behavior of plasma sprayed coatings is discussed. A study revealed that substrate preparation, coating technique, surface roughness, solidification mechanism, coating thickness, plasma temperature, the coefficient of thermal expansion, of the top coat and bond coat have a significant influence on the adhesion strength. A great deal of discussion is made on the mechanical strength of the plasma sprayed coatings.

1 Introduction The tensile strength is the preliminary requirement of any plasma sprayed coatings. The tensile adhesion test is a widely accepted adhesion test method carried out according to the ASTM C633 standard in the determination of adhesion strength of plasma sprayed coatings [1]. In the test, the coated sample is glued to another similarly coated counterpart and then tested in tension in a universal testing machine. The value of the tensile load, in which the separation of the two coated parts occurs, is registered and transformed into an adhesion value and can be calculated using load divided by area relation. The section at which fracture takes place shows the nature and characteristics of the failure mechanism. In the duplex thermal barrier coating system, the coatings are likely to fail from the bond coat (BC), top coat (TC) or at the BC/TC periphery.

Abhinav (B) · H. K. Kustagi · A. R. Shankar Alliance College of Engineering and Design, Alliance University, Bangalore, India e-mail: [email protected] H. K. Kustagi e-mail: [email protected] A. R. Shankar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_8

77

78

Abhinav et al.

In this paper, a critical review has been done on the adhesion strength of the plasma sprayed coating. Also, a method to conduct adhesion strength of plasma sprayed coatings, parameters affecting the adhesion strength of plasma sprayed coatings, selection criterion in plasma sprayed coatings and works in the related field extensively discussed.

1.1 Selection Criterion in Plasma Sprayed Coatings The criterion for obtaining good adhesion is based on the following three main rules. (a) To achieve a satisfactory matching of thermal expansion coefficient, it is essential either to allow the chemical composition of the coating to vary or to employ multilayers of different compositions and properties. (b) Wettability can be varied by introducing suitable additives that may affect the surface tension of the molten fluid during the enameling period and consequently the contact angle between coating and substrate. (c) To achieve the best affinity between a coating and a substrate, the former can be subjected to doping to avoid excessive ion diffusion. Care must be taken to prevent or minimize the risk that mechanical properties at the interface may decay as a consequence of the changes introduced into the system because such changes are conducive to more or less marked variations of the chemical and physical properties of the coating. The adhesion strength is found not to exceed l00 MPa in atmospheric-pressure plasma spraying coatings, where the adhesion mechanism is governed mostly by mechanical anchorage. It has been found that in the case of Low Pressure Plasma Spraying technique of diffusion bonding, or metallurgical bonding, develops a leading mechanism, since the substrate may be heated to an elevated temperature without any corrosion. The use of epoxy resin glue in the adhesion strength above 70 MPa is not recommended by the conventional method. The determination of high adhesion strength using LPPS coating technique reported is difficult [2]. The tensile strength and resistance against shear loads are the criterion for some adhesion experiments [3–6] although arguments about the experimental techniques and specific outcomes of each one are unambiguous [7–9]. Takeda [10] has developed a technique to quantify the adhesion strength. A thick molten glue deposits on the substrate, and then, it is machined to create a test sample similar to that of tensile specimen used in UTM. Finally, it was subjected to the tensile test to determine the bonding strength of the coating.

Adhesion Strength of Plasma Sprayed Coatings—A Review

79

2 Adhesion Strength of Plasma Sprayed Coatings Bond strength (adhesive bonding) is one of the critical properties of plasma sprayed coatings. The primary mechanism of adhesion can be described as mechanical keying, physical and chemical diffusive [11–13]. According to Sobolev et al. [13], mechanical interlocking improved due to high surface roughness caused due to the solidified droplet formed at the lower part of the splat and established to be a prominent mechanism behind the bond strength. Another critical factor which is found to affect the bond strength is the coadherence of the distinct coating grains, which is measured by the degree of flattening grains at the time of deposition, and hence by the velocity and temperature of the particle and that of the temperature of the substrate [14]. It has also been shown with the help of TEM investigations that an amorphous film ranging in thickness from 50–1000 nm is formed at the TC/BC interface. This thin film is a chemical compound which affects the mechanical strength of TBCs [15, 16]. Issues such as the effect of oxidation processes, coating morphology and residual stresses [14–16] have to be considered to assess the bond strength of plasma sprayed coatings. It is also found to be dependent on the combination of specimen cleaning and blasting technique, coating materials, coating spray parameters and environmental conditions. The composition of the substrate, and that of the BC as well, is found to affect the adhesion of alumina formed during the oxidation of bond. With the spalling of alumina, zirconia layer is seen to disintegrate contributing to the failure of TBCs. Segregation, migration and stress generation are the significant matters of concern in the thermally grown oxide (TGO) regions. Due to the concentration, gradient migration of aluminum ions from BC to metal alloy substrate arises. A similar effect has been observed from both the substrate and BC into the TGO. It seems that restricted yttrium ion (Y+) migrated to the alumina improves mechanical bonding; however, impurities in the alumina commonly lead to supplementary stresses, which reduce the adhesion during thermal exposure [17]. High-temperature bulk Y ions diffuse into the YSZ surface and found to destabilize the TC [18, 19] enabling the transformation of Zr phase to monoclinic phase [20] resulting in a detrimental microscopic volume change which may lead to coating spallation. Earlier researchers have also suggested that both neutral and ionic aluminum diffusion into the YSZ layer induces spallation [21]. Diffusion-controlled migration in the TBC is found to promote harmful phase transitions besides enhancing the thermal stresses. Though a concentration gradient is exposed to facilitate the movement via dispersion, the oxygen chemical potential gradient, occurring during oxidation, is found to supply another driving force. Studies have also shown that the potential chemical gradient penetrates into the TC, BC and the specimen (metal substrate). It also affects the reactive oxygen species in all these layers, including those which were initially in the form of carbides, stable oxides, nitrides and sulfides [22–25]. For example, sulfur is found to segregate the interface, which can prove to be detrimental to the life of TBC. The interfacial sulfur is found to increase the thickness of the TGO layer, decrease the mechanical bonding of the oxide layer to the metal, besides enhancing

80

Abhinav et al.

the transformation of metastable alumina to the α-phase and hence formation at the interface, within the oxide layer and TGO. It also increases the formation of micro-voids and interfacial coarsening. The creation of micro-voids is found to be challenging since they act as stress risers within the oxidation layer [23, 26–30]. Desulphurization of the specimen to less than 1 ppm may assist in preventing this spallation [30–33]. It has also been found that the existence of other species such as Y and Zr in the BC and substrate might suppress the detrimental effects of sulfur, thus reducing the desulphurization infractions [23, 34–37]. Further, the scale stresses resulting from the isothermally generated stress during the oxide formation or from uneven cooling stress owing to thermal expansion inconsistency during the thermal cycling process [13]. Sobolev et al. [38] have conducted experiments to analyze the factors which would impact adhesion strength. The magnitude of corrosion product and the degree of the residual stresses is found to be unfavorable for good mechanical bonding. It is found that both should be at the highest levels in the case of fine coatings which possess the second lowest levels of adherence among the IN625 coatings [39]. Residual stresses are also found to be high for graded and intermediate coats, both with little oxide proportion and with extraordinary adhesive strength. It has been observed that ceramic adhesion increases by increasing the particle velocity as well as an increase in the thermal energy of droplet. The poor inter-splat mechanical interaction is found in the case of uneven ceramic coatings. This was the reason attributed to the particle velocities, which are at the lower end of the plasma spray spectrum, thereby resulting in poor mechanical bonding with the substrate. The plasma temperature is also found to be a contributing factor. Fine and mixed composite coatings have small microporosities as compared to granular composite coating and display a very little or no inter-splat separation at all in microhardness test, thereby indicating higher particle velocities. These factors indirectly suggest that oxide content, possibly present as a coagulated surface layer formed before particle impact, is a key factor contributing to low adhesion levels. Another possible reason found to affect the adhesion is the amalgamation of splat size and the substrate surface history. An accusation for the existence of surface peak spacing and splat size ideal combination has also been found [40, 41], where it has been experimentally revealed, increase in the mechanical interlocking comparably with the average surface roughness but is maximum at a specific ratio of grain size to peak topography. In order to enhance the adhesion strength of ceramics, a range of techniques have been reported, one among is the application of silane and also by tailoring the composition of a ceramic [42]. According to the literature, it has been found that hydrofluoric and phosphoric acid etching followed by silane coupling greatly improve the resin ceramic strength [43–45]. A similar kind of work is reported by Romanini-Junior DDSA et al. on adhesive/silane application on bond strength durability to a lithium disilicate ceramic and found promising results [46, 47]. A great deal of review work has been carried out on the effects of lasers on bond strength to ceramic materials and found that the use of lasers significantly improves the adhesion strength of composite to the ceramic surface. The CO2 laser irradiation applied for 10 s to ceramic surfaces found positive and enhanced results [48]. From

Adhesion Strength of Plasma Sprayed Coatings—A Review

81

the above reviews on the adhesion strength, it is recommended to follow the standard protocols intended for specific applications.

3 Conclusions The following important conclusions can be drawn from the adhesion study of plasma sprayed coatings: 1. Adhesion mechanism is governed mostly by mechanical anchorage, and improved adhesion can be achieved by having high surface roughness. 2. Micro-voids are found to be challenging since they act as stress risers and may reduce the adhesion strength. 3. Plasma temperature during the coating process, corrosion product, blasting technique, coating materials, coating composition, coating spray parameters and environmental conditions has a significant role in deciding the adhesion strength of composite materials. Above said parameters must be carefully designed to have the highest degree of adhesion strength. 4. Secondary process, viz. laser technique and the following standard adhesion protocols intended for specific application can significantly improve the adhesion strength of composite to ceramics.

References 1. Standard test method for adhesion and cohesion strength of thermal spray coating, ASTM standard C-633-01, ASTM, West Conshocken, PA, USA (2001) 2. Smeggil, J.G., Funkenbusch, A.W., Bornstein, N.S.: Metal Trans. 17A, 923 (1986) 3. Pint, B.A., Wright, I.G., Lee, W.Y., Zhang, Y., Prubner, K., Alexander, K.B.: Mat. Sci. Eng. A245, 201 (1998) 4. Sarioglu, C., Blachre, J.R., Pettit, F.S., Meier, G.H.: Microscopy of oxidation 3. In: Ewcomb, S.B., Little, J.A. (eds.) Institute of Metals, p. 4150. London, United Kingdom (1997) 5. Sobolev, V.V.: J. Therm. Spray Technol. 9(1), 100–106 (2000) 6. Gnaeupel-Herold, T., Prask, H.J., Barker, J., Biancaniello, F.S.: Microstructure, mechanical properties, and adhesion in IN625 air plasma sprayed coatings. Mater. Sci. Eng. A 421, 77–85 (2006) 7. Bahbou, M.F.: J. Therm. Spray Technol. 13(4), 508–514 (2004) 8. Gawne, D.T., Griffiths, B.J., Dong, G.: Trans. Inst. Met. Finish. 75(6), 205–207 (1997) 9. Jackson, M., Rairden, J., Smith, J., Smith, R.: J. Met. 33, 23 (1981) 10. Takeda, K., Ito, Michihisa, Takeuchi, Sunao: Properties of coatings and applications of low pressure plasma spray. Pure Appl. Chem. 62(9), 1772–1782 (1990) 11. Krishnan, R., Dash, S., Kesavamoorthy, R., Babu Rao, C., Tyagi, A.K., Raj, B.: Laser surface modification and characterization of air plasma sprayed alumina coatings. Surf. Coat Technol. 200, 2791–2799 (2006) 12. Spraying, T.: Practice, theory and application. Miami FL Am. Weld. Soc. (1985) 13. Sobolev, V.V., Guilemany, J.M., Nutting, J., Miguel, J.R.: Int. Mater. Rev. 42(3), 117 (1997)

82

Abhinav et al.

14. Houben, J.M.: Relationship between the Adhesion of Plasma Sprayed Coatings to the Process Parameters Size, Velocity and Heat Content of the Spray Particles, Ph.D. thesis. Eindhoven University of Technology, Eindhoven, the Netherlands (1988) 15. Harmsworth, P.D., Stevens, R.: Microstructure of zirconia-yttria plasma-sprayed thermal barrier coatings. J. Mat. Sci. 27, 616–624 (1992) 16. Bartuli, C., Bertamini, L., Matera, S., Sturlese, S.: Investigation of the formation of an amorphous film at the ZrO2 –Y2 O3 /NiCoCrAlY interface of thermal barrier coatings produced by plasma spraying. Mater. Sci. Eng. A 199, 229–237 (1995) 17. Bahbou, M.F., Nylén, P., Wigren, J.: J. Therm. Spray Technol. 3(4), 508 (2004) 18. Berndt, C.C.: Cracking processes in thermally sprayed ceramic coatings. Mater. Sci. Forum 34–36, 457–46 (1988) 19. Era, H., Otsubo, F., Uchida, T., Fukuda, S., Kishitake, K.: Mater. Sci. Eng. A Struct. Mater. Prop. Microstruct. Process. 251, 166 (1998) 20. Zhu, Y.L., Ma, S.N., Xu, B.S.J.: Therm. Spray Technol. 8(2), 328 (1999) 21. Siegman, S., Dvorak, M., Grutzner, H., Nassenstein, K., Walter, A.: In: Lugscheider, E. (ed) Thermal Spray Connects: Explore its Surface Potential, Proceedings of the International Thermal Spray Conference ITSC, p. 823. ASM International/DVS, Dusseldorf, Basel, Switzerland, Germany (2005) 22. Hadad, M., Marot, G., Lesage, J., Michler, J., Siegmann, S.: In: Lugscheider, E. (ed.) Thermal Spray Connects: Explore its Surface Potential, Proceedings of the International Thermal Spray Conference ITSC, p. 759. ASM International/DVS, Dusseldorf, Basel, Switzerland, Germany (2005) 23. Trevisan, R.E., Fals, H.C., Lima, C.R.C.: Inf. Tecnol. Chile 11(4), 103 (2000) 24. Kuenzly, J.D., Douglass, D.L.: Oxid. Met. 8, 139–178 (1974) 25. Theunissen, G.S.A.M., Winnubst, A.J.A., Burggraaf, A.J.: Surfaces and Interfaces of Ceramic Materials, pp. 365–372. Kluwer (1989) 26. Schindler, K., Schmeisser, D., Vohrer, U., Wiemhoffer, H.D., Gopel, W.: Sens. Actuators 17, 555 (1989) 27. Ingo, G.M., Padeletti, G.: Surf. Interface Anal. 21, 450 (1994) 28. Daloe, J., Boone, D.: Failure mechanisms of coating systems applied to advanced turbine engine components. In: Proceedings 42nd ASME Gas Turbine and Aero Engineering Congress Orlando (1997) 29. Pint, B.A., GarratReed, A.J., Hobbs, L.W.: Mater. High Temp. 13, 3 (1995) 30. Pint, B.A.: Oxid. Met. 45(1) (1996) 31. Hou, P.Y., Prubner, K., Fairbrother, D.H., Roberts, J.G.: Scripta Mater. 40, 241 (1998) 32. Hou, P.Y., Stringer, J.: Oxide. Met. 38, 323 (1992) 33. Pint, B.A.: MRS Bull. 19, 26 (1994) 34. Clyne, T.W., Humphreys, C.J.: Improvements in plasma sprayed thermal barrier coatings for use in advanced gas turbines. Department of Materials Science and Metallurgy. University of Cambridge, Pembroke Street, Cambridge CB23QZ 35. Grabke, H.J., Kurbatov, G., Schmutzler, H.J.: Oxide. Met. 43, 97 (1995) 36. Prubner, K., Schumann, E., Ruhle, M.: In: Shores, D.A. et al. (ed.) Fundamental Aspects of High Temperature Corrosion VI, pp. 344–356. Electrochemical Society, Pennington (1996) 37. Pint, B.A.: Oxid. Met. 48, 303 (1997) 38. Allen, W.P., Bornstein, N.S.: In: Dahotre, N., et al. (eds.) High Temperature Coatings I, pp. 193– 202. TMS, Warren dale, PA (1995) 39. Meier, G.H., Pettit, F.S., Smialek, J.L.: Mater. Corros. 46, 232 (1995) 40. Smith, M.A., Frazier, W.E., Pregger, B.A.: Mater. Sci. Eng. A 203, 388 (1995) 41. Standard test method for adhesion and cohesion strength of thermal spray coating, ASTM standard C-633–01, ASTM, West Conshocken, PA, USA (2001) 42. Uzun, I.H., MalkocË, M.A., Polat, N.T., Ö˘greten, A.T.: The effect of repair protocols on shear bond strength to zirconia and veneering porcelain. J Adhes Sci. Technol. 30, 1741–1753 (2016) 43. Stella, J.P., Oliveira, A.B., Nojima, L.I., Marquezan, M.: Four chemical methods of porcelain conditioning and their influence over bond strength and surface integrity. Dent. Press J. Orthod. 20, 51–56 (2015)

Adhesion Strength of Plasma Sprayed Coatings—A Review

83

44. Guimaraes, H.A.B., Cardoso, P.C., et al.: Simplified surface treatments for ceramic cementation: use of universal adhesive and self-etching ceramic primer. Int. J. Biomater. (2018) 45. AlRabiah, M., Labban, N., et al.: Bond strength and durability of universal adhesive agents with lithium disilicate ceramics: a shear bond strength study. J. Adhes. Sci. Technol. 32(6) (2018) 46. Romanini-Junior, J.C., Kumagai, R.Y., et al.: Adhesive/silane application effects on bond strength durability to a lithium disilicate ceramic. J. Esthet. Restor. Dent. 30(4) May (2018) 47. Paula, A., Magalhaes, R., et al.: Simplified surface treatments for ceramic cementation: use of universal adhesive and self-etching ceramic primer. Int. Biomater. (2018) 48. Garcia-Sanz, V., Paredes-Gallardo, V., et al.: The effects of lasers on bond strength to ceramic materials: a systematic review and meta-analysis. Int. J. PLoS ONE 13(1) (2018)

An Experimental Studies on the Polymer Hybrid Composites—Effect of Fibers on Characterization M. Ashok Kumar, K. Mallikarjuna, P. V. Sanjeev Kumar and P. Hari Sankar

Abstract The present research focused on the polymer hybrid composite fabrication and its characterization. Kevlar fibers (also called Aramid fibers, KF) are mixed with Sansevieria trifasciata fiber (snake plant leaf fibers, STF) to improve the performance of the epoxy matrix. Former fiber is synthetic fiber and the later is natural fibers are combined proportionately by the rule of mixtures KF and STF treated fiber systems. Wet-hand layup was used to organize systems with weight ratios of KF/STF for treated, viz. 1:0; 0.5:0.5; 0.75:0.25; 0.25:0.75; 0:1 (typically named as A, B, C, D, and E systems from the left). It was found that tensile strength for system-D (treated) was found improvement due to the fact that dust-free, rough, and improved surface area. Impact strength was found significant for the system-D when compared with others. The interface and voids at the fracture surface were improved for the systems C and D which were observed from the SEM images. Chemical resistance found good all the samples except carbon tetrachloride due to the hit of carbon atoms which consequently imparted erosion of the fiber out of the matrix.

M. Ashok Kumar School of Mechanical Engineering, Rajeev Gandhi Memorial College of Engineering and Technology (Autonomous), Nandyala, A.P, India e-mail: [email protected] K. Mallikarjuna (B) Department of Mechanical Engineering, G.Pullaiah College of Engineering and Technology (Autonomous), Kurnool, AP, India e-mail: [email protected] P. V. Sanjeev Kumar Department of Mechanical Engineering, Annamacharya Institute of Technology and Sciences, Rajampeta, Kadapa, AP 516126, India P. Hari Sankar Department of Mechanical Engineering, G. Pulla Reddy Engineering College, Kurnool, AP, India © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_9

85

86

M. Ashok Kumar et al.

1 Introduction Polymers are taking their lead in all fields due to their unique properties. The lighter the weight of the product not only save material, but also improves fuel economy. For example, 10% of the vehicle weight shrinks, and then a corresponding ratio of fuel is saved. These materials demand lightweight and strength are the two parameters that are essential. Polymers/fibers are the very light metal (ρ = 1 – 1.5 g/cm3 ) probably comes out lightweight composite at the end. According to the stress, strain diagram slope of the curve increases strength, stiffness increases on other hand, and toughness decreases due to the decrease in ductile nature. Fiber-reinforced composites have drawn more attention as there are used for weight reduction applications due to their excellent strength and stiffness which comes with less weight. Natural fibers are used as reinforcing agents over the years and more recently these have used with synthetic fiber in order to improve specific properties that cannot be obtained from the individual fibers. These fibers have renewable, nice appeal, less density, high specific volume, biodegradability less cost, etc., is made them to versatile candidates for specific application. So far flax, hemp, jute, straw, wood fiber, rice husks, wheat, barley, oats, rye, cane (sugar and bamboo), grass reeds, kenaf, ramie, oil palm empty fruit bunch, sisal, coir, water hyacinth, pennywort, kapok, papermulberry, raphia, banana fiber, pineapple leaf fiber, and papyrus. Manufacturers are paying attention in incorporating natural fiber composites into both internal as well as external parts of the transportation vehicles. This serves two-fold goal of the companies to lower the overall weight of the vehicle, thus, ever-increasing fuel efficiency in addition to increase the sustainability of the mechanized process. Mercedes Benz, Toyota, and Daimler Chrysler are already started and looking to expand [1–8]. The advantage of these fibers can cost as less as $0.5/kg, whereas synthetic fibers cost $2.50/kg and their densities (1.15 – 1.50 g/cm3 ) are coming with even lesser than the synthetic fibers. Strength and stiffness are not high when compared with synthetic fibers. Kevlar fiber is a temperature resistance, high rigidity modulus, high young’s modulus, and strong synthetic fiber and it was commercially used as a replacement of steel in racing tires. These fibers are used for a lot of applications, ranging from bicycle tires and racing sails to bulletproof vests; due to their high tensile strength to weight ratio by this way, it is five times stronger than steel [9]. It is also used modern marching drumheads that withstand high impact loads. Having a higher weight of the Kevlar fibers made them to be balanced with the natural fiber which has less weight imparts them more surface area consequently corresponding bonding strength is the indication of the inbuilt mechanical strength in the composites. Therefore, Sansevieria trifasciata is chosen as reinforcement as it is very compatible. Current research is focused on the epoxy as matrix, STF/KF as a reinforcing hybrid fiber is chosen to fabricate the polymer hybrid composites and followed by tensile strength, impact strength, DSC, TGA, and SEM measurements evaluated to assess the material to suit specific applications.

An Experimental Studies on the Polymer …

87

2 Materials and Methods Epoxy, hardener, Kevlar fibers, and all the chemicals were employed form the Sriram composites from the Hyderabad. STF stuff was received from the formers from Enumuladoddi village, Anantapur district, Andhra Pradesh, India. Glass molds were prepared as per the ASTM standards. Then, finally, composites specimen cut with the help of cutter on par with the standards. Dumbbell shape specimens have been cut to test the tensile strength by using the UTM. Tensile strength test specimens (100 × 20 × 3 mm, ASTM D638) and impact strength test specimens (63.5 × 12.7 × 12.7 mm; ASTM D256) are prepared, respectively, on par with standards. DSC and TGA measurements were observed to check glass transition temperature and thermal stability of the composites. The following chemicals were taken to check the chemical resistance acids: hydrochloric acid (HCl) (10%), acetic acid (CHzCOOH) (5%), nitric acid (HNOz) (40%); alkalis: sodium hydroxide (NaOH) (10%), sodium carbonate (Na2 COz) (20%), ammonium hydroxide (NH4 OH) (10%); solvents: benzene, toluene, carbon tetrachloride (CCl4 ), and distilled water (H2 O). Specimens were weighed before and after soaking in the predetermined chemicals. Then, allowed 24 h to soak and removed from the chemical and then weights are measured again. The difference in weights is used to calculate the percentage of weight loss or gain. Fractured surfaces were evaluated by SEM images to evaluate reasons for the decrease or increase in the performance based on the voids, interface, and pull-outs theories observed from the literatures. STF fibers were extracted from the natural retting process in which corresponding plant leaves were soaked in the running water for about 1 month. Then, beated with the help of stick for several numbers of times until all the cellulose and lignin substances come out of the fibers and followed by through washed out with distilled water. This is so-called untreated natural fiber. Untreated fibers are dipped in the NaOH solution to get rid of foreign substances and this is so-called treated fiber.

2.1 Fabrication of Composites At the outset, the mold surface was coated with mold releasing agent. Continuous fibers of STF (treated) and KF are employed to fabricate the composites. Epoxy and hardener were taken in predetermined quantities such as 10:1 ratio, respectively. Polymers and chemical agents mentioned above are mixed with the help of spatula for about 10 min. Catalyzed resin of 25% spread all over the mold with the help of brush and rollers to ensure wetting at the interstices of the fiber. Then, the layer of KF was stacked and then 25% of the resin is again poured on the KF and makes sure all the fiber ought to be wet and resin has to be spread with roller all over the mold. By doing so, voids could be reduced considerably. Then, another layer of the STF was stacked and followed by remaining resin is poured and spread all over the surface with roller. Then remaining resin poured all over the mold and spread the resin with

88 Table 1 Different treated polymer composite systems

M. Ashok Kumar et al. Composite system

KF (wt%)

STF (wt%)

A

100

0

B

50

50

C

75

25

D

25

75

E

0

100

the help of brush and roller once again. Finally, allow the mold covered with OHP sheet and make sure to pull the roller all over again mold gently. Then the mold is loaded with some weight and make sure without spoiling the uniform thickness of the mold. Then allow the mold to cure for 1 day. In order to remove the composites with ease, post-cured casting is placed in the oven for 45 min at 100 °C. The same procedure was used for the remaining composites (refer Table 1).

3 Results and Discussions In the present study, ASTM G543-87 is used to prepare samples to measure the chemical resistance of the treated samples. The consequence of some acids such as glacial acetic acid, nitric acid, hydrochloric acid, alkalis are ammonium hydroxide, aqueous sodium carbonate, aqueous sodium ammonium hydroxide, and solvents are carbon tetrachloride, benzene, distilled water, and toluene, respectively, were used (refer Tables 1 and 2). Table 3 shows the mechanical characterizations that are carried out for different samples and found three reasons. Firstly, tensile strength was found maximum for sample ‘D’ and it was attributed that high specific volume of STF imparts corresponding improvements in the surface area. Consequently, high surface area means high Table 2 Measurements of chemical resistance of the treated KF/STF composites

Name of the chemical

wt% gain(+)/loss(−) for composites

Hydrochloric acid

+0.875

Acetic acid

+0.230

Nitric acid

+1.523

Sodium hydroxide

+0.530

Sodium carbonate

+0.234

Benzene

+5.632

Ammonium hydroxide

+0.702

Toluene

+3.253

Carbon tetrachloride

−0.253

Distilled water

+1.036

An Experimental Studies on the Polymer … Table 3 Results of mechanical measurements as a function of specimens

89

Specimen name

Tensile strength (MPa)

Tensile modulus (GPa)

Impact strength (J/m2 )

A

25.10

0.30

15.20

B

38.44

0.38

13.74

C

39.48

0.39

13.08

D

42.65

0.45

16.43

E

20.63

0.28

9.35

bonding strength. Second reason was treated STF surfaces free from dust, cellulose, and lignin materials which ultimately improves the performance. Lastly, interface between the fiber and matrix was also found significant from the epoxy and KF/STF standpoint (refer Fig. 1). Tensile modulus is also increased for all the samples which is due to the increase in natural fiber content which has paid bonding strength than the other fiber. Impact strength was decreased for specimens B and C when compared to A on the other hand specimen-D has got highest impact strength which is due to the improved STF fiber volume. Tensile modulus and impact strength were found maximum for the sample D when compared with other specimens (refer Table 3). Figures 1 and 2 show the graph which indicates the optimization of tensile strength, tensile modulus, and impact strength as a function of sample. In Fig. 1, tensile strength and tensile modulus were optimized for except specimenD and remaining samples have registered less due to the poor interface and voids. Poor interface and voids result in crack initiation at early loadings therefore failure takes place. In Fig. 2, impact strength was improved for the same specimen-D due to the high specific volume of the STF makes more volume of the fiber, furthermore, treated STF sticks well with the matrix due to rough surface might be the reason.

Fig. 1 Tensile strength and tensile modulus measurements of the specimens

90

M. Ashok Kumar et al.

Fig. 2 Impact strength measurements of the polymer hybrid composite specimens

Figure 3 shows the SEM images of the cross-sectional areas of the hybrid composites are discussed as mentioned below. Figure 3a shows the fibers of both natural

Fig. 3 SEM images for specimens KF/STF a 50/50 b 75/25 c25/75 d 0/100, respectively

An Experimental Studies on the Polymer …

91

and synthetic fibers. This image is also depicted some voids that might be the reason for reduced performance. Poor interface might be the one reason for decreased performance, and another reason is pulled outs which are not seen in the image. Natural fiber strand was not coated resin properly due to that fiber pull-out has been observed in the same image [2]. Figure 3b shows that image was not turned to ductile nature due to decrease in natural fiber content. In Fig. 3c, the natural fiber weight ratio has been increased as a result of that brittle nature comes down significantly when compared with the 75/25(KF/STF) composites. Improved interface and void were reduced significantly better when compared with the others. The last image of Fig. 3d shows that the fiber pull-outs and voids observed [5, 6].

4 Conclusion This paper was dealt with the preparation of polymer hybrid composites and its characterization, viz. chemical resistance, tensile strength, tensile modulus, impact strength, and morphology studies. For specimen-D, tensile strength increased by 70% when compared with specimen-A. Tensile modulus, impact strength was increased up to 50, 8% for the specimen-D, respectively, when compared with specimen-A. In scanning electron microscope analysis, we noticed interface, treatment of the fiber surfaces, pull-outs, and voids were played a significant role in finalizing the performance of the composites. Specimen-D has higher improvement of strengths when compared with the other specimens. Therefore, these composites with specific ratios can be used for the fuel tanks, structural purpose, aero-plane fuselages, and wings.

References 1. Ashok Kumar, M., Hemachandra Reddy, K., Mohana Reddy, Y.V., Ramachandra Reddy, G., Venkata Naidu, S.: Improvement of tensile and flexural properties in epoxy/clay nanocomposites reinforced with weave glass fiber reel. Int. J. Polym. Mater. 59, 854–860 (2010) 2. Ashok Kumar, M., Ramachandra Reddy, G., Siva Bharathi, Y., Venkata Naidu, S., Naga Prasad Naidu, V.: Frictional coefficient, hardness, impact strength and chemical resistance of reinforced sisal-glass fiber epoxy hybrid composites. J. Compos. Mater. 46(26), 3195–3200 (2010) 3. Chakradhar, K.V.P., Venkata Subbaiah, K., Ashok Kumar, M., Ramachandra Reddy, G.: Blended epoxy/polyester polymer nanocomposites: effect of “nano” on mechanical properties. J. Polym. Plast. Technol. Eng. 51, 92–96 (2011) 4. Ashok Kumar, M., Ramachandra Reddy, G., Raghavendra Rao, H., Hemachandra Reddy, K., Nanjunda Reddy, B.H.: Assessment of glass/drumstick fruit fiber (moringa oleifera) reinforced epoxy hybrid composites. Int. J. Polym. Mater. 61, 759–764 (2012) 5. Ashok kumar, M., Ramachandra Reddy, G., Harinatha Reddy, G., Vishnu Mahesh, KR.: Light weight epoxy composites from short sansevieria cylindrica fibers on mechanical and thermal parameters. Fiber Polym. 13(6), 769–773 (2012)

92

M. Ashok Kumar et al.

6. Ashok Kumar, M., Ramachandra Reddy, G., Harinath Reddy, G., Subbarami Reddy, N.: Tensile and thermal properties of polymer coatings effect on sorghum vulgaris stalk fiber. J. Polym. Mater. MD publications. 29(1), 4:71–75 (2012) 7. Ramachandra Reddy, G., Ashok Kumar, M., Karthikeyan, N., Mahaboob Basha, S.: Tamarind fruit fiber (Tf) and glass fiber reinforced polyester composites. Mech. Adv. Mater. Struct. 22, 770–775 (2015) 8. Singh, J., Kumar, M., Kumar, S., Mohapatra, S.K.: Properties of glass fiber hybrid composites: a review. Polym. Plast. Technol. Eng. 56(5), 455–460 (2017) 9. Cao, Y., Cameron, J.: Impact properties of silica particle modified glass fiber reinforced epoxy composite. J. Reinf. Plast. Compos. 25(7), 761–766 (2006)

Design Analysis and Pressure Loss Optimization of Automobile Muffler Vikram Kumar, Naresh Prasad, M. K. Paswan, Pankaj Kumar and Sanjoy Biswas

Abstract Noise from exhaust system of automobile is the major concern and cause of noise pollution. That is why it is a major concern and important area of research and development. Muffler is an important part of the engine which is used to minimize sound from exhaust, backpressure that affects fuel efficiency, noise as well as emission. The proposed design of muffler is chosen through so much of iteration on different designs. This paper mainly concerns to establish the relation between the pressure loss and the exhaust gas inlet velocity. So, the backpressure must be kept minimum. It also deals with the design methodology of muffler which will be an improvement to the existing design of muffler, improvement in terms of design, pressure loss, and noise level. It also focuses on modern cad tools and simulation tools which provide maximum advantage for optimizing the design in short time and for better result.

1 Introduction New government norms and standard forced automobile companies to design engine exhaust to decrease exhaust noise. Exhaust noise is one of the sources and other sources are like intake noise, noise from vibrating engine body, and mechanical V. Kumar (B) · N. Prasad · M. K. Paswan Department of Mechanical Engineering, NIT Jamshedpur, Jamshedpur 831014, India e-mail: [email protected] N. Prasad e-mail: [email protected] M. K. Paswan e-mail: [email protected] P. Kumar · S. Biswas Tata Motors Ltd., Jamshedpur 831010, India e-mail: [email protected] S. Biswas e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_10

93

94

V. Kumar et al.

noise. And noise from automobile exhaust is the major source of noise pollution. Puneetha et al. described the important function of a muffler. In automotive industry, the reduction in noise is an emerging concern [1]. There are various designs in the market available today; difference in them is only the path of the exhaust gas. Designing of muffler must consider all the factors that are correlated to each other and must be optimum [2]. Increasing back pressure will increase the torque [3]. The sole purpose of an automotive muffler is to reduce engine noise emission. A car running without a muffler will have an appreciation for the significant difference in noise level a muffler can make. If vehicles did not have a muffler, there would be unbearable amount of exhaust noise in our environment [4]. Hence, it is the need of time that a well robust design exhaust muffler should be prompted to meet the environmental norms as standardized by the government. Generally, exhaust system consists of exhaust manifold, catalytic converter, resonator, muffler, and exhaust pipes. But the paper’s prime focus is on muffler. In simple words, automobile exhaust muffler’s primary role is to reduce noise keeping other exhaust parameters in the optimum limit. While designing muffler, some functional requirements must be considered. Some of them are backpressure, sound, insertion loss, transmission loss, shape, size weight, manufacturability, and cost. So, in this study, muffler inlet velocity and pressure drop will be studied across the tubes and backpressure is measured. Pattern of pressure variation is obtained through CFD analysis. Fang et al. CFD technique helps designer to include all the parameters while designing parts of model [5]. Fang et al. CFD method is used to calculate pressure loss in the muffler. With the help of CFD technique, at a certain engine exhaust inlet velocity, the fluid flow inside muffler was simulated and the distribution of total pressure was analyzed [6]. For analysis of fluid flow inside the muffler, proposed design of muffler was drawn via Solid Works platform. Then, model was analyzed via Ansys Workbench. Actually, muffler’s performance is analyzed by pressure variation, length of muffler chambers, and flow pattern of exhaust gas.

1.1 Backpressure The difference between mean exhaust pressure and ambient pressure is termed as backpressure. Net power available at the crankshaft depends on the level of backpressure. The amount of power loss depends on many factors, but a good rule of thumb is that one inch (25.4 mm) of mercury backpressure causes about 1.0% loss of maximum engine power [7]. Engine brake mean effective pressure decreases by the backpressure produced by muffler which results in the decrease in volumetric efficiency of automobile engine. Engine manufacturer specifies some max allowable backpressure for all engines. Allowable maximum back pressure limits was determined by the Swiss VERT program in order to allow diesel particulate filter to install within the system (Mayer 2004) [8]. VERT Maximum Recommended Exhaust Back Pressure limit of automobile exhaust has been shown in Table 1.

Design Analysis and Pressure Loss Optimization … Table 1 VERT maximum recommended exhaust back pressure

95

Engine size

Back pressure limit

Less than 50 kW

40 kPa

50–500 kW

20 kPa

500 kW and above

10 kPa

But engine manufacturing industries are more conservative regarding backpressure limit. In order to set the backpressure limit, many factors are considered. It depends upon design, exhaust emission and temperature, type of fuel capacity of engine, and performance of turbocharger.

1.2 Effects of Increased Back Pressure With an increase in backpressure, piston has to compress the exhaust gas with more pressure in which additional work is done that subtracts the work generated from the combustion of fuel. This, in turn, affects the fuel consumption as for the same amount of work more fuel is required. This increases the exhaust temperature which results in overheating which also increases the emission of NOx . Also, increase in backpressure affects the performance of turbochargers which affects the performance of the engine.

2 Methodology In order to estimate the variation in pressure in muffler due to change in the velocity of exhaust gas, following method is adopted. Figure 1 elaborates the research work starting from the study of exhaust systems of different commercial vehicles and ends with an analysis of design.

3 Geometry Tata mini-truck muffler is used as a reference muffler in order to study the relation between exhaust inlet velocity and backpressure. Design of a muffler is always having been done in an iterative way by trial and error method. But companies nowadays shortened the PDC time. The design of muffler is as shown in Fig. 2. It is absorptivetype muffler that uses absorption process to reduce sound energy. In this type of muffler, sound energy is converted into heat energy by absorptive material. Rockwool is used as an absorbent as it reduces the intensity of sound up to maximum limit of

96

V. Kumar et al.

Fig. 1 Design methodology

Fig. 2 Model of exhaust muffler

50 dB depending on geometry. Tata mini-truck engine comes with 4-cylinder 1396 cc petrol engine with 85 Hp @ 5500–6000 rpm. • • • • •

Tata mini-truck muffler data Total length (l) = 889 mm, Width (w) = 280 mm, Height (h) = 210 mm, Outer diameter of inlet pipe = 91 mm,

Design Analysis and Pressure Loss Optimization …

97

Fig. 3 Perforated inlet pipe with hole diameter 5 mm

• • • • •

Inner diameter of inlet pipe = 89 mm, Outer diameter of outer pipe = 91 mm, Inner diameter of outlet pipe = 89 mm, No. of perforated holes = 1390 Perforated holes’ diameter (d) = 5 mm.

Figure 2 shows the muffler that consists of four pipes: inlet pipe, outlet pipe, and two pipes connecting chamber 1 and 2. Each pipe contains perforated holes. It contains a total of five chambers. It is known that more is the number of chambers (with increase in corresponding total length) better is the insertion loss. Figure 3 shows the snap of perforated inlet pipe with hole diameter 5 mm. Model which has been taken in use has been designed in Solid Works.

3.1 Meshing Domain of interest was isolated and meshed for CFD simulation which was done automatically. Meshing used is tetrahedron. Figures 4 and 5 depict the meshing method. The model is complex and due to its complexity tetrahedron, mesh is selected to do meshing in Ansys Workbench. Fine meshing was done in order to get the accuracy and high quality. The number of elements generated in mesh is 13,069,037, and number of nodes is 2,387,768.

3.2 Assumption and Boundary Conditions 1. Fluid flow is assumed steady and turbulent (K–ε Model). 2. Air is assumed to be the fluid for computations.

98

V. Kumar et al.

Fig. 4 Meshed body

Fig. 5 Sectional view of meshed body

3. Mass flow rate is assumed as 320 kg/hr 4. Pressure at outlet 101,325 pa (opened to atmosphere) (Tables 2 and 3). Analysis for these boundary conditions has been done in CFD tools of ANSYS. Each case is analyzed under same boundary conditions. Table 2 Material property details Material

Young’s modulus

Poisson’s ratio

Density (constant at temperature)

Steel

210 MPa

0.3

1.225 kg/m3

Design Analysis and Pressure Loss Optimization … Table 3 Initial values for inlet velocity as the inlet boundary conditions

99

Inlet temperature

873 K

Thermal conductivity λ (W m−1 K−1 )

6.222 × 10−2

Specific heat capacity cp (J kg−1 K−1 )

1089

dynamic viscosity μ (kg m−1 s−1 )

3.91 × 10−5

Inlet turbulence density

10%

Viscosity (kg/ms)

2.7e−05

Enthalpy (J/kg)

749,575.3

Ratio of specific heats

1.4

Outlet temperature

300 K

Outlet turbulence density

10%

Gas density

Air defined as incompressible ideal gas

Energy

On

4 Results and Discussions Proposed muffler is designed in Solid Works and then simulated in Ansys R18.2 (Licensed Version). After the meshing part is over, the fluid flow analysis was done by fluent by setting different parameters for running solutions and to get the result. Since the fluid flow in the exhaust muffler is turbulent, K–ε model is used. Also, inside the muffler, temperature is high so density change is assumed. Therefore, energy equation is used. Simulation is performed at three different muffler inlet velocities 39, 44, and 49 m/s.

4.1 Inlet Velocity: Case 1: 39 M/S See Figs. 6, 7, 8, 9, 10, and 11.

4.2 Inlet Velocity: Case 2: 44 M/S See Figs. 12, 13, 14, 15, 16, and 17.

100

V. Kumar et al.

Fig. 6 Pressure contours

Fig. 7 Residuals

Fig. 8 Velocity streamlines

4.3 Inlet Velocity: Case 3: 49 M/S From the image of pressure contour Figs. 6, 12, and 18 obtained as a result of simulation, it can be concluded that pressure is maximum at the exhaust inlet and gradually decreases in the subsequent chambers. Vortex is the main reason for generating pressure loss and the secondary noise of the muffler [9]. Figures 8, 14, and 20 show the velocity streamlines that indicate the path of exhaust gas in relation to velocity.

Design Analysis and Pressure Loss Optimization …

Fig. 9 Velocity contours

Fig. 10 Pressure variation in inlet pipe

Fig. 11 Pressure variation in outlet pipe

101

102 Fig. 12 Pressure contours

Fig. 13 Residuals

Fig. 14 Velocity streamlines

Fig. 15 Velocity contours

V. Kumar et al.

Design Analysis and Pressure Loss Optimization …

Fig. 16 Pressure variation in inlet pipe

Fig. 17 Pressure variation in outlet pipe Fig. 18 Pressure contours

103

104

V. Kumar et al.

From Figs. 9, 15, and 21 of velocity contours, it can be observed that the velocity of exhaust gas is greater at the inlet and outlet pipe compared to the velocities in the subsequent chambers. Figures 10, 16, and 22 show the variation of pressure loss along the length of inlet tube, and Figs. 11, 17, and 23 show the pressure variation along length of outlet tube for corresponding velocities 39, 44, 49 m/s (Fig. 19). From the above data, in Table 4, it can be inferred that with increase in exhaust inlet velocity, the backpressure also increases linearly. One can also infer that pressure inside muffler decreases as it reaches exit. A CFD analysis result was verified with physical test results. There was little difference between CFD and experimental Fig. 19 Residuals

Fig. 20 Velocity streamlines

Fig. 21 Velocity contours

Design Analysis and Pressure Loss Optimization …

105

Fig. 22 Pressure variation in inlet pipe

Fig. 23 Pressure variation in outlet pipe Table 4 Comparison table for exhaust back pressure Exhaust inlet velocity

Result of existing muffler [9]

Case 1: 39 m/s

Case 2: 44 m/s

Case 3: 49 m/s

At velocity 50 m/s

Muffler abs inlet pressure (Pa)

102,013

102,183

102,365



Muffler abs outlet pressure (Pa)

101,325

101,325

101,325

101,325

Muffler abs pressure difference/backpressure (Pa)

688.047

858.461

1040.37

5139.4731

106

V. Kumar et al.

values may be due to various parameters that were not taken into consideration during simulation. For example, friction was not considered. We also know that type of meshing also affects the simulation results. The improvement of pressure loss is 75% in respect of the author Xu et al. [9] which can be observed in Table 4.

5 Conclusion and Future Work In the present work, study of pressure variation along the length of exhaust muffler at different velocities has been analyzed to understand the correlation between pressure drop and flow velocities. For different velocities, we observe different pressure drop. It can be observed that when the velocity of exhaust increases backpressure also increases. Also, it can be observed that percentage increment in backpressure is linearly dependent on the increase of velocity. In this computational study, drawing a relation between backpressure and exhaust gas velocity can help in further detailed analysis. This result can be used as a reference for designing new muffler. Proposed design must be analyzed for the other key parameters like acoustic characteristics, vibration durability analysis and correlation between backpressure and noise. Acknowledgements Thanks to Mechanical Engineering Department, NIT Jamshedpur for valuable contribution in developing the ICIMES article template. Also, thanks to ERC, Tata Motors Ltd. for their guidance in design calculation and other necessary inputs toward completion of this project. Authors also feel privileged to express gratitude to them who motivated us and also mentor all through the project work. This project cannot have been completed without the help of mentors.

References 1. Puneetha, C.G., Manjunath, H., Shashidhar, M.R.: Backpressure study in exhaust muffler of single cylinder diesel engine using CFD analysis. Altair Technology Conference (2015) 2. Kamble, P.R., Ingle, S.S.: Copper plate catalytic converter: an emission control. SAE Tech. Pap. No. 2008-28-0104 (2008) 3. Ashok, P., Shivdayal, P., Umashanker, G., Mohan, S.: Commercial vehicles muffler volume optimization using CFD simulation. SAE Tech. Pap. No. 2014-01-2440 (2014) 4. Potente, D.: General design principles for an automotive muffler. Acoustics 2005, Busselton, Australia (2005) 5. Fang, G.U., Bo-tan, L.I.U., Hong-liang, L.I., Shu-jie, P.A.N.: Structural analyses for the vehicle exhaust system based on CFD simulation. Trans. Csice 25, 358–362 (2007). https://doi.org/10. 3321/j.issn:1000-0909.2007.04.012 6. Fang, J., Zhou, Y., Jiao, P., Ling, Z.: Study on pressure loss for a muffler based on CFD and experiment. International Conference on Measuring Technology and Mechatronics Automation (2009) 7. Mondt, J.R.: Cleaner cars: the history and technology of emission control since the 1960s, p. 262. SAE International, Warrendale, PA (2000)

Design Analysis and Pressure Loss Optimization …

107

8. Mayer, A.: Number-based emission limits. VERT standards and procedures for retrofit to reduce diesel engine emissions. Transport and Clean Air Moscow, Russia. December 11–12 (2013) 9. Xu, J., Zhou, S., Li, K.: Analysis of flow field and pressure loss for fork truck muffler based on the finite volume method. 33 (3), 85–90 (2015). http://dx.doi.org/10.18280/ijht.330312

Different Modules for Car Parking System Demonstrated Using Hough Transform for Smart City Development Janak D. Trivedi, M. Sarada Devi and Dhara H. Dave

Abstract Searching for the parking space is a time-consuming task while visiting for shopping or unknown cities. Real-time parking management gets benefited for the development of the smart city and also reduces time for finding the parking place. In this chapter, different two types of modules for car parking are simulated using a combination of Hough transform, edge detection, and color enhancement method. Different car parking modules are circular shape-based: (i) parallel parking with a different radius and (ii) angle parking with the same radius. Real-time video is captured using an android smartphone with an IP camera. The proposed research work can enhance the solution for real-time parking solution in big malls and theater. Limitation and future scope of this will give motivation toward research work on an intelligent transportation system (ITS) development in India. For small-scale version here, we have used toy cars and bus with a different color for parking modules.

1 Introduction Due to increase in population of the world numbers of vehicles on the roads are increasing day by day. One of the major challenges in the cities is to manage the parking of vehicles. Reports suggest that as much as 30% of traffic flow delays and restrictions are due to drivers in search of parking. Smart parking is a vehicle parking

J. D. Trivedi (B) · D. H. Dave Electronics and Communication Department, GEC Bhavnagar, Gujarat Technological University, Gujarat, India e-mail: [email protected] D. H. Dave e-mail: [email protected] M. Sarada Devi Electronics and Communication Department, AIT Ahmedabad, Gujarat Technological University, Gujarat, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_11

109

110

J.D. Trivedi et al.

system that helps drivers find a vacant spot by using real-time data which can be used to direct drivers to available parking spaces. Lack of sufficient space available for parking gives a headache to vehicle owners, where to park their own cars in minimum time? In the past few years, the Indian central government also has been trying to develop cities to become smart cities as an urban development program. Intelligent transportation system (ITS) plays an important role in smart cities development. In real-time traffic management, ITS covers accident monitoring and helping them immediately, parking space availability, minimum traffic congestions, automatic traffic light signals management, automatic number plate recognition, and many more things. In this chapter, automatic parking management is discussed with different parking modules like circles with the same radius and circle with a different radius. In these, all parking modules’ feature selection plays an important role. Section 2 discusses the technology gap between theoretical and practical implementation of the smart parking. Next, Sect. 3 explains about various different methods or techniques available earlier for real-time parking management. Section 4 is focused on Hough transform for the circle and straight-line detection. Section 5 gives a basic idea of IP webcam smartphone applications, and is related to the proposed methods with a flowchart. Simulation results for all the different modules are explained in Sect. 6, and the last section gives future scope with the limitation of this work. We would like to draw your attention to the fact that it is not possible to modify a paper in any way, once it has been published. This applies to both the printed book and the online version of the publication. Every detail, including the order of the names of the authors, should be checked before the paper is sent to the volume editors.

2 Technology Gap Currently, the manual parking system available in the Indian city, so there is no direct information available, for all the users. The man who manages parking only knows where to park a vehicle for maximum time duration and effective parking within the parking area. If I had to fix my parking slot in advance or I want to know the actual parking system with EMPTY and BLOCK parking spot identification, I can save both my time and fuel. This way I can save energy and also reduce noise and traffic consumption and definitely my money. I can decide my parking by seeing all the information available for a parking spot in my smartphone before I will go for the parking place, and with the help of parking fees, I have assured my parking spot also. The use of available land space to the parking with the help of a smartphone and real-time traffic management possibility here is presented in this chapter with earlier method descriptions. The novel proposed method is discussed using flowchart and its credibility verified using simulation results.

Different Modules for Car Parking System Demonstrated …

111

3 Background In past, [1] gives a mathematical model for parking space detection near the zoo area in terms of supply and demands of the available parking spaces. For every parking space, the demand for parking space for the specific time period is measured, and according to that, effective collection of parking fees is done. Parking fees is in accordance with the final destinations and parked vehicle distance. Maximum likelihood method [2] was explained for various choices of parking places with different parameters like parking charges, the walking time, the required time for parking, etc. Bus parking facility in [3] was discussed. Computer-based assignment and mathematical modeling system is developed for bus parking to prevent blockage and reduce time for arriving and departure of buses. In this method, probability of assigning a correct label for each bus with dynamic control buses is an important thing to be validated. Monday to Friday was a different situation compared to public holidays and Sundays. Minimum parking effect in [4] concludes deregulating the quantity and increasing the quality of parking will improve transportation and land use. Parking fees was based on the number of registered vehicle per year, effective land prize and different situation during parking. Parking fees collection should be with a different strategy like higher parking fees with near to destination location and lower parking fees for far to destination location. Parking fees handled by private firms in Taiwan was explained in [5], with the proper understanding between private and government firm with the different stages of selections. In this proposed agenda, the government is a leader and a private firm is a support system to the government. This is the best balance system for any government system like in India where our Honorable Prime Minister is trying to develop smart cities for a better future. Real-time parking information was discussed in [6] based on a genetic algorithm. Finding a proper parking place for a driver to reduce search in time, less pollution, and noise. Bayesian network-based parking was discussed in [7] with three main parameters—parking period, parking location, and parking duration with information available from China. Global climate change and smart city development force to think about intelligent transpiration system with automatic parking management. On-street parking fees compare to off-street parking, and traffic congestion issues, illegal parking, how it can be solved using proper parking management! explained in the same. This paper gives parking management for China, the country which has more or less same population as India. In [8] many parking facilities in Beijing are investigated and chosen six indexes, including nature of the parking lot, bus traffic, subway traffic, location, number of parking spaces and distribution of the parking lot. Through those, area of surrounding parking facility was classified based on land use functionality. On-street parking model [9] was discussed for truck delivery in the urban area and also included Toronto City for parking location as a case study. Revenue generation and its management [10] were explained in the parking field. Parking request is managed using multiple-garage intelligent model. Parking, Safety and design characteristics-based level of service for street parking is discussed in

112

J.D. Trivedi et al.

[11] for Kolkata. Blob analysis-based perpendicular and forward parking detection using image processing is explained in [12]. We proposed real-time parking management for small modules using Hough transforms which aim to locate and find real-time parking slot detection and identification. Research work based on finding a location of parking using circle Hough transform and proper thresholding operation are elaborated in this chapter.

4 Hough Transform Hough transform (HT) is a very good transformation technique for detecting an object using different shapes. Image segmentation is the first step in digital image processing for segmentation of the particular object. The basic idea is very simple—how to detect the straight line! Figure 1a shows, straight line AB with two different points (xi, yi) and (xj, yj) in image plane, x-y. Straight line AB in the image plane used basic line equation as shown in Eq. (1), P = x cos θ + y sin θ

(1)

where P is the distance from the coordinate system center to the straight line AB (as shown in the above figure) and θ is the angle from one of the axes with the line connecting to the center. Every point in the straight line AB follows Eq. (1). Now with the help of Hough transform, take a point from the image which is nothing but point that exists in the edge. For every point, we have an infinite number of lines with particular P and particular θ which is shown on the straight line AB. Let us draw another straight line passing through the same point (xi, yi) CD which has different P1 and different θ 1, the similar way an infinite number of a straight line passing through the point (xi, yi) with different (P, θ ). The second point (xj, yj) has an infinite number of straight lines with different (P, θ ) which are passing through

Fig. 1 a A simple illustration of how Hough transform works, for a straight line and point. b xy coordinate to P-θ mapping for a line [13]

Different Modules for Car Parking System Demonstrated …

113

the same point. From those an infinite number of straight lines, there is one common (P, θ ), so there is only one line which has two points on the same line with two different votes and that is the procedure for finding P and θ for straight line AB. If we consider third point x on the same straight line AB, we have also found a different straight line passing through the same point with different (P, θ ). But for common (P, θ ) it has a maximum three votes, for that particular point x indicated edge detected line from that particular image. This is basic Hough transform: For every point, that edge detector found votes for the possibilities in these case lines. If we have multiple points on the same line, then we have multiple votes. So, for working with a real-time condition, we are going to go x-y coordinate system which is the image coordinate system to a new coordinate system, which has theta θ and row P, shown in Fig. 1b. Figure 2a, discretized in Fig. 2b in P, θ index. Now, with the help of Eq. (1), each and every point is discretized in P, θ form. Like put the value of the point (xi, yi) with different θ degree value, from θ = 1, 2, 3, etc., and we can able to find rho value for respective θ . Sometimes, that point value measurement is called ‘accumulator.’ So keep voting for every single point. θ varies from θ min to θ max, and rho value changes in sinusoidal fashion as shown in Fig. 2. Sinusoidal nature comes from the sin and cosine nature of Eq. (1). For the second point (xj, yj), get another sinusoidal waveform with different votes and there would be a place which corresponds to rho of the straight line AB that is starting to accumulate votes for every point. The next step is from the accumulator to find the maximum value of that line. There would be a chance of multiple accumulators for multiple lines.

Fig. 2 Variation of theta value affecting change in rho value [13]

114

J.D. Trivedi et al.

4.1 Find Circle Using Hough Transform A circle can be written down in Eq. (2), (x − x0)2 + (y − y0)2 = r 2

(2)

where r indicates the radius of the circle, with respect to (x0, y0) coordinates and (x, y) point. Every point in a circle holds this Eq. (2), with center (x0, y0) coordinates. As we have to find rho and theta for straight lines, in circles, we have to find centers and radius for a circle. In this case, we have to find three parameters x0, y0, and r. Therefore, our accumulator will not be on a plane, but it will be a three-dimensional accumulator. The procedure is same, as in the earlier case, find the point with edge detector, every point is not a vote for a circle, but it gives us coordinates x0, y0 and then runs through accumulator with x-y coordinates of respective x0, y0 centers. Every point votes for multiple centers and multiple radii. We can limit the center and the radius; if we know the dimensions of that object and limit the votes of the radius, then we can know how the large object is in an image!

5 IP Webcam Designed modules can be processed with the help of IP webcam installed in the smartphone. IP webcam is useful for connecting android device to an operating device like desktop and laptop. Hypertext Transfer Protocol (HTTP) gives IP address; using that IP address, first connect our system with an android device by creating hotspot features in a smartphone. Then, click on start server button in whichever smartphone we have used. Here, Asus ZenFone Max Pro M1 device is used, which has inbuilt 16 megapixel camera resolution for a back and front camera, 6 GB RAM, and 64 GB available storage capacity. We can also view on the Web browser with the help of HTTP address, and we can change video resolution, orientation, and many other parameters related to audio, image, and video operations. So, in nutshell, IP webcam application provides us a real-time camera, which is interfaced to the current operating system. There is much more application like this available, which will be useful as IP camera application.

6 Proposed Method Circular Hough transform (CHT) is applied to one of the parking modules with the three-step requirement: first, computation of accumulator array; second, center estimation; and third, radius estimation. Circle Hough transform is implemented by

Different Modules for Car Parking System Demonstrated …

115

Fig. 3 Flowchart of the proposed method for parking module

two different techniques: (1) two-stage method and (2) phase-coding method. The two-stage method is based on radial histogram computations, and the phase-coding method is based on computations of complex value in accumulator array, and radius information encoded in the phase of the array. For parking module with different radii as shown in Fig. 4 here, we have used the two-stage method. The flowchart of the proposed method is shown in Fig. 3.

6.1 Step-by-Step Procedure First step: Read real-time video using Hypertext Transfer Protocol and link that to the MATLAB for further image processing and real-time parking detection. The second step applied general Hough transform and circle Hough transform (CHT) which are

116

J.D. Trivedi et al.

explained in the earlier section with equations. The third step is to check different light conditions and select coordinate and its locations. The fourth step added sensitivity level and accordingly controlled light conditions, dust, and color combinations of different cars. The fifth step checked reference saved data and compared continuously with real-time available information of parking status for both different radius and same radius modules. The sixth step is to find real-time circle location for parking status. The seventh step is to find parking status availability using circle detection with same and different radius. The final step checked available parking slots with the help of Eqs. (3) and (4), N 

[Iref (r, c) > Icur (r, c)] − T

(3)

{[Iref (r, c) < Icur (r, c)] + T } > 0

(4)

i=1 N  i=1

where N indicates a number of array elements, I ref is reference image, I cur is a current image, r indicates row number of the image, c indicates column number of image, and T indicates threshold values. Generally, HT is used for finding straight lines, circles, ellipse, etc., which have controllable number of parameters; otherwise, accumulator takes lot of memory and lot of computational time to find the maxima. Figure 3 represents a flowchart of the parking module and respective parking module, and its graphical user interface (GUI) is shown in Fig. 4. The circle shape detection for a different radius will not affect final real-time parking simulation results.

Fig. 4 Parking module and its graphical user interface (GUI) with a different radius and a parallel parking

Different Modules for Car Parking System Demonstrated …

117

Fig. 5 All empty and all block parking spot

Fig. 6 Individual parking slot, checking from 1 to 12

7 Simulation Results The proposed method is applied for two different cases: (1) circle with different radius parallel parking and (2) circle with the same radius—angle parking, using live video streaming from IP-based smartphone. Performance analysis parameters like true positive (which is higher) and false negative (which is nil) have checked with different colors, size, numbers, and location of available small toy cars. Their respective modules with simulation results using a graphical user interface (GUI) are shown in Figs. 4, 5, 6, 7, 8, and 9.

7.1 Circle with Different Radius—Parallel Parking Parking module is running with several assumptions and with particular lighting conditions. Light changes reflect false identification of circle, and parking module

118

J.D. Trivedi et al.

Fig. 7 Real-time parking slot checking for different available parking locations of different radius module

will not give accurate results. For this parking module, we have used an android smartphone with 1920 × 1820 resolutions with full light source available. Here, parking slot ‘Empty’ represents parking for a vehicle available and parking slot ‘Block’ means parking not available for that particular slot. Figure 5 shows the result for all parking slot available as EMPTY and BLOCK for not available car parking slot with green and red colors, respectively. Figure 6 represents 12 different parking slots with one car parking and remaining eleven empty slots. Figure 7 shows more than one filled parking slot with the help of small toy cars with different colors which is giving 100% accurate result, with fixed camera position and without varionations in light intensity.

Different Modules for Car Parking System Demonstrated …

119

Fig. 8 Parking module and its simulation results with same radius—angle parking

Fig. 9 Individual and random parking slot, checking from 1 to 32 locations

7.2 Circle with the Same Radius, Angle Parking In the earlier case, different radius-based circle shape detection is helpful for implementation of real-time parking. Now consider another case for circle, with the same radius, with an increasing number of parking slots up to 32 for small-scale implementation, simulation results are shown in Fig. 8 . In this, total of 32 parking spots have been detected using CHT. First check all the parking spot EMPTY, with red

120

J.D. Trivedi et al.

color circle mark on it as simulation results. Then, in Fig. 9, the vehicle with different color toys is placed, and we can see changes in simulation results and also notice that when toy cars are not present in a parking place, then parking availability is not affected. Performance measurements such as true positive value is high and false negative value is almost nil for the controlled light conditions, which are best in terms of results.

8 Conclusion and Future Work Simulation results are shown here for two different cases: a circle with different radius shape parallel parking and circle with same radius parallel parking real-time detection using simple Hough transform and circular Hough techniques with the help of IP webcam smartphone-based application. For easy in visibility Empty parking slot is indicated using ‘Green’ color and parked slot is indicated using ‘Red’ color. Limitation of this work is change in light effect on simulation results for all the cases discussed above. So here fixed camera position and minimum changing light are the assumptions for real-time implementation of this work. In future work, we will address this limitations and develop real-time parking system with minimum processing time and maximum efficiency.

References 1. Dirickx, Y.M., Jennergren, L.P.: An analysis of the parking situation in the downtown area of West Berlin. Transp. Res. 11(3), 1–11 (1975) 2. der Goot, D.: A model to describe the choice of parking places. Transp. Res. Part A Gen. 16(2), 109–115 (1982) 3. Marlin, P.G., Nauss, R.M., Smith, L.D., Rhoades, M.: Computer support for operator assignment and dispatching in an urban transit system. Transp. Res. Part A Gen. 22(1), 13–26 (1988) 4. Shoup, D.C.: The trouble with minimum parking requirements. Transp. Res. Part A Policy Pract. 33, 550 (1999) 5. Tsai, J.F., Chu, C.P.: Economic analysis of collecting parking fees by a private firm. Transp. Res. Part A Policy Pract. 40(8), 690–697 (2006) 6. Caicedo, F.: Real-time parking information management to reduce search time, vehicle displacement and emissions. Transp. Res. Part D Transp. Environ. 15(4), 228–234 (2010) 7. Zong, F., Wang, M.: Understanding parking decisions with a Bayesian network. Transp. Plan. Technol. 38(6), 585–600 (2015) 8. Yin, J., He, Y., Sun, X.: Area classification of surrounding parking facility based on land use functionality. Open J. Appl. Sci. 06(07), 380–385 (2016) 9. Amer, A., Chow, J.Y.J.: A downtown on-street parking model with urban truck delivery behavior. Transp. Res. Part A Policy Pract. 102, 51–67 (2017) 10. Roper, M.A., Triantis, K., Taylor, G.D., Teodorovi´c, D.: Revenue management in the parking industry: a multiple garage intelligent reservation model. Transp. Plan. Technol. 41(3), 286–300 (2018)

Different Modules for Car Parking System Demonstrated …

121

11. Das, D., Ahmed, M.A.: Level of service for on-street parking. KSCE J. Civ. Eng. 22(1), 330–340 (2018) 12. Campbell, S., Holguín-Veras, J., Ramirez-Rios, D.G., González-Calderón, C., Kalahasthi, L., Wojtowicz, J.: Freight and service parking needs and the role of demand management. Eur. Transp. Res. Rev. 10(2), (2018) 13. Gonzalez, R.C., Woods, R.E.: Digital image processing. Chapter-10, (2002) 14. Davarci, A., Schick, N., Marchthaler, R.: Detection of perpendicular parking spaces with a mono camera. ATZ Worldw. 120(12), 66–69 (2018)

Effect of Arrangement and Number of Water Mist Spray Nozzles on Air Humidity Pravinth Balthazar, Mohd Azmi Ismail, Andyqa Abdul Wahab, Mohammad Nazmi Nasir, Muhammad Iftishah Ramdan and Hussin Bin Mamat Abstract Water spray technique is used in many applications like cooling, humidifying, and firefighting applications. The performances of 1, 4, and 9 number of nozzles under horizontal parallel, counter, and vertical flow arrangement have been experimentally analyzed for air volume flow rate from 0.34 to 2.15 m3 /s, room temperature from 28.5 to 30.2 °C, and relative humidity between 59 and 78%. The data show clear trend between relative humidity and number of nozzles. 9.5, 10.5, and 20% higher humidification by percentage achieved for the highest number of nozzles than lower number of nozzles in vertical, parallel, and counter flow arrangement, respectively. However, flow evaporation is greatly affected than mist evaporation performance under parallel flow placement at single-nozzle arrangement. Vertical arrangement with 9 nozzles showed 20.3% higher relative humidity than 1 nozzle under counter flow arrangement. P. Balthazar · M. A. Ismail (B) · A. A. Wahab · M. N. Nasir · M. I. Ramdan School of Mechanical Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Pulau Pinang, Malaysia e-mail: [email protected] P. Balthazar e-mail: [email protected] A. A. Wahab e-mail: [email protected] M. N. Nasir e-mail: [email protected] M. I. Ramdan e-mail: [email protected] P. Balthazar Department of Mechanical Engineering, South Eastern University of Sri Lanka, 32360 Oluvil, Eastern Province, Sri Lanka H. B. Mamat School of Aerospace Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal, Pulau Pinang, Malaysia e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_12

123

124

P. Balthazar et al.

1 Introduction Energy is wealth, a major element in any countries’ development, especially developing countries like Malaysia, India, and Sri Lanka. It is a root cause to economy development and improvement of the quality of human life. Nowadays, HVAC systems are consuming approximately 20% of the total energy consumption [1]. Evaporative cooling is an alternative to mechanical vapor compression for thermal comfort applications. These systems usually require only a quarter of the electric power that mechanical vapor compression uses for conventional air-conditioning [2]. Therefore, such system will help in reducing electricity requirements and also contributes to reduce carbon credit emissions. Evaporative cooling technique uses water spray system that could produce cloud of very fine mist using atomization of nozzles. Fine spray mist enhances mixing between air and water mist; then, it increases the contact of surface area between the air stream and the water mist. As a consequence, it increases humidity level significantly in cool and low-humid climate zone. Water and mist spray is an efficient and environmentally friendly approach to enhance thermal comfort in indoors [3–6] which gives building designers to be flexible and be innovative in system design and renovation projects. Water mist evaporation in spray systems is influenced by several variables including free stream velocity, ambient temperature and relative humidity, water temperature, and mist size distribution. However, detailed knowledge of the every single parameter which is related to each other is still need to be thoroughly understood. Water spraying technique in direct evaporative cooling (DEC) can be classified according to spraying direction. Cooling towers mostly spray water in cross-flow to air stream and into gravitational direction, in order to enhance the performance of the evaporative cooling tower units. Muangnoi et al. [7] conducted experiments to determine the influences of moist air velocity, temperature, and humidity using exergy analysis in 2008. They concluded that exergy change of water is higher than air and exergy change in air dominated the evaporative heat transfer. Kachhwaha et al. [8, 9] and Suresh Kumar et al. [10, 11] employed wind tunnel to experiment misting system under various parameters including nozzle diameter, water temperature, velocities at nozzle angle, and water pressure under controlled hot and humid conditions. They used air which flows in horizontal direction inside square cross section channel, and spray is arranged in parallel to flow [8, 10] and counter to flow [9, 10] configurations. Parallel and counter flow mist spray arrangements show slight difference in outlet temperature and humidity measurement. Suresh Kumar et al. [11] showed numerical predictions for cooling and moisture addition in good agreement with the experiments within ±15% for parallel flow and within ±30% for counter flow configuration. However, if larger droplets produced only a fraction of evaporates, the rest remains in liquid phase, thus decreasing the cooling performance [12]. The drift eliminators have been deployed to remove droplets. It reveals the fact that droplet/mist diameter size is important and higher water pressure enhances droplet breakup [13].

Effect of Arrangement and Number of Water …

125

Subsequently, the researches on spray humidification become more based on CFD analysis that leads to fewer options to obtain experimental data validation. Montazeri et al. [14, 15] used Suresh Kumar et al.’s [10] experimental data to validate their CFD model. Later in 2016, Alkhedhair et al. [16] developed a numerical simulation with precooling air with water spray. They conducted experiments to validate their Lagrangian–Eulerian 3D model. Recently, Balthazar et al. [17] investigated the influence of indoor portable ultrasonic humidification system which is distributed by piccolo tube (PT). They varied diameter of hole sizes of PT (5, 7, 10, 12 mm) and air velocity (1, 3, 5 m/s) and observed the rate of water vapor added to the HVAC system and variation of relative humidity of the air. The authors concluded that relative humidity increased with the hole size and decreased with air velocities. Even though relative humidity is decreased with the air velocity, the rate of water vapor added to the system is increased with air velocities. The present study focuses on comparing of free stream velocity through a pipe under constant water pressure head and inlet air volume flow rate from 0.34 to 2.15 m3 /s. Water mist spray nozzles’ number (1, 4, and 9) and nozzle arrangements (counter, parallel, and vertical to airflow) have been investigated in the present study.

2 Equipment Modification The equipment in Fig. 1 originally was used to study drying system in School of Mechanical Engineering, Universiti Sains Malaysia. Now, this equipment has been modified in order to investigate desiccant dehumidification performance. Water mist spray was added into this system to control relative humidity (RH) of approach air velocity. After modification completion, it was tested for higher humidification Air outlet

Gyro meter

Axial Fan Air inlet Gyro meter Position

Flow conditioning system

Water flow meter Anemometer position

Fig. 1 Experimental rig setup

126

P. Balthazar et al.

Fig. 2 a Parallel, b vertical, and c counter flow arrangement

possibilities and to improve for future development. Subsequently, water is supplied with 40 kPa pressure through a rubber horse and flow control valve at position between flow meter and pump is used to control the water flow rate of nozzle. Water spray nozzles were located inside the 210-mm-diameter PVC pipe. Parallel, vertical, and counter flow arrangements are illustrated in Fig. 2a–c respectively. Airflow direction in Fig. 2 is considered to come from white color bottom surface to upward. Nozzles used in the experimental study as one, four, and nine were arranged in a 1 × 1, 2 × 2, and 3 × 3 combination as shown in Fig. 3a–c respectively. Water supplied to the nozzles through 4.9-mm-internal diameter pipe and water mist were produced with average 70 µm diameter as mentioned in the manual. Then, experiment was conducted under at seven different air volume flow rates between 0.34 and 2.15 m3 /s. Air inlet and outlet conditions were measured using gyrometer iTHX-SD model as shown in Fig. 1. The humidity sensor has ±2% error margin accuracy. Anemometer model TESTO 452 with velocity range up to 10 m/s and error measurement with ±0.05 m/s was positioned after flow conditioning compartment. All the data were measured in real time, and measuring equipment was made sure not affected by the external factors. Finally, experiments were conducted for parallel and counter sprayer nozzle positions and sequentially as mentioned above.

Fig. 3 a 1 × 1, b 2 × 2, and c 3 × 3 vertical nozzle arrangement

Effect of Arrangement and Number of Water …

127

Malaysia enjoys tropical weather year round; however, due to its proximity to water, the climate is often quite humid temperature ranged from a mild 20 to 30 °C average throughout the year. However, monsoon season varies depending on the destination. The southwest experiences its monsoon season from May to September, while November to March is the wettest in the northeast. The experiments were conducted in School of Mechanical Engineering, Universiti Sains Malaysia, situated in Nibong Tebal, Penang, during hottest weather months of February and March. The experiment has been repeated three times in order to increase reliability of experimental data. Experimental ambient conditions were between 28.5 and 30.2 °C and RH between 59 and 78%. All data points were measured continuously throughout the study until getting stable observational readings.

3 Results and Discussion Figures 4, 5, and 6 show the comparison of 1 × 1, 2 × 2, and 3 × 3 nozzles, respectively, under vertical, parallel, and counter flow arrangements. All the figures show similar pattern, which expressed as irrespective of the flow nozzle arrangement, 3 × 3 nozzles shows the highest RH respect to air volumetric flow. However, curvature gradient under each flow type arrangement showed significant difference. Figure 4 illustrates that there was averagely 9.2% RH difference between 3 × 3 and 2 × 2 nozzle, and averagely 3.6% RH difference between 2 × 2 and 1 × 1 nozzle under vertical nozzle arrangement. Vitally, at lower volumetric flow rate between 0.34 and 0.52 m3 /s, 1 × 1 and 2 × 2 nozzles showed difference only around 1–3% RH by value. 1x1V

Average Relative Humidity %

90.0

2x2V

3x3V

85.0 80.0 75.0 70.0 65.0 60.0 0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 4 Comparison of 1 × 1, 2 × 2, and 3 × 3 nozzle under vertical arrangement at 64.8% and 30.1 °C

128

P. Balthazar et al. 1x1P

Average Relative Humidity %

90.0

2x2P

3x3P

85.0 80.0 75.0 70.0 65.0 60.0 0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 5 Comparison of 1 × 1, 2 × 2, and 3 × 3 nozzle under parallel flow arrangement at 61.4% and 29.6 °C 1x1C

Average Relative Humidity %

90.0

2x2C

3x3C

85.0

80.0

75.0

70.0

65.0

60.0 0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 6 Comparison of 1 × 1, 2 × 2, and 3 × 3 nozzle under counter flow arrangement at 60.4% and 30 °C

Figure 5 demonstrates the comparison of 1 × 1, 2 × 2, and 3 × 3 nozzle under parallel flow arrangement. Even though there are slight differences with Fig. 4, horizontal parallel results showed around evenly matched up with vertical results except 1 × 1 nozzle arrangement. Average relative humidity differences between 1 × 1 and 2 × 2, and 2 × 2 and 3 × 3 are 7.8% and 9.7%, respectively, by percentage throughout the air volumetric flow. But notably there is a significant gradient drop between 3 × 3 and 1 × 1 nozzle arrangements. Figure 6 illustrates that counter flow arrangement under horizontal flow showed that 2 × 2 nozzle is close to 3 × 3 arrangement by 3% RH difference by value.

Effect of Arrangement and Number of Water …

129

Comparison of vertical, horizontal parallel and counter flow arrangement have been conducted for 1 × 1, 2 × 2, and 3 × 3 number of nozzles as shown in Figs. 7, 8, and 9. All figures reveal the most effective arrangement under the given number of nozzles respect to air volumetric flow rate. Figure 7 illustrates that parallel nozzle arrangement has the highest humidity absorption for airflow from 0 to 1.5 m3 /s. Potential reason for this phenomenon is that the flow evaporation has greater effect than water mist evaporation. In the present study, flow evaporation is referred as the behavior of water vapor absorption occurred between water flows on bottom pipe surface and air free stream. This might happen because the water-removable hole was located at 1.1 m to the right near the flow conditioning section. The above-mentioned condition increases the influence of the flow evaporation under air and water flowing scenario. However, its influence faded off when parallel and counter flow arrangements showed almost similar humidity ratio value at higher than 1.75 m3 /s air volumetric flow. Vertical nozzle arrangement contributed lowest humidity addition in 1 × 1 arrangement by 1.2 g/kg water content average drop compared to parallel arrangement. Humidity ratio (w) is described as mass of water vapor per unit mass of dry air, and it is calculated during psychometric chart which contains the following equation. Moisture by volume is shown here as notation M w .  w = 0.622

Mw (100 − Mw )

 (1)

Figure 8 displays the 2 × 2 nozzle arrangement, which was completely different than in Fig. 7 pattern. There was enough water mist to mix with airflow inside the pipe and also uniformly mixed with inlet air stream. Vertical arrangement showed averagely 2% higher humidity ratio than parallel water mist nozzle arrangement for 1x1V

0.0222

1x1P

1x1C

Humidity ratio (kg/kg)

0.0220 0.0218 0.0216 0.0214 0.0212 0.0210 0.0208 0.0206 0.0204 0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 7 Comparison of 1 × 1 nozzle arrangements under vertical, parallel, and counter flow at 77% and 29 °C

130

P. Balthazar et al. 2x2V

0.0208

2x2P

2x2C

Humidity ratio (kg/kg)

0.0206

0.0204

0.0202

0.0200

0.0198

0.0196

0.0194

0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 8 Comparison of 2 × 2 nozzle arrangements under vertical, parallel, and counter flow at 67.1% and 29.4 °C 3x3V

0.0225

3x3P

3x3C

Humidity raƟo (kg/kg)

0.0220

0.0215

0.0210

0.0205

0.0200

0.0195

0.0190

0

0.5

1

1.5

2

2.5

Air Volumetric Flow (m3/s)

Fig. 9 Comparison of 3 × 3 nozzle arrangements under vertical, parallel, and counter flow at 64.6% and 29.8 °C

all airflow rate. Vertical arrangement showed the highest humidity ratio and followed by parallel flow and counter flow. Parallel is the second place due to influence of flow evaporation. However, this influence is 0.5% by percentage which is very low between 0.35 and 0.52 m3 /s and highly influenced 1–1.5% by percentage between 0.97 and 2.14 m3 /s.

Effect of Arrangement and Number of Water …

131

Finally, flow evaporation influence is almost completely wiped off (0.1% by percentage) between horizontal parallel and counter flow arrangement at 3 × 3 nozzle arrangement at airflow from 0.4 to 2.4 m3 /s. However, for airflow less than 0.4 m3 /s volume flow rate still showed 1.9% humidity ratio difference between parallel and counter types of flows. Vertical spray arrangement dominated the humidification comprehensively by 6.5%.

4 Conclusions The present work recorded experimental data and summarized that, • Relative humidity inversely proportional to air volume rate. • For all parallel, counter and vertical nozzle arrangements; 3 × 3 combination nozzle arrangement show the highest RH increment. • Vertical 3 × 3 nozzle arrangement shows the highest RH (88.4%) followed by parallel arrangement (87.7%) and counter arrangement (84.7%). • Flow evaporation phenomenon is dominant for counter flow under 1 × 1 nozzle arrangement; however, it is dominant fading as the number of nozzles increases. Acknowledgements This research was funded by Universiti Sains Malaysia, grant code 304/PMEKANIK/6315150 and South Eastern University of SriLanka, grant code UGC/VC/DRIC/PG2017 (II)/SEUSL/01.

References 1. Chengqin, R., Nianping, L., Guangfa, T.: Principles of exergy analysis in HVAC and evaluation of evaporative cooling schemes. Build. Environ. 37(11), 1045–1055 (2002) 2. Cerci, Y.: A new ideal evaporative freezing cycle. Int. J. Heat Mass Transf. 46(16), 2967–2974 (2003) 3. Wong, N.H., Chong, A.Z.: Performance evaluation of misting fans in hot and humid climate. Build. Environ. 45(12), 2666–2678 (2010) 4. Nishimura, N., Nomura, T., Iyota, H., Kimoto, S.: Novel water facilities for creation of comfortable urban micrometeorology. Sol. Energy 64(4–6), 197–207 (1998) 5. Calautit, J.K., Chaudhry, H.N., Hughes, B.R., Ghani, S.A.: Comparison between evaporative cooling and a heat pipe assisted thermal loop for a commercial wind tower in hot and dry climatic conditions. Appl. Energy 101, 740–755 (2013) 6. Wei, J., He, J.: Numerical simulation for analyzing the thermal improving effect of evaporative cooling urban surfaces on the urban built environment. Appl. Therm. Eng. 51(1–2), 144–154 (2013) 7. Muangnoi, T., Asvapoositkul, W., Wongwises, S.: Effects of inlet relative humidity and inlet temperature on the performance of counterflow wet cooling tower based on exergy analysis. Energy Convers. Manag. 49(10), 2795–2800 (2013) 8. Kachhwaha, S.S., Dhar, P.L., Kale, S.R.: Experimental studies and numerical simulation of evaporative cooling of air with a water spray-I. Horizontal parallel flow. Int. J. Heat. Mass. Transf. 41(2), 447–464 (1998)

132

P. Balthazar et al.

9. Kachhwaha, S.S., Dhar, P.L., Kale, S.R.: Experimental studies and numerical simulation of evaporative cooling of air with a water spray-II. Horizontal counter flow. Int. J. Heat. Mass. Transf. 41(2), 465–474 (1998) 10. Sureshkumar, R., Kale, S.R., Dhar, P.L.: Heat and mass transfer processes between a water spray and ambient air–I. Experimental data. Appl. Therm. Eng. 28(5–6), 349–360 (2008) 11. Sureshkumar, R., Kale, S.R., Dhar, P.L.: Heat and mass transfer processes between a water spray and ambient air–II. Simulations. Appl. Therm. Eng. 28(5–6), 361–371 (2008) 12. Belarbi, R., Ghiaus, C., Allard, F.: Modeling of water spray evaporation: application to passive cooling of buildings. Sol. Energy 80(12), 1540–1552 (2006) 13. Husted, B.P., Petersson, P., Lund, I., Holmstedt, G.: Comparison of PIV and PDA droplet velocity measurement techniques on two high-pressure water mist nozzles. Fire Saf. J. 44(8), 1030–1045 (2009) 14. Montazeri, H., Blocken, B., Hensen, J.L.: CFD analysis of the impact of physical parameters on evaporative cooling by a mist spray system. Appl. Therm. Eng. 75, 608–622 (2015) 15. Montazeri, H., Blocken, B., Hensen, J.L.M.: Evaporative cooling by water spray systems: CFD simulation, experimental validation and sensitivity analysis. Build. Environ. 83, 129–141 (2015) 16. Alkhedhair, A., Jahn, I., Gurgenci, H., Guan, Z., He, S., Lu, Y.: Numerical simulation of water spray in natural draft dry cooling towers with a new nozzle representation approach. Appl. Therm. Eng. 98, 924–935 (2016) 17. Balthazar, P., Ismail, M.A., Soberi, A.S.B.A., Muhammad Iftishah, R., Hussin, M.: Experimental study of the effect of piccolo tube pipe on the air-conditioning experimental rig. J. Adv. Res. Fluid Mech. Therm. Sci. 53(1), 95–105 (2019)

Vibration Condition Monitoring of Spur Gear Using Feature Extraction of EMD and Hilbert–Huang Transform A. Krishnakumari, M. Saravanan, M. Ramakrishnan, Sai Manikanta Ponnuri and Reddy Srinadh

Abstract The vibration condition monitoring is the process of monitoring the vibration signals in machinery to identify a significant change in the development of fault. Gears are important rotary devices for power and torque transmission. Study on gear teeth relationship is considered as one of the most complicated applications because the speed, load conditions, and application cause different failures, leading to nonstationary operating conditions. Hence, an appropriate signal processing technique to identify the gear fault diagnosis plays a vital role in condition monitoring system. This work attempts the Hilbert–Huang transform (HHT) to identify the effect of the new time–frequency distribution, which increases the performance of fault diagnosis in gear. Also, the method using HHT is compared with fast Fourier transform (FFT). As a novel approach, the statistical feature called energy was calculated for all intrinsic mode functions (IMF) obtained from the empirical mode decomposition (EMD) of the signal which is suitable for the selection of IMF for applying HHT. The fault diagnosis of gear is done clearly using the present approach.

A. Krishnakumari (B) · M. Saravanan · S. M. Ponnuri · R. Srinadh Department of Mechanical Engineering, School of Mechanical Science, Hindustan Institute of Technology and Science, Padur, Chennai, India e-mail: [email protected] M. Saravanan e-mail: [email protected] S. M. Ponnuri e-mail: [email protected] R. Srinadh e-mail: [email protected] M. Ramakrishnan Centre for Simulation and Engineering Design, Hindustan Institute of Technology and Science, Padur, Chennai, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_13

133

134

A. Krishnakumari et al.

1 Introduction The vibration signals and acoustic emissions are the most common waveform in condition monitoring techniques. In the past literature, three types of waveform signal analysis have been tried such as time-domain, frequency-domain, and time– frequency-domain analysis. In time domain, the signals are collected at a specific time interval. Traditional statistical characteristics such as the root mean square (RMS), energy, and peak were used to analyze the time-domain signal analysis to monitor changes in signals for fault diagnosis [1]. Research has been conducted on machine vibration to detect gear box failures in the initial stages using time-domain average. In this case, statistical fault detection is done by assuming signals are stationary and provide information about the mean signal over time. As a result, the effect of local transient phenomena may be lost [1]. Therefore, FFT was used to identify the fault frequency in rotating machines as frequency-domain signal processing technique. The authors [2–4] had applied spectrum analysis using FFT for fault diagnosis of rotating machines such as induction motors, bearings, and rotor. In all the specified works, the importance of frequency-domain analysis is that to identify the certain frequency components of researcher’s interest. The limitations of frequencydomain analysis were handling non-stationary waveform signals, which are more common, when machine faults occur [2–4]. [5] reported that, as the signals collected from the faulty machine are naturally non-stationary, the FFT is not suitable for non-stationary signals [5]. But non-stationary signals have symptoms of sinusoidal component, random noise, and broadband impulsive components. Hence, the author [6] stated that non-stationary, time–frequency-domain analysis is considered to be appropriate to capture all features of the signals [6–8]. The short-time Fourier transform (STFT)/spectrogram was being applied to compute the time–frequency distribution of the gear vibration signal, and the spectrogram was powerful tool for identifying the early detection of local gear. But then, it was stated that both Wigner– Ville distribution (WVD) and STFT transforms have been linked to high interference terms and window-based analysis [9]. The interpretation in terms of this intervention was difficult to understand the fault conditions. Hence, to overcome WVD and STFT defects, time–frequency analysis of non-stationary signals was analyzed using Cohen class distribution and wavelet transform (WT), respectively. Also, many researchers use gear, bearing, and broken bar fault diagnosis by Zhao–Atlas–Marks (ZAM) transform. The ZAM provides the best performance compared to the traditional technique of STFT [10–12]. [13] presented the review on the application of wavelet transforms in machine condition monitoring and fault diagnosis. Furthermore, [14] carried out testing of fault diagnosis in a multistage gearbox under transient load by using wavelet transforms [13, 14]. On the other hand, [15] had applied EMD and local Hilbert energy spectrum method which could extract the characteristics information of the gear fault vibrations [15]. [16] had applied a novel approach, which is a process of combining HHT and dimensionless frequency (DLF) normalization to explain the errors of gear transmission system [16]. [17] had attempted to analyze a gear fault diagnosis method based on HHT and self-organizing feature map

Vibration Condition Monitoring of Spur Gear …

135

(SOM) neural networks. Nonlinear and non-stationary gear vibration signals were processed using two above-mentioned methods [17]. [18] stated the HHT spectrum stated that it was built with almost monochromatic IMFs, not just with multiple frequency components or IMFs containing pseudo-components [18]. From the above literature, it is understood that there are many time–frequency methods which had been applied for identifying the fault frequency of interest. The previous authors did not inform the feature extraction of EMD also called as energy. One of the most important notations in signal analysis is how much of energy it generates to produce the signal, but this is not attempted in the past literature. Also, HHT with EMD had been applied in identifying the frequency by removing the noise. But then, identification of monochromatic signal in EMD and processing suitable IMFs were the tasks in the previous literatures as it requires knowledge. Hence, in this work, diagnosing gear fault was done using HHT with EMD. And the appropriate IMFs were selected using statistical feature called energy. So, the monotonic signals are identified and eliminated in the process. Further, HHT was applied only for suitable IMFs.

2 Gear Vibration Signals The vibration signals are taken from the single-stage spur gear box for three conditions such as normal, broken tooth, and broken tooth underloading. For this work, the first wheel is new and is assumed to free from defects. Errors were created using EDM in the two pinion wheels in order to control the size of errors. The signals are taken using accelerometer, which is explained in the author’s previous work [11]. The pinion gear rotates in 400 rpm (6.67 Hz) with 20 number of teeth, and the information was collected using 4 channel data acquisition module (NI9233) and Dewesoft application software. The length of the sample was taken as 2048 number of samples per second in all conditions. The characteristic fault frequency of gear can be represented and calculated as a function of rotational frequency (T *n), where “n” is the rotational frequency and “T ” is the number of teeth on gears. The vibration signals are shown in Fig. 1.

3 Time–Frequency Representations In this section, the time–frequency representation of empirical mode decomposition (EMD) and Hilbert–Huang transform (HHT) and statistical feature of IMF signals have been discussed.

136

A. Krishnakumari et al.

Normal signal

Acceleration (g)

Broken tooth

Broken tooth under loading

Time (sec) Fig. 1 Vibration signals for various conditions

3.1 HHT The process of EMD and Hilbert spectral analysis (HSA) results in Hilbert–Huang transform (HHT). The HHT provides a new approach to identify non-stationary and nonlinear time series data. EMD method is used in HHT to break the IMF. It is applied to HSA method for IMFs to gain instant frequency information. The length of the IMFs is the original signal because the signal time is depleted in the domain, and the HHT preserves different frequency properties.

3.1.1

EMD

EMD decomposes the signal X(t) using IMF and residual rk(t) shifting process [15]. Detailed description is given below in Eqs. (1) and (2). 1. The local maximum and minimum for signal X(t) was calculated to build a upper envelope “s + (t)” and lower envelope “s − (t)” Mean envelope for ith iteration, mk, i (t)

Vibration Condition Monitoring of Spur Gear …

m k, i (t) =

137

1 [(s + (t)) + (s − (t))] 2

(1)

2. As C k (t) = X(t) for the first iteration, subtract the mean envelope from residual signal, Ck (t) = Ck (t) − m k, i (t)

(2)

The iteration is repeated till C k (t) values satisfies criteria of IMF, otherwise steps 4 and 5 can be skipped. 3. A new residual is computed, if C k (t) matches the criteria of an IMF. In order to update the residual signal, subtract the kth IMF from the previous residual signal, rk (t) = rk−1 (t) − Ck (t)

(3)

4. Then start from step 1, using the residual signal obtained as a new signal r k (t), and store C k (t) as an intrinsic mode function. The original signal for “N” intrinsic mode functions is represented as, X (t) =

3.1.2

i=1 N

ci (t) + r N (t)

(4)

Hilbert Spectral Analysis (HSA)

Hilbert spectral parameters are as follows: 1. Using empirical mode decomposition, a signal which is complex in nature is considered as X c (t) and it is decomposed into N intrinsic mode functions. 2. The energy a2i (t) and its instantaneous frequency ωi (t) are computed as,  X c (t) = H

N  i=1

 ci (t) =

N 

ai (t)e

t 0

ωi (t)dt

(5)

i=1

3. By overlaying N time series [t, wi (i), a2i (t)] onto a discretized time–frequency plane HHT creates the Hilbert spectrum plot. Hence, imf.insf = [wi (t)] and imf.inse = [a2i (t)].

3.1.3

Feature Extraction

The statistical feature of energy is calculated from IMF signal which indicates effect of noise in signal condition. Time-domain statistical features are used as a parameter

138

A. Krishnakumari et al.

to identify the presence of fault as well as characteristics of signal. In the present work, this particular feature called “energy” is used for the selection of suitable IMF for applying HSA.   Energy = sum IMF.∧ 2

(6)

4 Results and Discussions The vibration signals of three conditions such as normal, broken tooth, and broken tooth under loading is subjected to signal processing techniques of HHT and FFT. The time–frequency toolbox in MATLAB 2018 was used for the signal analysis. The results of EMD, HHT, and FFT are discussed in the following sections.

4.1 EMD The signals are analyzed during empirical mode decomposition which is a timedomain function. Figure 2 shows the EMD signal of normal, broken tooth, and broken tooth under loading conditions. The signals are processed and decomposed into various IMF components by EMD. At the highest frequency band, which is compatible with the number of frequency families and the rest of the sounds are filtered. Therefore, the frequency families were separated and the noise decreases. The IMF component with high-frequency band was analyzed. From figures, it is clear that IMFs indicate difference of amplitude in every frequency range with respect to time. Therefore, the selection of suitable IMFs was done by using feature calculation. Using MATLAB, the feature energy is calculated for all the IMF functions which are obtained from EMD. The results are plotted and shown in Fig. 3 for three conditions of gear. From the results, it is observed that IMF1 will have most of the white Gaussian noise from measurement. From the literature, it is found that signal-to-noise ratio. SNR of IMF1 will be more. Hence, considering IMF1 may cause bias in fault diagnosis and it is not being considered for applying HSA. IMF2–IMF5 have shown low-frequency components and that has been considered for HSA. But from IMF6 to IMF10, monotonic functions are observed. Therefore, fault diagnosis of gear is being done for IMF2, IMF3, IMF4, and IMF5. But the results have been identified in better manner in IMF4 and IMF5. Also, it is identified that the energy level is reduced from normal to faulty condition of gear.

Vibration Condition Monitoring of Spur Gear …

139

Fig. 2 a Normal condition vibration signal, b broken tooth condition vibration signal, and c broken tooth under loading condition vibration signal

140

A. Krishnakumari et al.

Fig. 3 Energy under normal to faulty condition for gear

4.2 HSA and FFT The HSA and FFT are applied for all the vibration signals of normal and faulty conditions of the gear. The results are shown in Figs. 4, 5, and 6, in which Fig. 4a, b shows HSA and FFT results of normal signal, respectively. From the results, it is observed that the frequency value of normal condition is as 123.5 and 129.9 Hz identified which is lesser than 133.4 Hz (T *n = 20*6.67 = 133.4 Hz) of gear characteristic frequency. Hence, both HSA and FFT are better in identifying the normal condition of gear. The results of HSA and FFT are shown in Fig. 5a, b for broken tooth condition of gear. In Fig. 5a, the HSA result shows that 133.8 Hz of characteristic frequency of gear clearly, whereas in Fig. 5b, the FFT result shows 135.2 Hz. Hence, it is observed that the performance of HSA in identifying the fault of gear is better than FFT. Similarly, Fig. 6a, b shows the results for broken tooth under loading condition of gear. Comparing the results, it is observed again HSA is showing better identification than FFT. Although many researchers have conducted the FFT for condition monitoring technique which is identified for non-stationary signals, it is not a better technique to identify the fault in non-stationary signal analysis. It is only identifying the beginning of error.

Vibration Condition Monitoring of Spur Gear …

(a)

Amplitude

(b)

Fig. 4 Normal condition of gear a HSA and b FFT

141

142

(a)

Amplitude

(b)

Fig. 5 Broken tooth conditions a HSA and b FFT

A. Krishnakumari et al.

Vibration Condition Monitoring of Spur Gear …

(a)

Amplitude

(b)

Fig. 6 Broken tooth load under conditions a HSA and b FFT

143

144

A. Krishnakumari et al.

5 Conclusion A novel method for analyzing a non-stationary vibration signal of gear by using Hilbert–Huang transform (HHT) with energy calculations was done in this work. The performance comparison of HHT in identifying fault with FFT was good as the noise has been removed through EMD. The estimation of energy is useful in understating the fault as well as the selection of appropriate IMF functions to apply the Hilbert spectral analysis. The energy level is getting reduced from normal condition when fault occurs in the gear. Hence, this method is suitable for the fault diagnosis of gear in precise manner. This work can be extended for bearing and rotary devices fault diagnosis in modern condition monitoring system.

References 1. Wang, W., McFadden, P.D.: Early detection of gear failure by vibration analysis I. Calculation of the time frequency distribution. Mech. Syst. Signal Process. 7(3), 193–203 (1993) 2. Schoen, R.R., Habetler, T.G.: Effects of time-varying loads on rotor fault detection in induction machines. IEEE Trans. Ind. Appl. 31(4), 900–906 (1995) 3. DeAlmeida, R.G.T., DaSilva, V.S.A., Padovese, L.R.: New technique for evaluation of global vibration levels in rolling bearings. Shock Vib. 9, 225–234 (2002) 4. Liu, Z., Yin, X., Zhang, Z., Chen, D., Chen, W.: Online rotor mixed fault diagnosis way based on spectrum analysis of instantaneous power in squirrel cage induction motors. IEEE Trans. Energy Convers. 19, 485–490 (2004) 5. Pan, M.C., Sas, P.: International conference on signal processing proceedings, ICSP 2, 1723– 1726 (1996) 6. Samuel, P.D., Darryll, P.J., David, L.G.: A comparison of stationary and non-stationary metrics for detecting faults in helicopter gearboxes. J. Am. Helicopter Soc. 45(2), April (2000) 7. Brennan, M.J., Chen, M.H., Reynolds, A.G.: Use of vibration measurement to detect local tooth defects in gears. Sound Vib. 31(11), 12–17 (1997) 8. Larder, B.: An Analysis of HUMS vibration diagnostic capabilities. In: Proceedings of the 53rd AHS International Annual Forum, American Helicopter Society, Alexandria, VA, 1997 9. Russel, P.C., Cosgrave, J., Tomtsis, D.: Extraction of information from acoustic vibration signals using Gabor transform type devices. Meas. Sci. Technol. 9, 1282–1290 (1998) 10. Aharamuthu, K., Ayyasamy, E.P.: Application of discrete wavelet transform and Zhao-AtlasMarks transforms in non-stationary gear fault diagnosis. J. Mech. Sci. Technol. 27(3), 641–647 (2013) 11. Krishnakumari, A., Saravanan, M., Venkatesan, G., Jain, S.: Application of Zhao-Atlas-Marks transforms in non-stationary bearing fault diagnosis. Procedia Eng. 144, 297–304 (2016) 12. Rajagopalan, S., Restrepo, J.A., Aller, J.M., Habetler, T.G., Harley, R.G.: Non-stationary motor fault detection using recent quadratic time-frequency representations. IEEE Trans. Ind. Appl. 44(3), 735–744 (2008) 13. Peng, Z.K., Chu, F.L.: Application of the wavelet transform in machine condition monitoring and fault diagnostics: a review with bibliography. Mech. Syst. Signal Process. 18, 199–221 (2004) 14. Kar, C., Mohanty, A.R.: Vibration and current transient monitoring for gearbox fault detection using multi-resolution Fourier transform. J. Sound Vib. 311(1–2), 109–132 (2008) 15. Cheng, J., Yu, D., Tang, J., Yang, Y.: Application of frequency family separation method based upon EMD and local Hilbert energy spectrum method to gear fault diagnosis. Mech. Mach. Theory. 43(6), 712–723 (2008)

Vibration Condition Monitoring of Spur Gear …

145

16. Wu, T.Y., Chen, J.C., Wang, C.C.: Characterization of gear faults in variable rotating speed using Hilbert-Huang transform and instantaneous dimensionless frequency normalization. Mech. Syst. Signal Process. 30, 103–122 (2012) 17. Cheng, G., Cheng, Y.L., Shen, L.H., Qiu, J.B., Zhang, S.: Gear fault identification based on Hilbert-Huang transform and SOM neural network. Measurement 46(3), 1137–1146 (2013) 18. Zhang, Y., Tang, B., Xiao, X.: Time–frequency interpretation of multi-frequency signal from rotating machinery using an improved Hilbert-Huang transform. Measurement 82, 221–239 (2016)

Comparative Study of PWM Technique for Switching Loss Reduction and Acoustic Noise Reduction in VSI-Fed Drives Tarang Kalaria, Tapankumar Trivedi, Vinod Patel, Rajendrasinh Jadeja and Chandresh Patel

Abstract Variable-frequency drives (VFDs) are widely used in the industry and offgrid PV-based pumping applications due to their versatile operation. The operating performance of VFDs in the given environment is mainly affected by switching losses and generation of acoustic noise. During the development of PWM techniques of VSI, both the attributes are considered separately. In this paper, different methods aimed at switching loss reduction and acoustic noise reduction are reported so that both the operating requirement of drives are satisfied. Different bus clamping methods such as DPWM0, DPWM1, DPWM2, and DPWM3 are studied and compared. Simulation results demonstrate the effectiveness of these schemes in reducing inverter switching losses and reducing acoustic noise of variable-frequency drive.

1 Introduction Nowadays, voltage-source inverters have become popular in many industrial drive applications. Adjustable speed drive (ASD) applications require variable-voltage variable-frequency output which is obtained from PWM-VSI. The off-grid PV-based pumping applications have necessitated reduction in losses and thereby maximizing utilization of installation. Numerous modulation schemes are available to accomplish task of obtaining variable-frequency variable-voltage output from inverter.

T. Kalaria · T. Trivedi (B) · R. Jadeja Electrical Engineering Department, Marwadi Education Foundations Group of Institutions, Rajkot, India e-mail: [email protected] R. Jadeja e-mail: [email protected] V. Patel · C. Patel Amtech Electronics Pvt. Ltd, Gandhinagar, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_14

147

148

T. Kalaria et al.

The space vector pulse-width modulation (SVPWM) is the most commonly found PWM scheme which utilizes the said approach. Modulation wave of SVPWM is continuous which results in more switching losses in inverter. On the contrary, development of advanced discontinuous pulse-width modulation (DPWM) scheme involves idea of freedom of distribution of zero vector dwell time between two possible zero voltage vectors which leads to possibility of many numbers of DPWM methods. DPWM schemes have discontinuous modulation wave which results in reduction in switching losses [1]. While switching loss reduction has always been a performance parameter, another problem associated with such drives is acoustic noise produced in the drive. The acoustic noise level must be maintained as specified by IEC 60034-9 norms [2]. Both the problems stated above are attributed to switching scheme adopted for the inverter. Switching of the PWM inverter at fixed frequency concentrates on acoustic spectrum around integer multiples of switching frequency which results in tonal frequency noise. A common practice to reduce acoustic noise is to spread harmonic spectrum of voltage and current [3]. The discussion leads to the fact that variablefrequency carrier wave can be used along with DPWM method to optimize switching loss as well as acoustic noise. Acoustic noise reduction by spread spectrum switching was first introduced by JT Boys et al. [3]. Acoustic noise reduction by variable-frequency carrier with DPWM was discussed in [4, 5]. Development of DPWM was discussed in [4, 6]. Switching loss reduction and its analysis was discussed in papers [4, 7, 8]. The present paper discusses optimization of acoustic noise and switching loss simultaneously using different DPWM methods such as DPWM0, DPWM1, DPWM2, and DPWM3 and compares the performance with the well-adopted SVPWM method. Optimization of switching loss is done by using discontinuous PWM schemes, while elimination/reduction of acoustic noise is obtained by spreading harmonics spectrum by using variable switching frequency instead of constant switching frequency. The improvement in performance is validated using mathematical model of IM and thermal module of VSI in PSIM simulation tool. The paper is organized as follows: Sect. 2 discusses the development of discontinuous PWM schemes, whereas Sect. 3 discusses the switching loss reduction and its analysis. The concept of acoustic noise prediction and reduction is revisited in Sect. 4, the simulation results for comparative study of different PWM schemes for acoustic noise reduction and switching loss reduction.

2 Advanced PWM Scheme In PWM of VSI, commanded voltage is given either by three-phase voltage references or by voltage vector. Each cycle of modulation wave contains two active states and one zero state which is divided into positive dc bus clamping (111) and negative dc bus clamping (000). The position of zero vector defines different pulse-width modulation techniques. If zero vector is located at end of cycle of modulation wave,

Comparative Study of PWM Technique for Switching Loss …

149

inverter is clamped at negative DC bus, while zero vector is located at middle of cycle of modulation wave, inverter is clamped at positive DC bus. The commonly used discontinuous PWM techniques using bus clamping are DPWM0, DPWM1, DPWM2, and DPWM3 [1]. Mathematically, PWM modulation wave is generated by conditioning reference voltages. Let Ua , Ub , Uc be three-phase reference voltages with Um ∗ Ts /Udc as magnitude and ω be angular frequency [1, 4, 7]. Um Ts sin(ωt) Udc   Um Ts 2π Ub = sin ωt − Udc 3   Um Ts 2π Uc = sin ωt + Udc 3

Ua =

(1)

Umax = max(Ua , Ub , Uc ) Umin = max(Ua , Ub , Uc )

(2)

where Umax be maximum from reference voltages, and Umin be minimum from reference voltages. Zero vector can be given as [4]: Uoffset = (Ts − Umax )(1 − k) − k ∗ Umin

(3)

Modulation wave U ∗ can be given by adding three-phase reference wave to zero vector wave. U ∗ = Ui + Uoffset where i = a, b, c

(4)

k is conditioning parameter whose value can be between 0 and 1. When k = 0, inverter is clamped to positive DC bus, while k = 1 represents negative dc bus clamping of inverter. The condition of different values of k can be derived from shifted reference voltages Uax , Ubx , and Ucx is writte as [1, 4]. Uc − Ub sin(ψm ) √ 3   √ √ 1 Uc − Ub 3Ua = Ub cos(ψm ) − 3sin(ψm − √ 2 2 3

Uax = Ua cos(ψm ) − Ubx

Ucx = −Uax − Ubx Umid = max(Uax , Ubx , Ucx ) + min(Uax , Ubx , Ucx )

(5) (6)

where ψm is phase shift from reference voltage signal [1]. If Umid > 0, put k = 0 else k = 1 in Eq. (3) [6]. If ψm = π6 , modulation wave is of DPWM0 (Fig. 1). If

150

T. Kalaria et al.

Fig. 1 Modulation signal for a SVPWM, b DPWM0, c DPWM1, and d DPWM2

ψm = 0, modulation wave is of DPWM1 (Fig. 1). If ψm = − π6 , modulation wave is of DPWM2 (Fig. 1). If ψm = − π3 , modulation wave is of DPWM3 [4, 6]. For k = 0.5, modulation wave is of SVPWM (Fig. 1). The switching loss analysis with the above-stated methods is discussed in the following section.

3 Switching Loss The switching losses of PWM-VSI drives depend upon load current and increase with magnitude of current. This relation is nearly linear; i.e., switching losses are proportional to magnitude of current [4]. With SVPWM method, IGBTs of inverter switch continuously, whereas with discontinuous methods, IGBTs of inverter switch for only 270° of modulation wave cycle. It is worth noting that degree of reduction of switching loss depends on power factor [4], i.e., location of peak of line current and interval of dc-rail clamping of inverter in each modulation cycle. Since switching losses are dependent on magnitude of phase current, choice of DPWM technique with reduced switching loss contributes to better performance of a drive. Average

Comparative Study of PWM Technique for Switching Loss …

151

switching loss per carrier cycle can be given as [4, 9]:  1 Udc (ton + toff ) 2π f i (θ )dθ Ps,wave = 2π 2Ts 0    0, Ua∗∗  ≥ U2dc f ia (θ ) = |i a |, Ua∗∗  < U2dc

(7) (8)

where f ia is function of current passing through power electronics switches. On normalizing Ps,wave to P0 , where P0 is power loss for SVPWM, switching loss function of a DPWM method can be found as [4]: P0 =

Udc Imax (ton + toff ) π TS SLF =

Ps,wave P0

(9) (10)

Switching loss function for DPWM developed in this paper can be given as [4]: ⎧√

4π  3 π π ⎪ ⎨ 2 cos 3 − ϕ , − 2 < ϕ < − 2 1 π π π SLFDPWM0 = 1√− 2 sin 3 − ϕ , − 2 < ϕ < 6 ⎪  ⎩ 3 π cos 3 − ϕ , + π6 < ϕ < π2 2 ⎧√

3π  3 ⎪ − π2 < ϕ < − π3 ⎨ 2 cos 2 − ϕ , 1 π SLFDPWM1 = 1√− 2 sin 2 − ϕ , − π3 < ϕ < π3 ⎪ 3 π  ⎩ cos 2 + ψ − ϕ , + π3 < ϕ < π2 2 √

 3 cos 5π − ϕ ,  − π2 < ϕ < − π6 2 3 SLFDPWM2 = 1 − 21 sin 2π − ϕ , − π6 < ϕ < π6 3

(11)

(12)

(13)

4 Acoustic Noise Prediction In PWM-VSI induction motor drive, VSI injects harmonics of significant amplitude in current of motor resulting in acoustic noise, vibration of stator core and magnetic force wave having same frequency components in the spectrum [2]. Impact of magnitude of harmonic current on acoustic noise and on magnetic force wave is well predicted. When induction motor is fed from VSI, components of stator current can be given in the form of fundamental current i 1 and harmonic current i n . These can be defined as follows [2]: i 1 = I1 cos(ω1 t − φ1 )

(14)

152

T. Kalaria et al.

i n = Im cos(mω1 t − φn )

(15)

When these currents are passed through stator winding, magnetic field is produced due to i 1 and i n [2]. This magnetic field can be given as [2]: f (θ, t) = f 1 (θ, t) + f n (θ, t) f (θ, t) = F1 cos(θ − ω1 t − φ1 ) +

 n

Fn cos(nθ − ωn t − φn )

(16) (17)

The interaction of this magnetic field and current results in pulsating torque which produces force wave whose mathematical representation is given as follows [2]: b(θ, t) = B1 cos(θ − ω1 t − φ1 ) +



Bm cos(mθ − ωm t − φm )

m

(18)

where m is mode order, ωm is angular frequency, and φm and Pm are phase angle and amplitude, respectively, of m th order force wave [2]. In fact, ωm = (ωn − ω1 ) if ωn and ω1 have same direction of rotation [2]. This force wave is directly related to emitted acoustic noise. Hence, acoustic noise behavior of IM drive can be predicted on studying FFT of current or voltage of induction motor [3, 5]. In fixed frequency PWM, acoustic noise is concentrated in narrow band around integral multiples of switching frequency [2, 5]. This acoustic noise can be reduced by using variable switching frequency due to spread of harmonics spectrum [2, 3, 5]. By creating gap in selected switching frequencies, excitation of resonant frequency can be avoided [2, 10].

5 Simulation Results and Discussions Simulation of advanced DPWM is performed under PSIM environment. PWM-VSI is simulated with RL load as well as induction motor. The rating of induction motor is 45 kW, 81 A, 415 V, 50 Hz, 1470 RPM. IGBTs used in VSI are Semikron 6pack module with parameters Vce,max = 1200 V, Ic,max = 150 A, and T j,max = 175 ◦ C. Figure 2a, b shows graphical comparison of THD and power loss for different PWM schemes obtained from simulation results for RL load, whereas Fig. 2c, d shows graphical comparison for IM load. The waveform obtained in simulation is represented by Fig. 3. For acoustic noise reduction, different PWM schemes with random frequency carrier wave are used. Out of all of discontinuous PWM techniques, DPWM2 has shown optimum results in terms of switching losses. Waveform of DPWM-RPWM is shown in Fig. 3. Comparison of different DPWM methods shows that DPWM1 has better performance in terms of switching loss whereas comparable performance in terms of THD. The reduction of switching losses is nearly 13% and 10% for DPWM1 and DPWM2,

Comparative Study of PWM Technique for Switching Loss …

153

Fig. 2 Comparison of switching losses and THD of different PWM schemes with a and b RL load; c and d IM load

Fig. 3 FFT of line current with fixed frequency and random PWM for IM load: fixed frequency a line current, b FFT of line current, random PWM: c line current, d FFT of line current

154

T. Kalaria et al.

respectively, than that of SVPWM at lower modulation index, while reduction of switching loss is nearly 8% and 17% for DPWM1 and DPWM2, respectively, than that of SVPWM at higher modulation index. It is observed that DPWM2 is more efficient than other DPWM schemes. This is due to the fact that bus clamping in DPWM2 occurs during peak value of line current in IM load. From waveform obtained, it can be observed that FFT of current with RPWM in Fig. 3 is spread evenly than that with constant frequency carrier wave in Fig. 2. Its harmonics around 5 kHz and its integral multiples reduce significantly, but overall harmonics of line current with DPWM2-RPWM increases due to increase in magnitude of higher-order harmonics. Reduction in harmonics component around 5 kHz is about 1 dB in FFT of current with constant switching frequency than that of with variable switching frequency. At higher fundamental frequencies, PWM techniques mentioned in the paper lose their linear modulation property and the inverter generates six step output. Although this reduces power losses in the semiconductor, the THD of the current is increased significantly. Nevertheless, most of the drives are operated at base speed or lower and hence the study carried out in the work is useful for the PWM-VSI-fed drives.

6 Conclusion In this paper, acoustic noise spectra and switching loss of VSI-fed induction motor is investigated using different advanced discontinuous PWM method with randomization in carrier wave its performance is compared. Simulation results show both current and voltage harmonics components around switching frequency with SVPWM and DPWM with constant switching frequency over with random switching frequency. Simulation work has reported power loss of inverter with different PWM schemes. Comparison of different PWM methods for VSI shows that DPWM2 method reduces switching loss by 17% in comparison with SVPWM, whereas randomized PWM reduces the acoustic noise by 2 dB by keeping average of switching frequency constant.

References 1. Holmes, D.G., Lipo, T.A.: Pulse width modulation for power converters : principles and practice. Wiley (2003) 2. Binojkumar, A.C., Saritha, B., Narayanan, G.: Experimental comparison of conventional and bus-clamping PWM methods based on electrical and acoustic noise spectra of induction motor drives. IEEE Trans. Ind. Appl. 52, 4061–4073 (2016) 3. Boys, J.T., Handley, P.G.: Spread spectrum switching: low noise modulation technique for PWM inverter drives. IEE Proc. B Electr. Power Appl. 139, 252 (1992) 4. Hava, A.M., Kerkman, R.J., Lipo, T.A.: A high-performance generalized discontinuous PWM algorithm. IEEE Trans. Ind. Appl. 34, 1059–1071 (1998)

Comparative Study of PWM Technique for Switching Loss …

155

5. Kumar, A.C.B., Narayanan, G.: Variable-switching frequency PWM technique for induction motor drive to spread acoustic noise spectrum with reduced current ripple. IEEE Trans. Ind. Appl. 52, 3927–3938 (2016) 6. Reddy, T.B., Ishwarya, K.: Simple and efficient generalized scalar PWM algorithm for VSI fed induction motor drives. In: 2012 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES), IEEE, 1–6 2012 7. Li, J., Liang, D.T.W.: Novel control scheme for reduced switching loss in inverter drives. In: 1997 IEEE International Electric Machines and Drives Conference Record, IEEE, MB3/5.1– MB3/5.3 8. Gowri, K.S., Reddy, T.B., Babu, C.S.: Switching loss characteristics of advanced DPWM methods using space vector based double switching clamping sequences. In: 2009 IEEE Symposium on Industrial Electronics & Applications, IEEE, 818–822 2009 9. Guillermo, R.A., Valenzuela, M.A., Weaver, M.D., Lorenz, R.D.: The impact of switching frequency on PWM AC drive efficiency. In: 2016 IEEE Pulp, Paper & Forest Industries Conference (PPFIC), IEEE, 153–163 2016 10. Jadeja, R., Ved, A., Chauhan, S.: An Investigation on the performance of random PWM controlled converters. Eng. Appl. Sci. Res. 5, 876–884 (2015)

An Improved Genetic Algorithm for Production Planning and Scheduling Optimization Problem Aditya Kunapareddy and Gopichand Allaka

Abstract This paper proposes an improvement of genetic algorithm (GA) for optimization of production planning and scheduling in the manufacturing industry. The problem is dynamic, combinatorial and multidimensional in nature with factors like production selection, production line allotment, manufacturing sequence, order quantity, etc., to be solved in sync. The efficiency and effectiveness of the proposed GA are demonstrated by a case study. Selection of parameters of the proposed GA is done using Taguchi experiment method. Performance comparison is done using six optimization solvers, namely pattern search solver, simulated annealing, tabu search, stochastic gradient descent, ant colony optimization and traditional genetic algorithm method.

1 Introduction Production planning and scheduling are the most complex problems which deal with unpredictable scenarios and overlapping requirements. This class of problem can be classified as an NP-hard problem. Due to large data sets and objectives, no optimal solution can be obtained as it would require indefinite time and the solution obtained is far from perfect owing to the multiplicity of constraints involved. Hence, only near optimal solutions can be obtained. A solution to this problem is usually solved using meta-heuristic class algorithm. To solve this problem, triple optimization: product selection (what to make), production line allocation (where to make) and manufacturing sequence (order to make) is to be done in such a manner that total profit of the company and customer satisfaction are maximized while satisfying the given constraints. Supply chain of a manufacturing company is a system of interconnected operations like procurement, processing and distribution. A number of sub-problems in production planning and scheduling exist. Khoshnevis and Chen [1] were among A. Kunapareddy (B) · G. Allaka Department of Mechanical Engineering, Swarnandhra College of Engineering and Technology, Seetharampuram, Narsapur, Andhra Pradesh 534280, India © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_15

157

158

A. Kunapareddy and G. Allaka

the first to approach integration of process planning and scheduling functions using heuristic approach. Morad and Zalzala [2] proposed a genetic algorithm for simultaneous optimization of production planning and scheduling as a multi-objective problem. Maravelias and Sung [3] presented three solution strategies to integrate production planning and scheduling, namely hierarchical, iterative and full-space strategy. The hierarchical and iterative strategy is decomposition-based method in which the problem is decomposed into master sub-problem with the planning model and surrogate scheduling as its core components and slave subprogram. The difference between hierarchical and iterative strategies lies in the flow of information from master subprogram to slave subprogram as the latter involves a feedback loop. In full-space strategy, planning model and scheduling are integrated. Though these strategies can be applied to solve large-scale problems, it is difficult to obtain a global optimal solution. Shao et al. [4] developed a genetic algorithm-based approach for integration and optimization of process planning and scheduling. However, the production planning and scheduling operations are unintegrated in this approach and the difficulty to obtain global optimal solution persists. Moon et al. [5] used a heuristicbased approach for integration of production planning and scheduling in multiple supply chain scenarios. A symbiotic evolutionary algorithm was presented by Kim et al. [6] for integrating and concurrently solving process planning and scheduling. The structural inefficacy of the solution proposed in the symbiotic evolutionary algorithm inhibits the solution from obtaining global optimization. Lui and Fang [7] designed a set of heuristics to solve the problem of integrating planning and scheduling though the scope of interaction is narrow. Zhang and Merchant [8] presented an integrated processing planning model with means of the distributed approach. Dao et al. [9] proposed a novel genetic algorithm to completely integrate production planning and scheduling. This robust algorithm provides best solution quality with less computation time, but material constraints used in this approach are limited and the effectiveness of the chromosome is tested against only few optimization solvers. The paper discusses an improved genetic algorithm with novel crossover, and mutation operation is introduced to optimize production planning and scheduling and its integration.

2 Research Methodology 2.1 Problem Modelling A manufacturing facility has N different production lines with capacity to produce n different products at any given point of time. A penalty is imposed for low-quality late delivery and returns, while the resources like labour, material, capital are limited. The following assumptions are made: 1 The manufacturing facility can produce any given combination of products.

An Improved Genetic Algorithm for Production Planning …

159

2 The manufacturing facility can function 24 h a day, 7 days a week without any breaks. 3 All machines are available on the call.

2.2 Chromosome Encoding Feasible solutions for the problem, i.e. production line, are encoded as chromosomes. Each chromosome encoded consists of string containing product and their manufacturing sequence. Chromosomes are randomly generated. Length of the string depends on the resource allocation constraints. Each chromosome consists of two parts: resource allocation and manufacturing sequence. The first column in resource allocation chromosome is labour for three production lines. The second and third columns in resource allocation chromosome are material required and capital, respectively. The values of the resource allocation chromosome are presented in percentage which is why the sum of each column is 100. In manufacturing sequence chromosome, the values in the cell represent the product and the location of the cell represents corresponding manufacturing sequence. The values in the cell are taken from case study data and can vary numerically depending on assigned values where 0 suggests that no product is allotted at that location.

2.3 Notations In order to develop the mathematical model of the proposed genetic algorithm, the notations used are given in Table 1.

2.4 Evaluation Objective function for the given problem to be maximized for optimization is defined as the sum of total profit realized by the company and customer satisfaction. Fitness value is calculated using the following expression:    Objective function =ω TR − MC − LC − OC − C p + LCO +(1 − ω) [NPS/SP]

(1)

160

A. Kunapareddy and G. Allaka

Table 1 Notations used Notation

Description

TR

Total revenue earned by the manufacturing facility for a particular sample financial year (in $)

Cp

Cost incurred due to the penalty (in $)

LCO

Loss incurred due to cancelled and returned orders (in $)

NPS

Net promoter score; a metric used for customer satisfaction; a score in range of 1–10 is assigned based on the calculations for that particular financial year

SP

Sales projection; estimate of total revenue that the manufacturing facility is expected to earn in future (in $)

TCmp

Total cost of material purchases; cost incurred due to purchase of raw materials required to produce the required products in the manufacturing facility (in $)

Cbi

Cost of beginning inventory; cost of inventory at the beginning of manufacturing period (in $)

Cei

Cost of ending inventory; cost of inventory at the end of manufacturing period (in $)

St

Total sales achieved by the manufacturing facility for a particular sample financial year (in $)

L%

Labour percentage; percentage of labour assigned for completing the total manufacturing activity till delivery of the finished product

LRavg/hr

Average hourly rate of labour; average cost incurred by the manufacturing facility per unit time (hour)

PT

Payroll taxes incurred at manufacturing facility; taxes paid by the manufacturing facility related to labour and related activities for a particular sample financial year (in $)

PS

Production salaries; salaries paid to labour by the manufacturing facility for a particular sample financial year (in $)

FR

Facility rent; facility rent paid by the manufacturing facility including miscellaneous rents (in $)

Crm

Cost of repair and maintenance; cost incurred to the manufacturing facility due to repair and maintenance of the machinery in shop floor and other ancillary machinery (in $)

ED

Equipment depreciation; depreciation expense over a period of time (5 years for given case study); variable depending on the type of equipment, product manufactured and manufacturing facility (in $)

ω

Weight coefficient; can be assumed as per convenience (for given case study, 0.65 for profit and 0.35 for customer satisfaction are assumed)

fv

Fitness values achieved

t

Computing time (in minutes)

P

Production time (in minutes); calculated as a number of products per unit man-hours involved (50 products in given case study)

V

Value of configuration of products; value to different combinations of products selected in a given chromosome

Piks

Selected product i in production line k in manufacturing sequence s (continued)

An Improved Genetic Algorithm for Production Planning …

161

Table 1 (continued) Notation

Description

nmin

Minimum number of required product diversity; minimum number of different products selected in given chromosome required to satisfy knapsack criteria of the problem

vmin

Minimum value of required product configuration; minimum value of different products selected in given chromosome required to satisfy knapsack criteria of the problem

Ci

Capital allotted to the production line i (in $)

Mi

Material allotted to the production line i

Li

Labour allotted to the production line i

Manufacturing cost (MC) is given as n  N    (TCmp + Cbi) − Cei

MC =

(2)

k=1 i=1

Labour cost (LC) is given as LC =

n  N    (St ∗ L%)/LRavg/hr + PT

(3)

k=1 i=1

Overhead cost (OC) is given as OC =

n  N 

[PS + FR + Crm + ED]

(4)

k=1 i=1

Minimize: Capital allotted N 

Ci = C

(5)

Mi = M

(6)

i=1

Material allotted N  i=1

Labour allotted N  i=1

Li = L

(7)

162

A. Kunapareddy and G. Allaka

Overall time N n  

Pik = P

(8)

k=1 i=1

Diversity of configuration of products n≥

n  N s  

Piks ≥ nmin

(9)

Piks ≥ vmin

(10)

j=1 k=1 i=1

Value of configuration of products V ≥

n  N s   j=1 k=1 i=1

2.5 Selection Selection is made using stochastic universal sampling to reduce premature convergence [10]. The optimum population pool is ensured by selecting good chromosomes from an earlier generation as the first generation in the current civilization maintaining the diversity.

2.6 Crossover In crossover, two previously selected parent chromosomes are combined to create two offspring chromosomes. There is crossover in resource allocation and secondly a crossover in manufacturing sequence part of the chromosome [9]. To initialize implementation, primary activities are activated first while the secondary and tertiary activities are deactivated. Correction strategy is used to optimally utilize resources as the length of the given production line is modified depending on the resource constraints. Changes are implemented as per resource availability. After the crossover operation, allocation of resources is amended. Crossover implemented is a hybrid crossover operation [12]. Factors like feasibility of the offspring chromosome, diversity of the selected configuration of products represented by Eq. (9) and value of the selected configuration of products represented by Eq. (10) must be satisfied.

An Improved Genetic Algorithm for Production Planning …

163

2.7 Mutation To create diversity in population that could be unachieved in crossover operation, offspring chromosomes of crossover are subjected to mutation in resource allocation and manufacturing sequence part of the chromosome [9]. Swap mutation is performed where the chromosome is selected randomly and column elements are swapped.

3 Case Study To test the effectiveness and efficiency of the proposed approach, a prototype was developed in Python. A manufacturing facility specializes in producing defence equipment. The facility produces a variety of products P1, P2 … P50. The production hour per week is 175 h. Every week, the facility is allotted 300 kg of material and $430 k of working capital. The penalty incurred due to delay is 7.5% of the manufacturing price of the product per day of delay. Any product made a week after the deadline is rejected by customers and will be returned. Any product changeover takes 1 h in any given production.

4 Experimental Results and Discussion The case study was solved using the proposed GA. Selection of parameters of proposed GA was done using Taguchi method-based experiment design using orthogonal arrays to categorize the parameters influencing the process and levels at which the parameters can be varied. The Taguchi orthogonal array design (L27-37) with three levels and seven factors was done using DOE (the experimental design package for Python). The experimental layout is given in Table 2. To maintain consistency in the solution quality, the proposed GA in each experiment was repeated for five times and was run for same computing time, i.e. 60 min. The parameters set for the proposed genetic algorithm are shown in Table 3. The results obtained in the ANOVA show the effect of seven parameters (population size, crossover rate in resource allocation, crossover rate in manufacturing sequence, mutation rate in resource allocation, mutation rate in manufacturing sequence, number of generations in civilization and number of good chromosomes for succeeding civilizations) on the performance of the proposed GA as shown in Table 4. Taguchi analysis results for obtaining optimum combination of seven parameters are shown in Fig. 1. Normality of the fitness values is determined using Shapiro–Wilk test [11] and Kolmogorov–Smirnov test. The parameter set for Shapiro–Wilk test is α = 0.05, and outliers are included. Critical value accepted range is observed to be 95% for all the optimization solvers. The results are shown in Tables 5 and 6 respectively.

E

F

G

2

2

2

2

2

2

2

3

3

13

14

15

16

17

18

19

20

1

8

12

1

7

2

1

6

11

1

5

1

1

4

2

1

3

10

1

9

1

2

1

1

3

3

3

2

2

2

1

1

1

3

3

3

2

2

2

1

1

1

3

3

1

1

1

3

3

3

2

2

2

3

3

3

2

2

2

1

1

1

2

2

2

2

2

1

1

1

3

3

3

3

3

3

2

2

2

1

1

1

2

1

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

1

3

1

3

2

1

3

2

1

3

2

3

2

1

3

2

1

3

2

1

3

2

2

1

3

2

1

3

2

1

3

3

2

1

3

2

1

3

2

1

15,450.8

17,175.1

17,189.7

17,467.2

16,369.8

18,624.3

18,394.4

17,557.5

16,333.3

16,272.9

16,213.8

17,371.2

17,821.5

16,172.5

18,277.3

18,842.9

16,179.5

18,283.8

18,736.9

17,378.3

RUN 1

D

Fitness values

C

A

B

Experimental factors

1

S. no.

Table 2 Taguchi experimental design

17,252.3

16,480.7

16,996.3

17,198.6

17,876.9

18,512.1

17,404.2

18,999.7

18,105.3

17,402.8

18,494.5

16,936.7

16,017.5

17,531.2

16,857.5

16,218.6

17,733.9

17,615.1

16,327.8

17,685.6

RUN 2

17,147.4

17,020.3

16,069.2

17,531.8

16,295.2

16,148.4

17,792.1

18,260.9

18,766.3

16,525.3

17,806.3

17,464.6

16,721.2

17,678.1

16,984.5

16,120.3

17,953.9

18,094.7

17,334.1

18,021.5

RUN 3

16,124.2

17,006.8

17,293.6

18,865.9

16,522.7

16,389.1

17,293.2

16,267.4

17,536.1

17,975.3

16,926.3

16,256.9

17,321.8

18,472.5

18,446.4

17,310.1

16,488.9

16,176.3

18,394.2

17,038.7

RUN 4

(continued)

17,098.4

17,140.1

18,669.9

16,586.6

17,889.2

18,227.8

17,934.7

16,490.4

16,286.6

17,501.2

18,457.9

16,277.1

17,492.7

17,341.2

16,113.4

16,456.2

17,715.5

18,327.9

16,320.4

16,353.1

RUN 5

164 A. Kunapareddy and G. Allaka

E

F

G

3

3

3

3

3

3

3

22

23

24

25

26

27

3

3

3

2

2

2

1

2

2

2

1

1

1

3

1

1

1

3

3

3

2

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

3

2

1

17,194.2

17,369.2

17,160.5

16,010.1

18,385.7

17,534.2

18,011.0

RUN 1

D

Fitness values

C

A

B

Experimental factors

21

S. no.

Table 2 (continued)

18,792.3

16,927.9

17,459.7

16,025.2

16,985.3

17,703.1

16,160.8

RUN 2

16,290.2

17,434.7

17,023.5

16,684.6

18,864.1

17,971.1

18,992.9

RUN 3

17,458.6

17,350.1

16,178.8

16,418.5

17,344.5

16,915.6

17,038.4

RUN 4

16,751.4

18,125.1

16,840.5

17,725.7

16,416.5

17,295.2

16,173.8

RUN 5

An Improved Genetic Algorithm for Production Planning … 165

166

A. Kunapareddy and G. Allaka

Table 3 Parameters and their levels Parameters

Levels

Code

GA parameter

1

A

Population

40

80

120

B

Crossover rate (resource allocation)

20

40

70

C

Crossover rate (manufacturing sequence)

20

40

70

D

Mutation (resource allocation)

15

30

45

E

Mutation (manufacturing sequence)

15

30

45

F

Number of generations in civilization

600

1000

1400

G

Number of good chromosomes for subsequent civilizations

10

25

40

2

3

Table 4 ANOVA for seven parameters DF

Adj SS

Adj MS

F-value

A

2

1,451,827

725,913

1.46

B

2

644,512

322,256

0.65

C

2

381,300

190,650

0.38

D

2

3,507,125

1,753,563

3.54

E

2

1,768,428

884,214

1.78

F

2

2,926,203

1,463,102

2.95

G

2

713,188

356,594

0.72

Error

12

5,948,060

495,672

Total

26

17,175,889

With this solution, the achieved fitness value is 18,992.9. The proposed genetic algorithm provides 31.50, 60.50, 68.03, 86.84, 22.98 and 77.01% better solution compared to pattern search solver [13], simulated annealing solver [14], tabu search [15], stochastic descent search [16], ant colony optimization [16, 17] and traditional GA [18], respectively. The computation time of the proposed genetic algorithm is observed to be 36.51, 88.12, 55.16, 86.35, 30.28 and 44.03% smaller than to pattern search solver, simulated annealing solver, tabu search, stochastic descent search, ant colony optimization and traditional GA, respectively. The proposed GA provides 6.10% better solution than the algorithm proposed by Dao et al. [9]. Performance comparison between pattern search, simulated annealing, traditional genetic algorithm, tabu search, stochastic gradient descent, ant colony optimization and proposed genetic algorithm is shown in Fig. 2, Tables 7 and 8. Figure 3 illustrates the evolution of solution quality of the proposed genetic algorithm where 75% crossover probability and 15% mutation probability demonstrate the best impact on fitness.

An Improved Genetic Algorithm for Production Planning …

167

Fig. 1 Taguchi analysis results for fitness values

Results obtained by experimental data indicate that the proposed improved GA yields better results than the traditional GA and outperforms it in terms of fitness values and computing time.

168

A. Kunapareddy and G. Allaka

Table 5 Shapiro–Wilk test Pattern search

Statistics

Degree of freedom

p value

0.9447

100

0.6485

Simulated annealing

0.9210

100

0.3799

Traditional genetic algorithm

0.9090

100

0.2813

Tabu search

0.9283

100

0.4524

Stochastic gradient descent

0.9363

100

0.5424

Ant colony optimization

0.9247

100

0.4148

Proposed genetic algorithm

0.9440

100

0.6391

Table 6 Kolmogorov–Smirnov test Mean

Median

Standard deviation

Skewness

Kurtosis

Pattern search

13,779.50

13,823.10

598.337393

−0.676991

−0.083685

Simulated annealing

11,290.25

11,414.05

402.636283

−0.344172

−1.103497

Traditional genetic algorithm

10,237.18

10,288.30

509.864986

−0.256847

−1.535443

Tabu search

10,783.79

10,610.50

530.209999

0.726648

−0.276948

9698.36

9756.95

536.330653

Stochastic gradient descent

−0.52815

−0.850898

Ant colony optimization

14,734.73

14,553.8

606.159374

0.954443

0.399776

Proposed genetic algorithm

18,120.93

18,029.9

625.639396

−0.036728

−1.329572

Fig. 2 Performance comparison

An Improved Genetic Algorithm for Production Planning …

169

Table 7 Performance comparison between pattern search, simulated annealing, traditional genetic algorithm and proposed genetic algorithm No. of trial

Pattern search

Simulated annealing

Genetic algorithm

Proposed genetic algorithm

fv

t

fv

t

fv

t

fv

t

1

13,716.9

104.6

11,792.8

560.3

10,809.8

119.4

17,821.5

64.1

2

13,144.9

107.1

11,545.6

546.9

9536.5

121.7

18,512.1

60.4

3

14,491.2

103.5

10,979.3

574.1

10,312.2

115.1

18,992.9

63.2

4

12,626.4

102.1

11,500.9

541.3

10,875.5

119.5

17,725.7

72.8

5

13,806.7

110.8

10,935.3

555.8

10,264.4

122.7

18,125.1

66.9

6

14,086.3

103.2

11,375.8

580.4

9694.5

117.7

18,766.3

71.1

7

13,355.7

104.9

10,928.6

562.9

10,679.9

115.3

17,344.5

61.5

8

14,460.2

104.7

11,452.3

571.1

10,054.1

119.2

17,194.2

69.5

9

13,839.5

105.9

10,607.7

569.1

10,586.6

116.9

18,792.3

75.1

10

14,267.2

103.7

11,784.2

554.6

9558.3

124.1

17,934.7

62.3

Table 8 Performance comparison between tabu search, stochastic gradient descent, ant colony optimization and proposed genetic algorithm No. of trial

Tabu search

Stochastic gradient descent

Ant colony optimization

fv

t

fv

t

fv

1

10,159.3

156.6

10,346.1

517.4

15,216.0

2

11,805.7

139.4

9731.7

481.1

3

10,451.3

158.7

10,122.9

544.5

4

10,224.1

130.5

9782.2

5

11,312.9

147.5

6

10,657.8

152.8

7

11,146.1

8 9 10

Proposed genetic algorithm

t

fv

t

92.9

17,821.5

64.1

14,031.2

96.6

18,512.1

60.4

14,152.8

91.1

18,992.9

63.2

439.4

15,973.5

100.0

17,725.7

72.8

8788.2

556.8

15,313.8

94.1

18,125.1

66.9

9530.5

477.9

14,827.9

99.7

18,766.3

71.1

138.1

10,076.8

412.3

14,626.7

98.6

17,344.5

61.5

11,071.0

163.9

10,277.3

513.5

14,480.9

98.1

17,194.2

69.5

10,563.2

144.0

8970.6

482.6

14,293.3

91.9

18,792.3

75.1

10,446.5

156.0

9357.3

463.1

14,431.2

93.6

17,934.7

62.3

5 Conclusion An alternative approach to integrate production planning and scheduling using an improved GA with variable chromosome length and parameters tuned by Taguchi experiment design is presented in this paper. The proposed improved GA searches for global optimal solution more effectively in comparison with other algorithms like pattern search, simulated annealing, traditional genetic algorithm, tabu search, stochastic

170

A. Kunapareddy and G. Allaka

Fig. 3 Evolutionary trajectory of solution quality

gradient descent, ant colony optimization. Further work will look into adding more constraints, applying techniques to shorten computing time and comparison with other heuristic-based algorithms.

References 1. Khoshnevis, B., Chen, Q.M.: Integration of process planning and scheduling functions. J. Intell. Manuf. 2(3), 165–175 (1991) 2. Morad, N., Zalzala, A.M.S.: Genetic algorithms in integrated process planning and scheduling. J. Intell. Manuf. 10(2), 169–179 (1999) 3. Maravelias, C.T., Sung, C.: Integration of production planning and scheduling: overview, challenges and opportunities. Comput. Chem. Eng. 33(12), 1919–1930 (2009) 4. Li, X., Gao, L., Zhang, G., Zhang, C., Shao, X.: A genetic algorithm for integration of process planning and scheduling problem. In: International Conference on Intelligent Robotics and Applications, Springer, Berlin, Heidelberg, 495–502 Oct 2008 5. Moon, C., Kim, J., Hur, S.: Integrated process planning and scheduling with minimizing total tardiness in multi-plants supply chain. Comput. Ind. Eng. 43(1–2), 331–349 (2002) 6. Kim, Y.K., Park, K., Ko, J.: A symbiotic evolutionary algorithm for the integration of process planning and job shop scheduling. Comput. Oper. Res. 30(8), 1151–1171 (2003) 7. Liu, Y., Fang, Y.: Boost the Integration of planning and scheduling: a heuristics approach. Procedia Eng. 29, 3348–3352 (2012) 8. Zhang, H.C., Merchant, M.E.: IPPM–a prototype to integrate process planning and job shop scheduling functions. CIRP Ann. 42(1), 513–518 (1993) 9. Dao, S.D., Abhary, K., Marian, R.: An improved genetic algorithm for multidimensional optimization of precedence-constrained production planning and scheduling. J. Ind. Eng. Int. 13(2), 143–159 (2017)

An Improved Genetic Algorithm for Production Planning …

171

10. Hussain, A., Muhammad, Y.S., Sajid, M.N.: Performance evaluation of best-worst selection criteria for genetic algorithm. Math. Comput. Sci. 2(6), 89–97 (2017) 11. Shapiro, S.S., Wilk, M.B.: An analysis of variance test for normality (complete samples). Biometrika 52(3/4), 591–611 (1965) 12. Rahman, R.A., Ramli, R.: Average concept of crossover operator in real coded genetic algorithm. Int. Proc. Econ. Dev. Res. 63, 73 (2013) 13. Goldberg, D.E.: Genetic algorithms in search, optimization & machine learning, Addison Wesley, 1997 14. Li, W.D., Mcmahon, C.A.: A simulated annealing-based optimization approach for integrated process planning and scheduling. Int. J. Comput. Integr. Manuf. 20(01), 80–95 (2006). Taylor Francis 15. Ponnambalam, S., Aravindan, P., Rajesh, S.: Int J Adv Manuf. Technol. 16, 765 (2000) 16. Meuleau, N., Dorigo, M.: Ant colony optimization and stochastic gradient descent. Artif. Life 8(2), 103–121 (2002) 17. Liu, X., Ni, Z., Qiu, X.: Int J Adv Manuf. Technol. 84, 393 (2016). https://doi.org/10.1007/ s00170-015-8145-4 18. Man, K.F., Tang, K.S., Kwong, S., Ip, W.H.: Genetic algorithm to production planning and scheduling problems for manufacturing systems. Prod. Plan. Control 11(5), 443–458 (2000)

Automatic Calibration for Residential Water Meters by Using Artificial Vision Edwin Pruna, Carlos Bustamante, Miguel Escudero, Santiago Mullo, Ivón Escobar and José Bucheli

Abstract The present work addresses the problem of automated calibration system for residential water meters, by using artificial vision. The project consists of a closed water flow circuit powered by a low-power pump; the data from water meter is taken by the computer using a USB camera; a calculation is made based on the time it takes to fully rotate the smaller-scale needle of the meter to determine the water flow in real time. At the same time, the actual flow data of the reference standard element, which is a rotameter, is obtained by a second camera. These values are compared to calculate an error that determines the adjustment action on the water meter.

1 Introduction The applications of machine vision in industry have increased in recent years and are submerged in different fields such as in the agro-industry for the inspection of the condition and quality of fruits and vegetables [1], in robotics to recreate a map with techniques such as stereoscopic vision [2], while in medicine it is used for the restoration of vision by electrical stimulation of the visual system [3]. In E. Pruna (B) · C. Bustamante · M. Escudero · S. Mullo · I. Escobar · J. Bucheli Universidad de Las Fuerzas Armadas ESPE, Sangolqui, Ecuador e-mail: [email protected] C. Bustamante e-mail: [email protected] M. Escudero e-mail: [email protected] S. Mullo e-mail: [email protected] I. Escobar e-mail: [email protected] J. Bucheli e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_16

173

174

E. Pruna et al.

addition, [4] proposes a reading recognition method for the automatic recognition of analog measuring instruments. Analog measuring instruments are widely used in control and measurement systems due to their low cost and simple structure. However, measurement precision may be limited by the human visual capacity, where factors such as the visual fatigue are latent. In this research, the measurement of analogical instruments using artificial vision is developed. Artificial vision is robust in industry because many processes require high precision and good real time, so some devices have been developed to cover the technological needs of artificial vision such as the Kinect which has advantages in the acquisition of 3D body movement through an RGB camera and a depth sensor [5]. However, many problems are present in the image processing due to the geometric changes caused by the shape of the object, the relative positioning of object and camera, perspective imaging and optical lens defects [6]. Figure 1 shows the essential components of a complete system of artificial vision. Large quantities of residential water meters may be outside the calibration range, causing problems for the water users or water companies due to the lack of accurate record of water consumption records. Despite the introduction of smart technology, it solves the main challenges of the growing demand of the population, such as smart water meters that provide consumption data with a good resolution, and in real time [7, 8], many of them are not accessible to everyone, due to different constraints (economic and political), so they are not the best way to manage and control water consumption due to their high cost. Nowadays, there are researches that resolve the economic limitations [9]. Therefore, a calibration automatic system to measure water at low cost has been developed based on an algorithm implemented in LabVIEW for the processing of

Fig. 1 Diagram of an artificial vision system

Automatic Calibration for Residential Water Meters …

175

images through the extraction of information applying various techniques (delimiting the region of interest (ROI), edge detection, color spectrum analysis, etc.) used in metrological applications [10] from low-cost USB cameras and control of servomechanisms.

2 Equipment Description

ROTAMETER WATER METER

VALVE

TANK

CENTRIFUGAL PUMP

CAMERA 1

Fig. 2 Schematic diagram of the automatic calibration system

CAMERA 2 FI

SERVOMECHANISM

The prototype of a calibration system is implemented with piping polymerization of vinyl chloride (PVC) of ¾ inch and water as a recurring liquid. Figure 2 shows the schematic diagram of the automatic calibration system consisting of a 10-gallon reservoir tank, which in turn is connected to a centrifugal pump to drive the liquid, and then the pump is connected to a series gate-type valve to regulate the flow, coupled to a servomechanism. The measuring instruments are also connected: a residential water meter and rotameter with a range of 4–40 L per minute LPM shown in Fig. 3. The data is obtained using two low-cost USB cameras connected to the computer. Camera 1 gets the data from the water meter needle, while camera 2 gets the data from the rotameter. Both cameras transmit information to the LabVIEW software that performs the image processing algorithm and thus calculates the control signal of the servomechanisms using an Arduino card. The lack of calibration of residential water meters due to various factors such as lack of maintenance and operating conditions causes the water meter to deteriorate and the reading to be inaccurate, as well an additional cost and human effort, so regular calibration is essential. Figure 4 shows the flow diagram of the entire calibration process implemented.

ARDUINO

176

E. Pruna et al.

Fig. 3 Prototype of the automatic calibration system implemented

START IMAGE ACQUISITION FROM THE CAMERAS EXTRACTION OF DATA FROM THE ROI POSITION ACQUISITION ROTAMETER AND NEEDLE ERROR = NEEDLE VALUE- ROTAMETER VALUE

NO

ERROR≠0 YES POSITIVE OR NEGATIVE ADJUSTMENT

END Fig. 4 Flow diagram of the automatic calibration system

Automatic Calibration for Residential Water Meters …

177

The calibration is performed with a measurement error generated by obtaining the value of the flow rate in the water meter and its comparison with the actual value measured on the standard instrument (rotameter). The error generates a correction signal that activates the servomechanism and makes the positive or negative adjustment by an external regulation screw of the residential meter, where the compensatory flow can be regulated (compare Fig. 3) and can eliminate the error. To obtain an automatic calibration process, one HMI is developed in LabVIEW and a second servomechanism is mechanically coupled to the control valve.

3 Instrument Measurement The image processing algorithm in LabVIEW starts with the reduction of processed pixels to increase the processing speed and obtain a higher response speed. Therefore, region of interest (ROI) is defined with the help of Vision Assistant from LabVIEW shown in Fig. 5. This region detects changes generated by the movements in the rotameter (variation in the y-axis) and the needle of water meter (variation in the color spectrum in the region). Non-lineal response of the rotameter leads to a curve fitting, taking 14 different points from the minimum to the maximum value, obtaining Eq. (1) of second degree

Fig. 5 Parameter configuration of rotameter and water meter images in LabVIEW

178

E. Pruna et al.

with an error of 1.94%. The adjustment made is shown in Fig. 6. p = 0.0000905 · y 2 − 0.1417 · y + 49.36

(1)

where y represents the values of the pixel positions on the y-axis and P is the value of liter per minute LPM. Figure 7 shows the measurement results of the standard element (rotameter). The image processing of the water meter starts with the detection of the circular edge around the smallest scale needle (compare the red circle in Fig. 5), and then

Fig. 6 Convertion from pixel to liters per minute (LPM)

Fig. 7 Flow measurement of rotameter with camera 2

Automatic Calibration for Residential Water Meters …

179

the rectangular ROI is determined around the number zero where the color spectrum changes in this region are detected. Each change will mean that the needle has completed one cycle and one liter of water has just flowed. To calculate the amount of flow in LPM is necessary to obtain the time it takes for the needle to complete a cycle and calculate the inverse given by Eq. (2). LPS =

1 tk − tk−1

(2)

where LPS is the value of liters per second, and tk and tk−1 represent the actual time and last, respectively, for which the needle has completed the cycle given in seconds. This data later is divided by 60 to obtain the desired units.

4 Results The color changes since the reflected electromagnetic radiation recorder in the image is affected by the sensitivity of the light recording medium and the transmission media. Therefore, calibration tests were conducted in an enclosed space under unchanging lighting conditions to avoid problems in data acquisition. The tests consist of varying flow of water through the HMI in LabVIEW by setting specific values. To obtain the percentage error, the values taken from water meter are compared with the reference values of the rotameter at nine different up and down points detailed in Table 1, resulting in an average error of 5.55%. Each value obtained after the calibration and adjustment has minimum percentage errors of approximately 1%. Moreover, it is necessary to take into account that the rotameter has an accuracy error of 1% given by the manufacturer. However, these errors are imperceptible and acceptable for system performance. In addition, processing speed is affected by computer capacity because LabVIEW’s Vision and Motion toolkits consume computing resources. Table 1 Calibration in 9 points

Real value (LPM)

Measured value (LPM)

Error (%)

12

14

6.250

17

19

6.250

22

23

3.125

27

29

6.250

32

33

3.125

27

29

6.250

22

24

6.250

17

19

6.250

12

14

6.250

180

E. Pruna et al.

5 Conclusions In this work, a prototype developed allows the calibration of the residential water meter, using a servomechanism coupled to the regulating screw to adjust the compensatory flow rate. The system has two cameras that allow the flow rate to be measured in the rotameter (standard instrument) and in the water meter using an image processing algorithm implemented in LabVIEW. The system of measuring analog instruments by artificial vision opens up the possibility of implementing this system in industries that have analog instruments, without having to alter or change the equipment. However, the accuracy of the devices and computer capacity for image processing define the limitations of the system. For future works, software with higher image processing speed and the improvement of the algorithm will be needed to obtain a better real time and reduce errors caused by changes in lighting conditions. The adjustment is made based on the error generated between the values of the water meter and the reference in the rotameter, causing the control actions executed by an Arduino card connected to the servomechanism. Finally, the tests were carried out at different points obtaining excellent results with errors of less than 1%, therefore obtaining a good calibration.

References 1. Cubero, S., Aleixos, N., Molto, E., Gomez-Sanchis, J., Blasco, J.: Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables. Food Bioproc. Techol. 4(4), 487–504 (2011) 2. Meier, L., Tanskanen, P., Heng, L., Lee, G.H., Fraundorfer, F., Pollefeys, M.: PIXHAWK: a micro aerial vehicle design for autonomous flight using onboard computer vision. Auton. Robot. 33(1–2), 21–39 (2012) 3. Fernandes, R.A.B., Diniz, B., Ribeiro, R., Humayun, M.: Artificial vision through neuronal stimulation. Neurosci. Lett. 519, 122–128 (2012) 4. Jiale, H., En, L., Bingjie, T., Ming, L.: Reading recognition method of analog measuring instruments based on improved hough transform. In: IEEE 2011 10th International Conference on Electronic Measurement & Instruments, Chengdu, 337–340 2011 5. Han, J., Shao, L., Xu, D., Shotton, J.: Enhanced computer vision with microsoft Kinect sensor: a review. In: IEEE Transactions on Cybernetics, vol. 43, no. 5, 1318–1334 Oct 2013 6. Luhmann, T., Robson, K., Kyle, S., Boehm J.: Close-range photogrammetry and 3D imgaging. Walter de Gruyter Gmbll Berlin (2014) 7. Cominola, A., Giuliani, M., Piga, D., Castelletti, A., Rizzoli, A.E.: Benefits and challenges of using smart meters for advancing residential water demand modeling and management: a review. Environ. Model. Softw. 72, 198–214, (2015) ISSN 1364-8152 8. Sonderlund, A.L., Smith, J.R., Hutton, C., Kapelan, Z.: Using smart meters for household water consumption feedback: knowns and unknowns. Procedia Eng. 89, 990–997 (2014) 9. Larson, E., Froehlich, J., Campbell, T., Haggerty, C., Atlas, L., Fogarty, J., Patel, S.N.: Disaggregated water sensing from a single, pressure-based sensor: an extended analysis of HydroSense using staged experiments. Pervasive Mob. Comput. 8(1), 82–102 (2012) 10. Lim, T.Y., Ratnam, M.M.: Edge detection and measurement of nose radii of cutting tool inserts from scanned 2-D images. Opt. Lasers Eng. 50(11), 1628–1642 (2012)

Hardware-in-the-Loop of a Flow Plant Embedded in FPGA, for Process Control Edwin Pruna, Icler Jimenez and Ivón Escobar

Abstract It presents the simulation through the Hardware-in-the-loop (HIL) technique of a flow plant. The mentioned system is designed in the LabVIEW software and implemented in an FPGA. As a result of the system tests performed in manual mode, an absolute error of 0.02 is obtained in the simulated instruments. In addition, two controllers are designed (continuous and discrete), and the results indicate that the system works in real time and does not generate disturbances in response to the implemented controls.

1 Introduction An industrial process is a set of operations to: obtain, transform or transport primary products [1]; the control of industrial processes allows to keep a desired amount in the dynamic variables such as flow, pressure and temperature. This increases the production, productivity and efficiency [2]. To learn about process control, there are teaching systems (industrial plants) that allow students to acquire skills and abilities when they develop laboratory practices [3], these didactic systems for the control of the variables: pressure, level, temperature, flow, etc. They allow to set up a system of closed-loop control, integrating the three fundamental components such as controller, sensor and the actuator. The tasks to be developed to perform the closed-loop control of the aforementioned processes are obtaining the mathematical model of the process, optimal controller design (analytical method) and implementation and validation of the control algorithm in the plant. The didactic systems of commercial brands used for the learning of process control are very expensive because of the industrial components that comprise it (sensors, E. Pruna (B) · I. Jimenez · I. Escobar Universidad de Las Fuerzas Armadas ESPE, Sangolquí, Ecuador e-mail: [email protected] I. Escobar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_17

181

182

E. Pruna et al.

transmitters, actuators, etc.), this price makes it difficult for some institutions to acquire these systems for teaching of process control. Nowadays, there is a lower cost alternative known as “Hardware-in-the-loop” that allows to simulate plants or control systems in real time, together with real elements of sensors or actuators [4–7]. Several scientific works have been developed in this area of research, in [8] explain how HIL can be used effectively to reinforce the theoretical concepts in control systems, in [9]. It is developed a system of simulation of a photovoltaic plant for the analysis of intelligent networks, using Simulink and Arduino. It is explained that the best result is obtained from the implementation of the entire system in a simulator in real time. In [10] develop the simulation using the technique of Hardware-in-the-loop of a Buck Converter. Good results are obtained from the simulation, and the data are compared for verifying that the system works efficiently when is implemented in an FPGA; in [11], a control system for a plant embedded in FPGA is developed. In this research, they implement a first-order temperature system in an FPGA. At the same time, they develop the PID controller programming in the LabVIEW software, and the communication between the two applications is made using the RS-232 serial interface. It is observed that the limitation that exists is the resolution of sending and the receiving information to 8 bits. The obtained results indicate the correct functioning of the implemented system. In this context, a simulation of a plant is presented for the flow control. In the design of the visualisation interface, the LabVIEW software is used, and the simulation of the plant is embedded in a card myRIO that contains a FPGA. This system includes the simulation of a sensor (0–5 v) and an actuator (0–5 v) for the interaction with the different controllers on the market. The structure of the system is presented in Fig. 1. “HIL” FLOW CONTROL PLANT HOST COMPUTER HOST VI

|

CONTROLLER SP

+

error (m)

X

-

CV Actuator

PID Variation % PV Caudal (GPM)

Fig. 1 Diagram of the proposed system

myRIO FPGA VI Analog input Analog output

Plant G( s ) =

1.1981 – 0.3883s e 1 + 0.2601s

Flow (GPM)

Hardware-in-the-Loop of a Flow Plant Embedded …

183

0-5V

FIC

FI

01

0-5V

SC 01

V-3

FE 01 ///

M V-2 WATER TANK

PALLETS

V-1 PUMP

Fig. 2 Piping and instrumentation diagram of the plant flow to be implemented

2 Development of the HIL 2.1 The Flow Plant Design The design of the flow plant is considered to be a didactic system as indicated in [12]. The process is a flow feedback system, which has a detailed flow sensor (FE). It sends a 0–5 Vdc voltage signal to a flow controller (FIC). In this controller, the PID control algorithm is performed. Finally, the control signal CV (0–5 Vdc) is sent to the final control element (SC). In Fig. 2, the design of the pipe and instrumentation diagram (P and ID) is presented. It is used as a reference for the development of the virtual environment.

2.2 Implementation of the Flow Plant The embedded system (myRIO) is used. It is composed of an FPGA. The design of the visualisation interface is developed in LabVIEW software. The LabVIEW FPGA module converts the designed programming in the FPGA VI to the FPGA hardware, through the compilation and generation of files for FPGA programming (bit file). The developed interface HIL consists of a flat sequence with three stages: First stage (A). It allows initialising the variables of the myRIO. Second stage (B). The execution of the program is developed in parallel due to the characteristics of the FPGA. It is made up of two While loop:

184

E. Pruna et al.

Fig. 3 Programming “HIL” of the flow plant, a front panel, b block diagram

1. While Loop #1. It allows the simulation of the transmitter and the actuator, generating and receiving electrical signals (0–5 Vdc), prior to the configuration of analogue inputs and outputs of the myRIO card. 2. While Loop #2. It allows the simulation of the flow plant. The mathematical model implemented corresponds to a first-order model with dead time, as shown below:

G(S) =

1.1211 ∗ e−0.7023s 1 + 1.12168s

(1)

Third stage (C). Close the connection with myRIO. The programming is presented in Fig. 3.

3 Controller Design To check the operation of the HIL system in closed loop, two types of controllers were designed and implemented: (i) a continuous controller is implemented in a PLC and (ii) a discrete controller is implemented in LabVIEW.

3.1 Design of the Continuous PID Controller For the design of the continuous controller, the analytical method is used. The mathematical model of the HIL system of the flow plant is obtained. Km ∗ e−τm s 1 + Tm s

(2)

1.1205 ∗ e−0.7557s 1 + 1.1648s

(3)

G(S) = G(S) =

Hardware-in-the-Loop of a Flow Plant Embedded …

185

The calculation of PID tuning constants is performed by LAMBDA method. K =

1 τ2m + Tm . ; where Tcl = 3Tm K m τ2m + Tcl τm Ti = Tm + 2 Td =

Tm τm τm + 2Tm

(4) (5) (6)

The PID obtained constants are K = 0.3555, Ti = 1.5427 and Td = 0.1. The PID continuous control algorithm designed is  PID(s) = 0.3555 1 +

1 + 0.1s 1.5427s

 (7)

3.2 Design of the Discrete PID Controller The action of the implemented digital controller is given by the following expression: PID(kh) = P(kh) + I (kh) + D(kh) PID(kh) = K p E(kh) + K i hE(kh) + I ((k − 1)h) +

(8)

Kd [E(kh) − E((k − 1)h) h (9)

To obtain the tuning constants and the sampling time, the auto-tuning tool is used. The constants obtained are k p = 0.6, K i = 0.5, K d = 0.26 and dt(s) = 0.1.

4 Tests and Results 4.1 Tests The procedure used for the tests of the system consists of performing a manual and automatic control, using a computer that has LabVIEW software installed and a Siemens S7-1200 PLC. There are three types of tests, which allow to verify the operation of the Hardware-in-the-Loop system, and this tests are described below:

186

E. Pruna et al.

1. Transmitter and actuator simulation offlow. Tests are done in open loop with random samples of variation of the transmitter and the actuator to verify the error of the implemented system. 2. Automatic control of the flow plant using continuous control. Through the programmable logic controller (PLC), the continuous PID control, designed in paragraph (3.1), is implemented. 3. Automatic control of the plant using discreet control flow. Through LabVIEW and data acquisition (DAQ 6008) produced by National Instruments, the discrete PID control, designed in paragraph (3.2), is implemented.

4.2 Results The obtained results from the HIL system are presented: 1. Test performed in open loop. The obtained measurements from the simulated instruments are presented in Tables 1 and 2. 2. Operation of the system Hardware-in-the-loop of the flow plant. The interface of the flow plant in operation is presented in Fig. 4. The animation in the pipes provides a realistic environment for the flow of water through the pipes, and as well as the amounts of the transmitter and actuator are displayed in real time. 3. Response of the implemented continuous PID control for the control of the HIL system. Table 1 Measurements of the simulated flow transmitter, determination of the error Measurements

Flow sensor HIL (GPM)

Sensor voltage (v)

Measured voltage(v)

Error

1 2

0,000

0,000

0,000

0

10,000

0,500

0,501

002

3

20,000

1,000

1,000

0

4

30,000

1,500

1,501

002

5

40,000

2,000

2,000

0

6

50,000

2,500

2,500

0

7

60,000

3,000

3,000

0

8

70,000

3,500

3,500

0

9

80,000

4,000

4,000

0

10

90,000

4,500

4,500

0

11

100,000

5,000

4,990

−0,2

Absolute error

0,0218

Hardware-in-the-Loop of a Flow Plant Embedded …

187

Table 2 Verification of the actuator working, determination of the error Measurements

Input voltage (V)

1

0,000

0

0,098

0098

2

0,500

10

9,912

−0088

3

1,000

20

19,992

−0008

4

1,500

30

29,985

−0015

5

2,000

40

39,997

−0003

6

2,500

50

49,997

−0003

7

3,000

60

59,987

−0013

8

3,500

70

70,020

0020

9

4,000

80

80,029

0029

10

4,500

90

90,015

0015

11

5,000

100

99,976

−0024

Absolute error

% of speed variation (reference)

% of speed variation (obtained)

Error

0,0287

Fig. 4 HIL system in operation

Figure 5 shows how the implemented continued PID controller in the PLC responds to a consignment signal and how the process variable stabilises at the 14 s and not presents overshoot.

Fig. 5 Response of the continuous PID control implemented in PLC for the control of “HIL” flow plant

188

E. Pruna et al.

Fig. 6 Response of the discreet control implemented in LabVIEW, for the control of “HIL” flow plant

4. Response of the discrete PID control implemented in LabVIEW software. Figure 6 shows how the discrete PID controller implemented in LabVIEW software responds to consignment signal and how the process variable stabilises at the 9 s and does not present overshoots.

5 Conclusions The implemented HIL system provides performance animation similar to a flow plant. Its response is in real time with two types of implemented controllers (continuous controller and discrete controller). The HIL system is a low-cost proposal for teaching process control. It allows students to design continuous and discrete controllers. Finally, as future work, a Hardware-in-the-loop system could be developed for advanced and multivariable control.

References 1. Peixoto, D.C., Resende, R.F., Pádua, C.I.: An educational simulation model derived from academic and industrial experiences. In: 2013 IEEE Frontiers in Education Conference (FIE), Oklahoma City. OK, 691–697 2013 2. Chao, Z., et al.: Automatic control process analysis of gas pressure in electrostatic discharge measurement system. In: 2015 7th Asia-Pacific Conference on Environmental Electromagnetics (CEEM). Hangzhou, 202–205 2015 3. Pruna, E., Chang, O., Jimenez, D., Perez, A., Avila, G., Escobar, I., Constante, P., Gordon, A.: Building a training module for modern control. In: 2015 CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON) 2015 4. Krishnan, B., Pillai, A.: Digital sensor simulation frame work for hardware-in-the-loop testing. In: 2017 International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT), Kannur, 813–817 2017. https://doi.org/10.1109/icicict1.2017. 8342669

Hardware-in-the-Loop of a Flow Plant Embedded …

189

5. Daffaie, H., Baniyounis, M., Tutunji, T., Lohöfener, M.: Temperature control of a heat sink based on hardware in the loop. In: 2018 15th International Multi-Conference on Systems. Signals & Devices (SSD), Hammamet, 366–370 2018 6. Bertoletti, L., Ragaini, E., Liu, J.: Hardware-in-the-loop simulation for testing low voltage circuit breakers selectivity. In: 2017 IEEE International Conference on Environment and Electrical Engineering and 2017 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I & CPS Europe), Milan, 1–5 2017 7. Kumarin, A., Kuznetsov, A., Makaryants, G.: Hardware-in-the-loop neuro-based simulation for testing gas turbine engine control system. In: 2018 Global Fluid Power Society Ph.D. Symposium (GFPS), Samara, 1–5 2018 8. Zhou, Y., Qi, B., Huang, S., Jia, Z.: Fuzzy PID Controller for FOPDT system based on a hardware-in-the-loop simulation. In: 2018 37th Chinese Control Conference (CCC), Wuhan, 3382–3387 2018. https://doi.org/10.23919/chicc.2018.8482632 9. Huo, Y., Gruosso, G., Piegari, L.: Power hardware in the loop simulator of photovoltaic plant for smart grid interation analysis. In: 2017 IEEE International Conference on Environment and Electrical Engineering and 2017 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I & CPS Europe), Milan, 1–5 2017. https://doi.org/10.1109/eeeic.2017.7977629 10. Casellas, F, et al.: Simulaciónmediante “hardware in the loop” de un convertidor Buck. A: Annual Seminar on Automation, Industrial Electronics and Instrumentation. In: Proceedings of the XXI Annual Seminar on Automation, Industrial Electronics and Instrumentation. Universitat Rovira i Virgili, Tangier, 1–5 2014 11. Caldas, O., et al.: Sistema de control de una planta embebida en fpga empleando hardware-inthe-loop. Dyna. 80(179), 51–59 (2013). ISSN 2346-2183 12. Pruna, E., Andaluz, V, Proaño, L.E., Carvajal, C.P., Escobar, I., Pilatásig, M.: Construction and analysis of PID, fuzzy and predictive controllers in flow system. In: 2016 IEEE International Conference on Automatica (ICA-ACCA), Curico, 1–7 2016 13. Pruna, E., Rosero, M., Pogo, R., Escobar, I., Acosta, J.: Didactic system for process control learning: case study flow control. In: Rocha, A., Adeli, H., Reis, L., Costanzo, S. (eds.) Trends and Advances in Information Systems and Technologies. World CIST’18 2018. Advances in Intelligent Systems and Computing, vol. 746. Springer, Cham (2018)

Error Diagnosis in Space Navigation Integration Using Wavelet Multi-Resolution Analysis with General Regression Neural Network Ramanan Gopalakrishnan, Diju Samuel Gnanadhas and Madhu Kiran Reddy Muli Abstract In this work, recent navigation systems rely on Kalman filtering for the fusion of data obtained from Global Positioning System (GPS) and the inertial navigation system (INS). The navigation combination offers consistent solutions of navigation by avoiding the generation of position errors with the consideration of time in case of INS. In current scenario, Kalman filtering INS/GPS method of integration contain some limitations related to stochastic error models of inertial sensors and the immunity to noise. The aim of this paper is to propose a system integration technique for the fusion of data from INS and GPS with the architecture of alternative general regression neural network (GRNN). The wavelet multi-resolution analysis (WMRA) is processed for comparing the position outputs of GPS and INS at various levels of resolution. The GR-ANN module is trained for predicting the position error of INS and offers vehicle positioning with increased accuracy.

1 Introduction In space vehicles, inertial navigation is one of the most versatile forms, which is used for automatic navigation. It uses accelerometers and gyroscopes for the measurement of state of the vehicle motion. By knowing the starting position of vehicle and the changes in its speed and direction, one can track the present position of the vehicle [1]. GPS is a navigation system operated by using satellites in space. The NAVSTAR constellation is located at an altitude of 20,200 km in six different orbits each inclined at 55° to the equator. One of the most important features of GPS is that it has a facility for degrading the accuracy for unauthorized users [2]. The hybrid R. Gopalakrishnan (B) · D. S. Gnanadhas · M. K. R. Muli Aeronautical Engineering, ACS College of Engineering, Bangalore, Karnataka 560074, India e-mail: [email protected] D. S. Gnanadhas e-mail: [email protected] M. K. R. Muli e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_18

191

192

R. Gopalakrishnan et al.

navigation system has been designed to exploit the advantages of both the systems while overcoming their inherent drawbacks. The INS can be calibrated using the GPS and the values obtained through GPS can be compared or cross checked with that of the values of INS. Therefore, the accuracy is assured in the navigation parameters thus obtained [3]. The process of obtaining a hybrid navigation system by combining INS and GPS is commonly known as integration of INS and GPS. Authors presented an iterative algorithm for two-scale wavelet decomposition. This an iterative method to decompose an arbitrary mother wavelet into a low pass filter and a band pass filter, where the filter pair will reproduce the mother wavelet through the two-scale equations [4]. Nourelin et al., utilized data fusion method of GPS/INS using radial basis function NNs that gives the architecture of multilayer perceptions, their training methods and drawbacks in them [5]. Author described learning algorithms for learning the network structure, in either a supervised or unsupervised manner in RBFN for multi-task [6]. Wavelet transforms and multi-resolution signal decomposition in case of machinery monitoring and diagnosis describes the multi-resolution signal decomposition depending on WT [7]. So this work aims to utilize general regression neural network to get better solution for the integration of GPS/INS. This paper proposes a new technique to estimate the error of the vehicle integration based on the INS outputs for signal outages in GPS. This technique uses the wavelet output for analysis and compares the outputs of GPS and INS position at various levels of resolution and finds the position error after processing them by general regression neural network (GRNN) for the reduction of error in position at the time of signal outages in GPS.

2 Inertial Navigation Systems INS is a self-contained system, which integrates three components of acceleration and three components of angular velocity in terms of time and converts them as navigation frame to produce the components of velocity, position, and attitude [1]. GPS make in use of a one-way ranging method from GPS satellites that broadcasts their evaluated positions. Signals obtained from the fourth satellite are utilized along the replica signal generated by the user and the relative phase is estimated. By using triangulation, the location of the receiver is fixed. In this work, only the first three parameters have been considered. The ranging signal of the GPS is broadcasted at two different frequencies, such as a primary signal of about 1575.42 MHz (L1) and a secondary broadcast of about 1227.6 MHz (L2). The frequency L1 is used for civilian purposes which have two modulations, via Clear Acquisition code and Protected Code.

Error Diagnosis in Space Navigation Integration Using …

193

2.1 Prediction of Errors in GPS As it is already known that though the accuracy of GPS does not degrade with time, it also has certain inherent errors. There are six major errors in GPS. The orbit parameters of satellite are termed as ephemeris. Errors present in the satellite position directly influence the fixing errors [8]. The ground station controls the orbital position of the satellites to ensure the range of errors resulting from inaccuracies in ephemeris remains within the specified limit of 0.5 m. Timing of transmission and the range measurement relies on satellite clock. The clock errors are corrected using the control stations. Multipath error is generated with the reflected signals from the surfaces nearer to the receiver that may either interface with, or be mistaken for the signal that takes straight line path from the satellite. Orbital perturbation errors occur due to distortion in the orbit. Instrument or receiver errors are the errors that are produced in GPS receiver as a result of electrical noise, time measurement, and position/range computation [9]. The range errors of about 1 m are possible in such causes. Geometric dilution of precision (GDOP) error is caused when two satellites are present very close to each other and the angle of cut among the range circles is shallow. In some of the GPS receivers, the alternative satellites can be selected to remove the GDOP; however, it must be noted that fixing continuity may be interrupted for a short period of time at the time of changeover.

2.2 Parameters Obtained Using the GPS signals obtained from the satellites, the position of the aircraft is known. The signals are further processed and the software is designed in such a way that it gives the vehicle position in terms of longitude, latitude, and altitude, in terms of which, the output of INS is also obtained, and hence, these parameters are used to integrate both the systems and make a hybrid navigation system [8]. Through validation, the wavelet multi-resolution analysis has shown improved performance by providing a maximum of 74% improvement in prediction accuracy in comparison to artificial neural networks.

3 Wavelet Analysis WMRA is carried out depending on the details and approximation provided by the WT. The INS or GPS signal is decomposed by the WMRA into both coarse resolutions that constitute the data related to the components of low frequency and provides the major characteristics of original signal, and fine resolution with the data related to the components of high frequency. This function is used in the signal successively to produce various components of fine resolution. In addition, this function is reversible

194

R. Gopalakrishnan et al.

since the reconstruction of signal is termed as the addition of detail components of all levels and the final approximation component. GPS and INS signals are obtained and synchronized in the time domain, but the time domain has poor importance as it is not capable of showing the properties of GPS and/or INS signal except their changes with respect to time [8]. However, the representation of frequency shows various frequency components present in the signal. In order to overcome this limitation, a new method is proposed with the concept of wavelet transform (WT).

4 Neural Networks An ANN is a processor that is parallel-distributed, normally used in the modeling of highly non-linear and complex problems. The learning process of ANN consists of two levels, where the first level involves in training using learning algorithms, and the second level involves in prediction with the processing of the data obtained as input to find the output depending on stored knowledge. In this research, the GRNN is utilized in modeling the INS position errors [5] (Fig. 1). The instantaneous time and position of INS act as inputs to the network in obtaining the position error of INS. The training of the network is done when there is a presence of GPS signal [8]. The variation among GPS signal and INS signal after processing with WMRA module at the three resolution levels acts as expected signal, which is utilized as the output of target network. The training algorithm makes modifications in the parameters of the network in order to lower the value of mean square error (Fig. 2). The architecture of GRNN operates in the prediction stage at the time of outages in GPS. The position of INS and the instantaneous time are then fed as inputs to the GRNN at this stage that involve in predicting the corresponding position error [10]. The corrected position component is obtained by removing the position error

Fig. 1 Architecture of GRNN networks

Error Diagnosis in Space Navigation Integration Using …

195

Fig. 2 Training procedure of GRNN module for the modeling of INS position errors

from the corresponding position component of INS. The error among the output of network and the expected output termed as INSEPS error is utilized in the update of output layer weight values. The INS/GPS navigation system depends on the GPS to give the data related to position. In the blockage of satellite signal, the INS position is assumed as the input of trained GRNN for the prediction of corresponding error in position. The predicted errors in position are neglected from outputs of INS to offer the accurate data related to position [8].

5 Result and Discussion The kinematic data used in this paper is generated by using the Flight Gear Simulator. All developments in this research are done in MATLAB software with the wavelets and GRNN functions (Fig. 3). In first phase, the INS/GPS is operated in the integration mode in the presence of continuous GPS signal. Processing of data is performed as: Initially, the decomposition of the signals of INS and GPS are done to the module of WMRA to three resolution levels. At every decomposition level, the details of INS and GPS are compared with each other and the variation obtained is considered as INS error. While achieving the third level of resolution, reconstructing and smoothening of error signal is done by removing lowest detail level possessing noises of high frequency [9] (Fig. 4). The INS/GPS error signal is used as expected output for training the GR-ANN. In order to analyze the developed method of data fusion, the execution of second phase is done, where the navigation system operates in the integration mode of INS/GPS in the presence of some simulated outage periods of GPS. There are two outages selected with intervals of 25 s. The position signals of both INS and GPS were decomposed with the wavelet transform. In the wavelet analysis, the approximations are normally of high scale and high frequency components [8].

196

Fig. 3 Decomposition of INS latitude component

Fig. 4 Decomposition of INS longitude component

R. Gopalakrishnan et al.

Error Diagnosis in Space Navigation Integration Using …

197

From Figs. 5 and 6, predicted that there is no error signal is available at the time of outages in GPS signal. At this instant, the only available signal is the erroneous INS signal. In order to reduce the error at this time of outages, the neural network

Fig. 5 INS/GPS real position error signal

Fig. 6 Predicted INS/GPS position error signal

198

R. Gopalakrishnan et al.

is used. At the outages of error signal, the values are predicted by using the general regression neural network. Figure shows the position error signal of INS and GPS without any outages. These error signals are eliminated from the erroneous INS signal without any inconvenience [9]. The capability of the projected model is validated with increasing the data accuracy is recognized by comparing WMRA to existing random forest regression (RFR) and artificial neural networks (ANN) models [10].

6 Conclusion In this paper, a novelty technique for signal error reduction in integrating GPS and INS signals in the navigation of vehicle is proposed with existing models. The fusion of data is carried out using the wavelet analysis and general regression neural network. The WMRA is utilized for the reduction of noise present in the signals of GPS and INS, and then given to GRNN. At the same time, the GRNN predicts the INS error at the time of outage in GPS providing 74% improvement. The developed WMRA and GRNN techniques are effective in lowering position errors with high accuracy.

References 1. Grochowski, M., Schweigler, M., Alrifaee, B., Kowalewski, S.: A GPS-aided inertial navigation system for vehicular navigation using a smartphone. IFAC-Pap. Online 51(10), 121–126 (2018) 2. Beaudoin, Y., Desbiens, A., Gagnon, E., Landry Jr., R.: Observability of satellite launcher navigation with INS, GPS, attitude sensors and reference trajectory. Acta Astronaut. 142, 277–288 (2018) 3. Oh, S.H., Hwang, D.H.: Low-cost and high performance ultra-tightly coupled GPS/INS integrated navigation method. Adv. Space Res. 60(12), 2691–2706 (2017) 4. PrešEren, P.P., Stopar, B.: Wavelet neural network employment for continuous GNSS orbit function construction: application for the assisted-GNSS principle. Appl. Soft Comput. 13(5), 2526–2536 (2013) 5. Chen, X., Shen, C., Zhang, W.B., Tomizuka, M., Xu, Y., Chiu, K.: Novel hybrid of strong tracking Kalman filter and wavelet neural network for GPS/INS during GPS outages. Measurement 46(10), 3847–3854 (2013) 6. Noureldin, A., El-Shafie, A., Bayoumi, M.: GPS/INS integration utilizing dynamic neural networks for vehicular navigation. Inf. Fusion 12(1), 48–57 (2011) 7. Wang, J.J., Wang, J., Sinclair, D., Watts, L.: A neural network and Kalman filter hybrid approach for GPS/INS integration. In: 12th IAIN Congress & International Symposium on GPS/GNSS, Jeju, Korea, 18–20, 2006 8. Hossain, M.A., Madkour, A.A.M., Dahal, K.P., Zhang, L.: A real-time dynamic optimal guidance scheme using a general regression neural network. Eng. Appl. Artif. Intell. 26(4), 1230–1236 (2013) 9. Zhang, T., Xu, X.: A new method of seamless land navigation for GPS/INS integrated system. Measurement 45(4), 691–701 (2012) 10. Adusumilli, S., Bhatt, D., Wang, H., Bhattacharya, P., Devabhaktuni, V.: A low-cost INS/GPS integration methodology based on random forest regression. Expert Syst. Appl. 40(11), 4653– 4659 (2013)

Hydromagnetic Squeeze Film Performance of Two Conducting Longitudinally Rough Elliptical Plates J. V. Adeshara, M. B. Prajapati, G. M. Deheri, and R. M. Patel

Abstract This article aims to analyze and discuss the presentation of the HS film between two conducting LR elliptical plates. The surface of bearing characterized by stochastic averaging model is assumed to be rough (longitudinally). The stochastically average type equation of Reynolds is solved with the boundary conditions concerned in order to obtain the characteristics of the bearing performance like pressure distribution and load-carrying capacity. The results establish that LR is more helpful as compared to transverse roughness. The calculated results are presented graphically and from this presentation it is clearly seen that the hydromagnetic lubrication substantially maximizes the load of the system of bearing. In addition, the LCC is maximized in the case of (−ve) skewed roughness due to increased plates conductivity and standard deviation associated with the Longitudinally Roughness. Moreover, in the case of (−ve) skewed roughness and (−ve) variance, the adverse effect of variance (+ve), positive skewness and aspect ratio of the plates can be compensated to some extent by the appropriate combination of conductivity and magnetization. Thus, this study makes it clear that longitudinal roughness must be given due respect while preparing the bearing systems.

Nomenclature BC B0 DP

Boundary condition Uniform transverse magnetic field Dimensionless pressure

J. V. Adeshara · M. B. Prajapati Mathematics Department, H. N. G. U, Patan–65, Gujarat, India e-mail: [email protected] G. M. Deheri Mathematics Department, S. P. University, Vallabh Vidyanagar, India R. M. Patel (B) Gujarath University, 15 Badrinath Society Ghodasar, Ahmedabad, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_19

199

200

HL HS h0 LBC LCC LR M MF p P r s TR w W α* σ* ε* μ

J. V. Adeshara et al.

Hydromagnetic lubrication Hydromagnetic squeeze Initial film thickness Load bearing capacity Load-carrying capacity Longitudinal roughness Hartmann number Magnetic field Lubricant pressure Non-dimensional pressure Radial coordinate Electrical conductivity Transverse roughness Load-carrying capacity Dimensionless load-carrying capacity Dimensionless variance (α/h) Dimensionless standard deviation (σ /h) Dimensionless skewness (ε/h3 ) Viscosity

1 Introduction Generally, roughness has a direct impact on the electrical, optical and mechanical solid thin-film instruments. For any this type of material roughness of the surface will have a very important effect on the macroscopic contact angle estimation on its surface which is flat, which may play an important role in different processes such as: humidifying, scattering and drenching. Due to the progress of up-to-date machine elements, various types of lubricants are used to fulfill the requirements for bearings operating under numerous conditions. To evade unpredicted viscosity variation with temperature, the use of liquid metals as lubricants has received extensive interest as compared to the conventional nonconducting lubricating oils; liquid metals possess a higher thermal conductivity and a higher electrical conductivity. This property of high thermal conductivity tells that the heat from the source of generation is readily conducted very far away. Further, this same behavior also implies that hydrodynamic flow behavior can be compensated up to certain extent by the application of an external magnetic field. The electric-field intensity results in a current density cooperating with the magnetic field to produce a Lorentz body force acting on the lubricant film, and this may give an element reverse to the direction of motion. As a result, the hydrodynamic characteristics of thin-film bearings with electrically conducting lubricants can be improved by the application of external magnetic fields.

Hydromagnetic Squeeze Film Performance of Two Conducting …

201

It is well known that hydromagnetic squeeze film behavior is employed when excessive temperature and speed occur. For instance, space entry vehicles make use of hydromagnetic effect. Further, investigation concerning the properties of a liquid metal lubrication reveals that if liquid metals like sodium and mercury could be inserted amid the moving surface of a bearing than a larger load are supported, with the help of large magnetic field. When electrically conducting lubricant is supported to large external electromagnetic field the electromagnetic pressurization came into effect. The investigation discussed in the literature talks about good amount of theoretical and experimental discussions regarding the hydromagnetic lubrication. (Elco-Huges [1], Kuzma [2], Shukla [3], Patel-Hingu [4], Sinha-Gupta [5]) Simplification of the scrutiny of HS films in between two plates of innumerable geometric shapes was modified by Patel-Gupta [6] resorting to Morgan-Cameron approximation. Prakash and Vij [7] modified the study of Wu [8] to obtain the LCC for a SF between spongy plates of several shapes including elliptical ones. Prajapati [9] dealt with the performance of a MF-based spongy squeeze film in elliptical plates. Here, the analysis of Reynolds’ (1886) was modified and developed. Because of the significant influence of roughness on the bearing performance, it has been studied of number of investigations. Both the type of roughness (transverse as well as longitudinally) has been studied. The investigations of Vadher et al. (circular, triangular and conical) suggested that magneto-hydrodynamic lubrication may counteract the effect of roughness (Figs. 1 and 2). Further, Vadher et al. [10] considered the transverse roughness effect on the hydromagnetic squeeze film performance. Therefore, it was thought to investigate the longitudinal roughness on this bearing configuration of Vadher et al. [10].

Fig. 1 Coordinate and geometry of the bearing system

202

J. V. Adeshara et al.

Fig. 2 Configuration of the roughness of the bearing system

2 Analysis The upper plate travels along its normal toward a non-porous fixed plate. Both plates are electrically conducted and lubricant packs the space between two plates. A standardized transverse magnetic field is used among the conical plates. The surfaces are longitudinally rough. The lubricated thickness of film is taken from (Christensen and Tonder [11–13]). As usual hydromagnetic lubrication assumption, the modified equation of Reynolds for lubricant film pressure is [9, 10].      • μ h h −3 1 − αh −1 + 6h −2 σ 2 + α 2 − 10h −3 ε + 3σ 2 α + α 3 ∂2 p ∂2 p  + 2 =   φ0 +φ1 +1  2 M M ∂x2 ∂z − tan h 3 M M M 2 2 φ0 +φ1 +(tan h 2 ) ( 2 ) (1) φ 0 and φ 1 are permeability of upper and lower plate where h is lubrication film thickness The concerned BC are P(x0 , z 0 ) = 0

(2)

z 02 x02 + =1 a2 b2 where a is major axis and b is minor axis. Solving Eq. (1) with BC Eq. (2) obtains the DP P=−

 P=

k (k)2 +1

 . 1−

x2 a2

p(πab)−1 h 3 •

μh

    2  − bz 2 1 − 3α ∗ +6 σ ∗2 +σ ∗2 − 10 ε ∗ +3σ ∗2 α ∗ +α∗3    φ0 +φ1 +1  4π M M − tan h M3 2 2 φ0 +φ1 +(tan h M2 ) ( M2 ) (3)

Hydromagnetic Squeeze Film Performance of Two Conducting …

203

Here, k(= a/b) is aspect ratio. Then, the LCC given by z= ab

x=a



a

2 −x 2

w=

pdxdz x=−a

√ z=− ab a 2 −x 2

is found in non-dimensional form as W =−

wh 3 •

μ h π 2 a 2 b2      k . 1 − 3α ∗ +6 σ ∗2 +α∗2 − 10 ε ∗ +3σ ∗2 α ∗ +α∗3 (k)2 +1  W =   φ0 + φ1 +1  8π M M tan h − M3 2 2 φ0 + φ1 +(tan h M2 ) ( M2 ) 

(4)

3 Results and Discussions One can see that for smooth bearing surfaces this investigation reduces to the study of Prajapati [9]. Further, when conductivities are taken to be zero the investigation of Prakash and Vij [7] can be obtained by considering M→0 (Tables 1, 2, 3, 4 and 5). The factor 1 φ0 + φ1 h(M/2) φ0 + φ1 + tan(M/2)

is responsible for the effect of conductivity on W. However, as tan h (M) ∼ = 1 and (2/M) ∼ 0 for large values M, the above factor tends to = φ0 + φ1 . φ 0 + φ1 + 1 So, it can be easily seen that the LCC gets maximize as the values of parameter of conductivity (φ 0 + φ 1 ) increases (Tables 6, 7, 8 and 9). A closed scrutiny of equation (W ) indicates that in the absence of flow the bearing with the MF can support a certain amount of load. Variation of LCC with respect to the magnetization parameter presented in Tables 1, 2, 3, 4 and 5 suggest that the LCC increases as the magnetization parameters increases. A noticeable difference in the role of standard deviation (Vadher et al. [10]) is found here as the load increases with increase in standard deviation (Tables 10, 11

204

J. V. Adeshara et al.

Table 1 Variation of load-carrying capacity with respect to M and φ 0 + φ 1 M

φ0 + φ1 = 0

φ0 + φ1 = 1

φ0 + φ1 = 2

φ0 + φ1 = 3

φ0 + φ1 = 4

4

0.68008513

1.04550487

1.16731146

1.22821475

1.26475672

6

0.81611055

1.63830498

1.91236980

2.04940220

2.13162165

8

0.97358144

2.43526044

2.92248677

3.16609993

3.31226783

10

1.14180712

3.42568055

4.18697169

4.56761727

4.79600461

12

1.31549170

4.60426944

5.70052869

6.24865832

6.57753609

Table 2 Distribution of load with respect to M and σ * M

σ* = 0

σ * = 0.05

σ * = 0.10

σ * = 0.15

σ * = 0.20

4

1.15205913

1.16731146

1.21306844

1.28933008

1.39609637

6

1.88738239

1.91236980

1.98733203

2.11226908

2.28718095

8

2.88430097

2.92248677

3.03704418

3.22797319

3.49527381

10

4.13226388

4.18697169

4.35109515

4.62463424

5.00758897

12

5.62604443

5.70052869

5.92398147

6.29640277

6.81779259

Table 3 Profile of load-bearing capacity with regards to M and α* M

α* = −0.10

α* = −0.05

α* = 0

α* = 0.05

α* = 0.10

4

1.33610388

1.16731146

1.02902369

0.91513964

0.81955839

6

2.18889714

1.91236980

1.68581728

1.49924461

1.34265684

8

3.34507632

2.92248677

2.57626882

2.29114816

2.05185047

10

4.79240488

4.18697169

3.69095414

3.28246910

2.93963344

12

6.52482117

5.70052869

5.02520473

4.46905559

4.00228757

Table 4 Variation of load-carrying capacity with respect to M and ε* M

ε* = −0.05

ε* = −0.025

ε* = 0

ε* = 0.025

ε* = 0.05

4

1.37067582

1.16731146

0.96394709

0.76058273

0.55721836

6

2.24553527

1.91236980

1.57920433

1.24603886

0.91287339

8

3.43163080

2.92248677

2.41334273

1.90419870

1.39505466

10

4.91640927

4.18697169

3.45753412

2.72809654

1.99865896

12

6.69365216

5.70052869

4.70740523

3.71428176

2.72115830

Table 5 Distribution of load with respect to M and k M

k = 1.50

k = 1.75

k = 2.0

k = 2.25

k = 2.50

4

1.34689784

1.25710465

1.16731146

1.08307249

1.00630298

6

2.20658053

2.05947517

1.91236980

1.77436373

1.64859465

8

3.37210012

3.14729344

2.92248677

2.71158566

2.51938515

10

4.83112119

4.50904644

4.18697169

3.88481910

3.60945836

12

6.57753311

6.13903090

5.70052869

5.28915033

4.91424887

Hydromagnetic Squeeze Film Performance of Two Conducting …

205

Table 6 Profile of load-bearing capacity with regards to φ 0 + φ 1 and σ * φ0 + φ1

σ* = 0

σ * = 0.05

σ * = 0.10

σ * = 0.15

σ * = 0.20

0

0.96086043

0.97358144

1.01174447

1.07534953

1.16439662

1

2.40344083

2.43526044

2.53071925

2.68981728

2.91255451

2

2.88430097

2.92248677

3.03704418

3.22797319

3.49527381

3

3.12473103

3.16609993

3.29020664

3.49705115

3.78663346

4

3.26898907

3.31226783

3.44210412

3.65849792

3.96144925

Table 7 Variation of load-carrying capacity with respect to φ 0 + φ 1 and α* φ0 + φ1

α* = −0.10

α* = −0.05

α* = 0

α* = 0.05

α* = 0.10

0

1.11639600

0.97561680

0.86027963

0.76529607

0.68557773

1

2.79248853

2.44035157

2.15185382

1.91426744

1.71486458

2

3.35118605

2.92859650

2.58237855

2.29725789

2.05796019

3

3.63053480

3.17271896

2.79764092

2.48875312

2.22950800

4

3.79814406

3.31919244

2.92679834

2.60365026

2.33243669

Table 8 Distribution of load with respect to φ 0 + φ 1 and ε* φ0 + φ1

ε* = −0.05

ε* = −0.025

ε* = 0

ε* = 0.025

ε* = 0.05

0

1.14319493

0.97358144

0.80396795

0.63435446

0.46474097

1

2.85952184

2.43526044

2.01099904

1.58673764

1.16247624

2

3.43163080

2.92248677

2.41334273

1.90419870

1.39505466

3

3.71768529

3.16609993

2.61451458

2.06292923

1.51134387

4

3.88931798

3.31226783

2.73521769

2.15816754

1.58111740

Table 9 Profile of load-bearing capacity with regards to φ 0 + φ 1 and k φ0 + φ1

k = 1.50

k = 1.75

k = 2.0

k = 2.25

k = 2.50

0

1.12336320

1.04847232

0.97358144

0.90332298

0.83929434

1

2.80991589

2.62258816

2.43526044

2.25951999

2.09936244

2

3.37210012

3.14729344

2.92248677

2.71158566

2.51938515

3

3.65319223

3.40964608

3.16609993

2.93761850

2.72939650

4

3.82184750

3.56705767

3.31226783

3.07323820

2.85540331

Table 10 Variation of load-carrying capacity with respect to σ * and α* σ*

α* = −0.10

α* = −0.05

α* = 0

α* = 0.05

α* = 0.10

0.00

3.29925336

2.88430097

2.54572018

2.26823668

2.03657615

0.05

3.34507632

2.92248677

2.57626882

2.29114816

2.05185047

0.10

3.48254521

3.03704418

2.66791475

2.35988261

2.09767343

0.15

3.71166002

3.22797319

2.82065796

2.47444002

2.17404504

0.20

4.03242077

3.49527381

3.03449846

2.63482039

2.28096528

206

J. V. Adeshara et al.

Table 11 Distribution of load with respect to σ * and ε* σ*

ε* = −0.05

ε* = −0.025

ε* = 0

ε* = 0.025

ε* = 0.05

0.00

3.39344500

2.88430097

2.37515693

1.86601289

1.35686886

0.05

3.43163080

2.92248677

2.41334273

1.90419870

1.39505466

0.10

3.54618821

3.03704418

2.52790014

2.01875610

1.50961207

0.15

3.73711723

3.22797319

2.71882915

2.20968512

1.70054108

0.20

4.00441785

3.49527381

2.98612977

2.47698574

1.96784170

and 12). However, the trends of LBC with respect to skewness and variance remain identical with the case of transverse roughness. Which can be seen from Tables 13, 14 and 15. It is also found that the influence of aspect ratio is registered to be equally strong. Besides, it is appealing to note that the effect of skewness remains almost marginal Table 12 Profile of load-bearing capacity with regards to σ * and k σ*

k = 1.50

k = 1.75

k = 2.0

k = 2.25

k = 2.50

0.00

3.32803958

3.10617027

2.88430097

2.67615554

2.48646635

0.05

3.37210012

3.14729344

2.92248677

2.71158566

2.51938515

0.10

3.50428174

3.27066296

3.03704418

2.81787604

2.61814153

0.15

3.72458445

3.47627882

3.22797319

2.99502667

2.78273551

0.20

4.03300824

3.76414103

3.49527381

3.24303756

3.01316708

Table 13 Variation of load-carrying capacity with respect to α* and ε* α*

ε* = −0.05

ε* = −0.025

ε* = 0

ε* = 0.025

ε* = 0.05

−0.10

3.85422036

3.34507632

2.83593228

2.32678825

1.81764421

−0.05

3.43163080

2.92248677

2.41334273

1.90419870

1.39505466

0.00

3.08541286

2.57626882

2.06712479

1.55798075

1.04883671

0.05

2.80029220

2.29114816

1.78200413

1.27286009

0.76371605

0.10

2.56099450

2.05185047

1.54270643

1.03356239

0.52441836

Table 14 Distribution of load with respect to α* and k α*

k = 1.50

k = 1.75

k = 2.0

k = 2.25

k = 2.50

−0.10

3.85970344

3.60238988

3.34507632

3.10367906

2.88368648

−0.05

3.37210012

3.14729344

2.92248677

2.71158566

2.51938515

0.00

2.97261787

2.77444335

2.57626882

2.39035252

2.22092140

0.05

2.64363250

2.46739033

2.29114816

2.12580757

1.97512773

0.10

2.36751977

2.20968512

2.05185047

1.90377878

1.76883661

Hydromagnetic Squeeze Film Performance of Two Conducting …

207

Table 15 Profile of load-bearing capacity with regards to ε* and k ε*

k = 1.50

k = 1.75

k = 2.0

k = 2.25

k = 2.50

−0.050

3.95957401

3.69560241

3.43163080

3.18398734

2.95830242

−0.025

3.37210012

3.14729344

2.92248677

2.71158566

2.51938515

0.000

2.78462623

2.59898448

2.41334273

2.23918398

2.08046787

0.025

2.19715234

2.05067552

1.90419870

1.76678230

1.64155060

0.050

1.60967845

1.50236656

1.39505466

1.29438061

1.20263333

in most of the cases (Tables 4, 8 and 11). It was observed that the performance of the bearing remains a little better when the plates are considered to be non-conducting. It is indicated by some of the graphical representations that the negative effect of longitudinal roughness can be completely countered by the effect of magnetic fluid lubrication, when the plates are conducting and negatively skewed roughness develops. Further, this positive effect goes ahead in the presence of variance (−ve). In fact, the standard deviation associated with roughness characterization changes the entire scenario until the case of transverse roughness, even in the absence of the magnetization the effect of longitudinal roughness turns out to be little favorable when the plates are non-conducting. A comparison of the present analysis with that of Vadher et al. [10] makes it clear that the negative effect introduced by roughness is comparatively more here.

4 Conclusion It is seen that in the case of longitudinal roughness, the bearing performance is more pronounced and reasonably better in comparison with the corresponding transverse case. For any type of improved performance, the role of aspect ratio remains predominant as can be noticed from the expression of load. Therefore, the graphical representations underline that this type of bearing system can be adoptable in industry by selecting a suitable aspect ratio. Furthermore, to extend the bearing system’s lifespan this study offers good measures. But from bearing design point of view the roughness must be treated with priority basis.

References 1. Elco, R.A., Huges, W.F.: Magnetohydrodynamic pressurization in liquid metal lubrication. Wear 5, 198–207 (1962) 2. Kuzma, D.C.: Magnetohydrodynamic squeeze films. J. Basic Eng. Trans. of ASME 86, 441–444 (1964) 3. Shukla, J.B.: Hydromagnetic theory of squeeze films. ASME 87, 142 (1965)

208

J. V. Adeshara et al.

4. Patel, K.C., Hingu, J.V.: Hydromagnetic squeeze film behaviour in porous circular disks. Wear 49, 239–246 (1978) 5. Sinha, P.C., Gupta, J.L.: Hydromagnetic squeeze films between porous annular disks. J. Math. Phys. Sci. 8, 413–422 (1974) 6. Patel, K.C., Gupta, J.L.: Behaviour of hydromagnetic squeeze film between porous plates. Wear 56, 327–339 (1979) 7. Prakash, J., Vij, S.K.: Load carrying capacity and time height relation for squeeze film between porous plates. Wear 24, 309–332 (1973) 8. Wu, H.: Squeeze film behaviour for porous annular disks. J. Lub. Tech. Trans. ASME 92, 593–596 (1970) 9. Prajapati, B.L.: On certain theoretical studies in hydrodynamics and electromagnetohydrodynamic lubrication. Dissertation, Ph. D. Thesis, S. P. University, V. V. Nagar (1995) 10. Vadher, P.A., Vinodkumar, P.C., Deheri, G.M., Patel, R.M.: Behaviour of hydromagnetic squeeze films between two conducting rough porous circular plates. J. Eng. Tribol. 222(4), 659–679 (2008) 11. Christensen, H., Tonder, K.C.: Tribology of rough surface: stochastic model of hydrodynamic lubrication, SINTEF Report No. 10/69-18 (1969a) 12. Christensen, H., Tonder, K.C.: Tribology of rough surface: parametric study and comparison of lubrication models, SINTEF Report No. 22/69-18 (1969b) 13. Christensen, H., Tonder, K.C.: The hydrodynamic lubrication of rough bearing surface of finite width, ASME-ASLE Lubrication Conference; Paper No. 70-Lub-7 (1970) 14. Radovi´c, N., Jokanovi´c, I., Mati´c, B., Šešlija, M.: A measurement of roughness as indicator of road network condition—case study. Tehniˇcki vjesnik 23(3), 881–884 881 (2016)

Performance and Emission Characteristics of Biodiesel from Rapeseed and Soybean in CI Engine R. Udayakumar, Vivek J. Shah and Sai Vijay Venkatesh

Abstract The use of renewable fuels is one approach for sustainable energy future for the world, with the crude oil prices going up day by day, and also keeping environment concerns in mind we need to look for other alternatives. The aim of this study is to identify the optimum blend ratio and compare the performance and emission characteristics of diesel with biodiesel and its blends and also compare theoretical values with the experimental values. In this study, a four stroke twin cylinder diesel engine is created in Diesel-RK software, and the engine is tested for various performance parameters like brake power, brake mean effective pressure, brake thermal efficiency, and brake specific fuel consumption, and emission characteristics like NOx emissions, CO2 emission, and particulate matter emission are noted, and then different biofuels like rapeseed methyl esters and soybean methyl esters are designed along with different blends (B100, B20, B40), after obtaining the theoretical values from the software the same parameters are obtained from an actual engine of similar setup as that of the software and then the values are compared. From the simulations, it was seen that the performance and emission characteristics of diesel were better than biodiesel and its blend. B20 blend is the optimum blend ratio.

1 Introduction Internal combustion engines as prime movers in roadways and railways have benefitted humans in many ways. It has been serving the humans for more than a century, but it has many problems associated with it like the harmful tailpipe emissions (NOx , CO2 , etc.). The price of petroleum is rising day by day. Petroleum fuel is the dominant source of global warming and its combustion poses a major threat to the environment. R. Udayakumar (B) · V. J. Shah · S. V. Venkatesh Department of Mechanical Engineering, Bits Pilani Dubai Campus, Dubai International Academic City, Dubai, UAE e-mail: [email protected] S. V. Venkatesh e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_20

209

210

R. Udayakumar et al.

IC engines run on two energy sources gasoline and diesel both of which will vanish from the earth’s surface in a few decades’ times. Therefore, it is very important to find alternative fuels to tackle all these issues. Biodiesel is made from non-edible seed oils, i.e. transesterified form of seed-based oils. Biodiesel is produced through a chemical reaction transesterification and esterification. In this process, vegetable oil or animal’s fat is reacted with alcohols (methanol, ethanol). The transesterification reaction can be catalyzed by either acids or bases. The most common means of production is the base-catalyzed transesterification. Properties of rapeseed oil are: (1) (2) (3) (4) (5)

Lower heating value (MJ/kg): 39.45 Cetane number: 54.4 Density of fuel (kg/m3 ): 874 Specific vaporization heat (KJ/kg): 325 Fuel thermal capacity (J/kg * K): 185. Properties of soy methyl esters are:

(1) (2) (3) (4) (5)

Lower heating value (MJ/kg): 36.22 Cetane number: 51.3 Density of fuel (kg/m3 ): 874 Specific vaporization heat (KJ/kg): 325 Fuel thermal capacity (J/kg * K): 1853.

In recent times, the most important type of alternative fuel is fatty acid methyl esters (FAME). It is predicted that the demand for such type of fuel will continue to increase in the near future. Soybean methyl esters were used because it is not yet exploited fully and a lot of research is going on regarding this FAME. Soybean oil in country like India is mainly used for edible purpose and is very important from economic point of view. Soybean oil is the second most used vegetable oil. In the recent years, the amount of soybean oil traded is developing at the rate of 4.05% which is around 25% of the global aggregate oils and fats production. Brazil, China, Argentina, and India alongside USA are the significant contributors of soybean oil for the development. In Paris, a decision was made to use 30% RME additives with diesel. There are around 400 buses in Stockholm which are powered by FAME. There are 27 buses powered by FAME utilized in Burgos (Bulgaria). The advantages of using rapeseed methyl esters and soybean methyl are: high cetane number, good lubrication qualities, high ignition temperature, reduce engine noise, low sulfur content, good biodegradability, and cheap production available in sufficient amounts miscible with diesel.

2 Literature Review Peterson et al. [1] in their paper performed test on a road vehicle on a transient chassis dynamometer. They concluded that 100% RME and 100% REE reduced HC, CO

Performance and Emission Characteristics of Biodiesel …

211

and NOx emissions and increases in CO2 and PM. Khatri [2] in his paper explained that there was 13% decrease in brake power and 9% increase on BSFC. The NOx emissions are decreased 40% also there were significant decrease was noted for CO and CO2 emission when rapeseed methyl ester was used. Labeckas et al. [3] in their paper operated engine on neat RME and its 5, 10, 20, and 35% blends with diesel fuel and concluded that specific fuel consumption is higher for RME compared to diesel and brake thermal efficiency for diesel is higher. The maximum NOx emissions increase proportionally with the mass percent of oxygen. The CO2 , emissions along with the fuel consumption, are slightly higher for the B20 and B35 blends and neat RME. Fasogbon et al. [4] in their experiment tested many blend ratios like (0/100, 10/90, 20/80, 30/70, and 40/60) of soybean methyl ester and diesel and concluded that B20 is the best blend ratio. Pexa et al. [5] in their paper chose 100% rapeseed methyl ester and 100% hydrogenated oil as fuel and concluded that the operating parameters of the internal combustion engine does not change significantly when using these fuels. Chlopek et al. [6] they concluded that brake power and torque for diesel were higher when compared to RME and the CO emission was reduced when RME was used. Amosu [7] compared RME and diesel at different loads (no load, 25, 50, 100%) thermal efficiency of RME B20 blend is 42.38% comparable to diesel and NOx emission is higher for B20 blend, hydrocarbon emission is low for RME. Bhatti et al. [8] in their paper indicated that fuel consumption for SME is 14.65% higher than diesel and CO emission for B20 and B100 SME is 11.36% and 41.7% lower than diesel, respectively.

3 Methodology A basic four stroke twin cylinder diesel engine was designed in Diesel-RK software, then two biodiesels rapeseed methyl ester and soybean methyl ester and its blends B100, B40, B20 were also designed in Diesel-RK and then these samples of biodiesel were tested in an actual engine of similar setup as done in the computer simulation, and then the theoretical and experimental values were compared and the results were interpreted. The simulation and engine parameters used are: (1) (2) (3) (4) (5) (6) (7)

Engine bore: 150 mm Engine stroke: 180 mm Compression ratio: 17.5:1 Values in the cylinder: 4 Engine RPM used: 200, 500, 1000, 1500, 2000, 2500, 3000, 3500 Injection pressure: 800–1000 bar Cooling system: water cooling.

Experimental methodology

212

R. Udayakumar et al.

An experimental test rig is developed to perform the experiment to obtain different performance and emission characteristics. The setup consists of a four stroke twin cylinder engine diesel engine. The RPM of the engine was varied from 200 to 3500 while keeping the load constant. Setup is provided with necessary instruments for combustion pressure and crank-angle measurements. These signals are interfaced to computer through engine indicator for Pθ –PV diagrams. Provision is also made for interfacing airflow, fuel flow, and temperatures. This setup helps us in getting performance parameters like brake power, thermal efficiency, specific fuel consumption, and heat balance sheet also helps in getting emission parameter values. The engine configuration is very similar to that used in theoretical setup in the simulation software. Rapeseed Methyl Ester Preparation: The collected rapeseeds are crushed and rapeseed oil is obtained. The rapeseed oil is poured in the conical flask. Methanol is added to the conical flask containing rapeseed oil. Slowly add potassium hydroxide solution. Adding the chemicals can be done directly into the conical flask with a top pan zeroing the balance after each reaction Heat the mixture at 30 °C for 60 min. Phase separation was performed in a separatory funnel to remove glycerol. Stir or swirl for 10 min. The biodiesel is ready. Soybean Methyl Ester Preparation: Soybean oil is collected in the conical flask. Methanol is added to the conical flask. Slowly add potassium hydroxide solution. Adding the chemicals can be done directly into the conical flask with a top pan zeroing the balance after each reaction. Heat the mixture at 30 ° C for 60 min. Phase separation was performed in a separatory funnel to remove glycerol. Stir or swirl for 10 min. The biodiesel is ready.

4 Results and Discussion The results obtained from the simulation were interpreted. The simulation was carried out at various engine rpm as mentioned before while keeping the load constant. Biodiesels (rapeseed methyl esters and soybean methyl esters) and its blends B100, B40, B20 all were tested and results were obtained. For comparing the simulation results with experimental results, the engine was made to run at the same rpms as that in the simulation and using the similar settings to that of the simulation. The results obtained in the experimental setup were compared to the simulation for various engine parameters like brake power, specific fuel consumption, brake thermal efficiency, and emission parameters like CO2 emissions, NOx emissions, and particulate matter emissions. The optimum blend ratio was determined after looking at the simulation and the experimental results.

Performance and Emission Characteristics of Biodiesel …

213

4.1 Brake Power From Figs. 1 and 2, it can be seen that the theoretical values for brake power are greater than the experimental values for brake power. The brake power in both Figs. 1 and 2 increases as the RPM increases and it increases until 2500 rpm and then starts to decrease. From both figures, it is noted that diesel produces the highest brake power and pure biodiesel B100 produces the least brake power. It can be seen that the brake power is lower for blends of biodiesel than diesel. This is attributed to the Fig. 1 Brake power versus engine RPM—theoretical

Fig. 2 Brake power versus engine RPM—experimental

214

R. Udayakumar et al.

lower calorific value of biodiesel when compared to pure diesel. The brake power values of soybean methyl ester B20 and rapeseed methyl ester B20 are very similar to that of diesel at all the rpms. Therefore, it can be said that as the biodiesel blend ratio increases brake power decreases. The experimental values are lower compared to the simulation values because various factors such as friction and other types of heat losses are not considered in the simulation.

4.2 Specific Fuel Consumption From Figs. 3 and 4, it is seen that theoretical specific fuel consumption values are low compared to experimental values. It is noted that as the engine rpm increases the specific fuel consumption values decreases till 1000 rpm and then starts to increase.

Fig. 3 Specific fuel consumption (theoretical)

Fig. 4 Specific fuel consumption (experimental)

Performance and Emission Characteristics of Biodiesel …

215

Diesel has the lowest fuel consumption values compared biodiesel. The reason for higher fuel consumption for blends is the lower heating value. The higher proportion blends increases the viscosity which in turn increases the fuel consumption due to poor atomization of the fuel. B20 blend of both the biodiesels used produces fuel consumption values very similar to that of diesel. The fuel consumption of experimental setup is higher as the losses such as heat and friction are not considered in the simulation.

4.3 Brake Thermal Efficiency From Figs. 5 and 6, it can be seen that theoretical values of brake thermal efficiency are higher than the experimental values. It can be seen from the graph that as the engine RPM increases the brake thermal efficiency decreases. Pure biodiesel B100 for both the biodiesel used had the least brake thermal efficiency compared to diesel, and B20 blend of both the biodiesels used had brake thermal efficiency similar to that of diesel. It is evident that diesel has the highest brake thermal efficiency among all the tested biodiesel this is because of higher calorific value of diesel, with higher calorific value more amount of heat is produced in the combustion chamber, and subsequently Fig. 5 Brake thermal efficiency versus engine RPM—theoretical

Fig. 6 Brake thermal efficiency versus engine RPM—experimental

216

R. Udayakumar et al.

more power is produced for a given amount of fuel and also the viscosity of diesel is less compared to biodiesels used, therefore, diesel has better atomization of fuel and hence better fuel economy and higher thermal efficiency. Therefore, as the blend ratio of biodiesel decreases the brake thermal efficiency increases.

4.4 NOx Emission From Figs. 7 and 8, it can be seen that as the engine RPM increases the NOx emission decreases. B100 (pure biodiesel) of both the biodiesels used has the highest NOx emission, when compared to diesel and B20 blend of both the biodiesels used have NOx emission values similar to that of diesel. NOx emissions decreases as the RPM increases because nitrogen reacts with oxygen only above 800 °C and as the engine speed increases the time required for both to react decreases. Therefore, two things happen less oxygen for nitrogen to react and as there is less oxygen for combustion less cylinder temperature and hence less NOx emissions. Biodiesels have higher Fig. 7 NOx emission versus engine RPM—theoretical

Fig. 8 NOx emission versus engine RPM—experimental

Performance and Emission Characteristics of Biodiesel …

217

NOx emissions compared to diesel because of higher oxygen content in biodiesels. Therefore, as the blend ratio of biodiesel decreases the amount of NOx produced also decreases.

4.5 CO2 Emission From Figs. 9 and 10, it can be seen that the theoretical values of CO2 emissions are lower than that of experimental values. It was noted that initially at lower rpms CO2 emissions are high as the rpm increases CO2 emission decreases till 1000 rpm and then it starts increasing again. Pure biodiesels B100 have the highest CO2 emissions compared to diesel. B20 blend of both the biodiesels has CO2 emissions value similar to that of diesel. As the RPM of the engine increases more and more fuel is injected into the combustion chamber and as more fuel is injected more and more CO2 is produced and emitted. Higher content of oxygen in biodiesel is the reason why it produces more CO2 . Carbon gets oxygenated inside the combustion chamber and as a result it produces more carbon dioxide. Therefore, as the blend ratio of biodiesel increases CO2 emission also increases. Fig. 9 CO2 emission versus engine RPM—theoretical

Fig. 10 CO2 emission versus engine RPM—experimental

218

R. Udayakumar et al.

Fig. 11 Particulate matter emission versus engine RPM—theoretical

Fig. 12 Particulate matter emission versus engine RPM—experimental

4.6 Particulate Matter Emission From Figs. 11 and 12, it can be seen that the theoretical values of particulate matter emission are lower than that of the experimental values. It can be seen that as the engine rpm increases particulate matter emission decreases till 1000 rpm and after 1000 rpm as the rpm increases particulate matter emission also increases. Diesel produces the least amount of particulate matter. It was also noted that B100 produces very high amount of particulate matter. B20 blend particulate matter emission values are similar to that of diesel. Therefore, as the blend ratio of biodiesel increases emission also increases.

5 Conclusion A four stroke twin cylinder diesel engine was designed in Diesel-RK software. The theoretical values obtained from the Diesel-RK software were compared with the experimental values which were obtained from an actual diesel engine of similar setup as that of the software and both the results were interpreted. It was concluded that diesel was better when compared biodiesels and its blends in terms of performance

Performance and Emission Characteristics of Biodiesel …

219

and emission. B20 blend ratio showed performance and emission characteristics values very similar to that of diesel both experimentally and theoretically. B100 (pure biodiesels) showed very poor results and it can be said that biodiesels should always be used in blends with diesel. Therefore, B20 blend ratio is the optimal blend ratio.

References 1. Peterson, C., Recce, D.: Emission characteristics of ethyl and methyl ester of rapeseed oil compared with low sulphur diesel control fuel in a chassis dynamometer test of a pickup truck. Am Soc Agri Eng 39(3), 805–816 (1996) 2. Khatri, P.: Comparison of performance and emission characteristics of rapeseed biodiesel and diesel. Indian J. Appl. Res. 6, 313–316 (2016) 3. Labeckas, G., Slavinskas, S.: The effect of rapeseed oil methyl ester on direct injection diesel engine performance and exhaust emission. Energy Convers. Manag. 47, 1954–1967 (2006) 4. Fasogbon, S.K., Asere, A.: Effects of soybean methyl ester on the performance characteristics of compression ignition engine. Int. J. Mech. Mechatron. Eng. 8, 485–488 (2014) 5. Pexa, M., Cedik, J., Marik, J., Honig, V., Hornickova, S., Kubin, K.: Comparison of the operating Characteristics of the internal combustion engine using rapeseed oil methyl ester and hydrogenated oil. Agron. Res. 13(2), 613–620 (2015) 6. Chlopek, Z., Jagiello, S., Juwa, S., Skrzek, T.: Comparative examination of performance Characteristics of an IC engine fueled with diesel oil and rape methyl esters. Archiwum Motoryzacji 74, 19–32 (2016) 7. Amosu, V., Bhatti, S.K.: Experimental investigation of performance and emission characteristics of DI diesel engine with rapeseed methyl ester. SSRG Int. J. Mech. Eng. 117–121 (2017) 8. Bhatti, S.K., Al Dawody, M.: Experimental and computational investigation for combustion, performance and emission parameters of diesel engine fueled with soybean biodiesel diesel blends. Energy Procedia 52, 421–430 (2014)

A Practical Approach to Camera Calibration for Part Alignment for Hybrid Additive Manufacturing Using Computer Vision Pallavi Kulkarni, Atul Magikar and Tejas Pendse

Abstract Additive manufacturing (AM) is a process that involves building up thin layers of material, one by one creating a three-dimensional object from a digital file. The technology can produce complex shapes that are not possible with traditional casting and machining methods, or subtractive techniques. Hybrid AM is the process in which the AM process takes place on an existing part in the AM chamber. The existing part could have been manufactured by the traditional subtractive manufacturing process. The powder bed fusion process includes the following commonly used printing techniques such as direct metal laser sintering, electron beam melting, selective heat sintering, selective laser melting (SLM) and selective laser sintering (SLS). Laser sintering requires a scan path by which the laser travels and welds the metal powder to form layers over layers. Computer vision can help to locate the part on which the laser has to be fired. Camera calibration is one of the techniques in computer vision by which the world (object) 3D coordinates are converted into 2D image coordinates. Practically, it becomes essential to know the exact position of the camera, i.e., camera intrinsic. Camera calibration technique provides camera intrinsic and gives out extrinsic, i.e., 2D coordinates.

1 Introduction Metal 3D printing or additive manufacturing (AM) involves manufacturing a metal part layer by layer, bonding each layer of metal alloy powder to one another using a high-powered laser using SLS [1]. Traditional 3D printing is done by constructing P. Kulkarni (B) · A. Magikar Symboisis Institute of Technology, Symbiosis International (Deemed) University, Lavale, Pune, India e-mail: [email protected] A. Magikar e-mail: [email protected] T. Pendse Renishaw Metrology Systems Limited, Pune, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_21

221

222

P. Kulkarni et al.

each layer from scratch. Hybrid 3D printing is the technique where only a section of an existing part is made using the normal 3D printing techniques. Since the bonding of adjacent layers is caused by melting metal by laser, the proximity tolerance for adjacent layers needs to be very low for optimal results. This means that the spatial accuracy of the laser is critical to the structural integrity of the final part. For this, we need to accurately know where the part is located on the base plate inside the AM chamber. Current techniques of locating the part involve placing the part on a tile plate using fixture such that its 3D coordinates are fixed to the tile plate as well. This plate is then moved into the AM chamber and the part coordinates can be used to ascertain where the laser should be fired. There are various drawbacks to this process. Fixing a part to the tile plate puts limitation on how the part can be positioned, for example, how the part is shaped or how a fixture can be used with it. Adding a fixture adds an additional complexity of how the laser can be fired without the fixture obstructing it. Additionally, there is no feedback once the part is moved inside the AM chamber. If the part moves during installation, it can only be detected once the printing process is done, leading to additional effort, time and cost to verify part integrity. The aim of this paper is to investigate if a camera mounted inside the AM chamber can be used to accurately locate the part. Using camera vision algorithms, we aim to establish a camera coordinate system and use image processing techniques to identify the location, orientation and size of the part inside the chamber. The expected results from the project are a workflow, which will allow the locating of any part kept on the base plate inside the AM chamber. For the first time, authors have decided to implement computer vision algorithms, i.e., camera calibration which gives the camera intrinsic and extrinsic matrices and image processing for locating the part in AM machine chamber [2].

1.1 Problem Identification The existing part which is manufactured by subtractive manufacturing or a broken part which needs a fix with AM having complex geometry, it becomes essential to locate the part in order to fire the laser at the correct position. The current process of AM follows a scan path generated in software for the CAD model of the part to be produced to generate part file which is then provided to the AM machine for part generation. Requisite for correct alignment of the part on the machine bed on which the laser firing has to take place for hybrid AM. Fixing a part to the tile plate puts limitation on how the part can be positioned, for example, how the part is shaped or how a fixture can be used with it. Adding a fixture adds an additional complexity of how the laser can be fired without the fixture obstructing it. Additionally, there is no feedback once the part is moved inside the AM chamber. If the part moves during installation, it can only be detected once the printing process is done, leading to additional effort, time and cost to verify part integrity.

A Practical Approach to Camera Calibration for Part …

223

The laser scan path of the CAD model can be generated anywhere on the workspace but the part which is already placed should match with the starting coordinates of the part that needs to be built over it. Therefore, to determine 3D coordinates (start point) where the laser is to be fired on the existing part to be additive manufactured with the help of CAD and image processing software, to develop a transformation matrix in order to find the coordinates of where to start firing the laser, i.e., the start point on the part, to calibrate the camera for the correct view of the component to be manufactured on a real AM machine and validate it using real-world measurements, is the objective of this paper.

1.2 Camera Calibration Geometric camera calibration is computation of the mapping between points in the scene and their corresponding points in the image. A lens distortion model can be used to find the correct mapping between the image points and the distorted ones [3]. The Zhang’s calibration model requires only a camera to observe a planar pattern shown at a few different orientations which can linearly calibrate the camera if the lens distortion is known [4]. OpenCV library is used to run the calibration algorithm [5, 15]. A transformation of the OpenCV algorithm is proposed which is a promising technique to find 3D coordinates from any random 2D image points. The proposed algorithm for 2D planar camera calibration is given in Fig. 1. The algorithm demonstrates how the output parameters from the OpenCV camera calibration can be used to find the inverse homography matrix to find 3D coordinates from any 2D image point. The output parameters from OpenCV are used to find the inverse homography matrix which then contributes to the formulation of corresponding 3D point. Reprojection error between the image points and the projected points is calculated for each image in the series which is the Euclidian distance between the image point and the projected point. The accuracy of this algorithm is determined using a test image with ground truth measurements.

2 Literature Review Chatterjee et al. explained the analytical solutions from a given set of image points when the world points lie on a two-dimensional plane for the determination of camera parameters [6]. Weng proposed a two steps calibration procedure, one with solution based on distortion-free camera model and parameters estimated in the first step are improved iteratively through nonlinear optimization taking into account camera distortions in the other. Correcting the tangential and radial distortion leads to improvement. As the results demonstrated that a wide-angle lens causes more distortion than a tele lens [7]. Shang et al. mentioned that solution for the perspective of four coplanar control points with known effective focal length can be found out

224

P. Kulkarni et al.

OpenCV camera calibration

Output rotation vectors, translation vectors, Camera Matrix(C), Image Points (I).

Convert rotation vectors to rotation matrix using Rodrigues rotation formula. Concatenate translation and rotation vectors to form transformation Matrix (T).

Assume z=0, get transformation Matrix(T’). Homography(H)= T’ * C………. [10] Find inverse (H) = H’

3D Point= H’ * (Ix, Iy)

Accuracy Test of the algorithm using any Random image points measurement. Fig. 1 Proposed algorithm

analytically using control lines as reference objects suppressing noise and giving precise results [8]. Jose Maurıcio et al. used a 3D vision-based measurement system consisting of a single CCD camera mounted on the robot tool flange. For modeling and performing robot calibration, including lens distortion processes and to measure the robot end-effector pose relative to a world coordinate system with portability, accuracy and low cost leading to average accuracy varying from 0.2 to 0.4 mm, at distances from 600 to 1000 mm from the target, respectively [9]. Park et al. presented the use of single camera in pose estimation for the vehicle’s wheel alignment system using results of camera calibration using both ideal pinhole model and lens distortion model. A coordinate matching algorithm was developed for calculating matched sequence of order in both extracted image coordinate and object coordinate for non-interactive calibration and pose estimation. [10]. Song et al. described methods for calibration of two cameras firstly by calibrating both individually followed by calibrating them together. The relationship between the world, right camera and left camera is given by four rotation matrices: world to right camera, world to left camera, right to left camera and left to right camera. Initially, the distortion parameters which include radial and tangential distortions of camera lens are calibrated followed by

A Practical Approach to Camera Calibration for Part …

225

the calibration of position parameters which include focal length, rotation-translation matrices and image center [11].

3 Experimentation The experiment was carried out using the AM machine camera. A series of images of checkerboards with different checker box sizes and with different control points were passed through the camera for the calibration process [12]. Based on the calibration results, the calibration pattern with minimum error is considered for the final calibration process and testing purpose. A test image with known measurements is also passed through the same camera and the pixel points of the nearest and farthest points are used for testing the accuracy of the process.

3.1 Setup The experimental setup with was a replica of the AM machine chamber as a setup. The camera is set at the same height and width as it is in the machine with more light and with no enclosure. A series of five images with each checkerboard size was passed through the calibration process. The noise in the images was removed using image processing [13, 14] (Fig. 2). Details of the equipment used are given. Table 1 shows total seven variations of checkerboard sizes with varying control points are used for the calibration process here. The results obtained are plotted in the graphs.

Fig. 2 Series of images for calibration

Table 1 Details of the setup

Camera name

mvBlueCOUGAR camera.

Sensor information

CMOS sensor 1/2.3

Lens

Kowa make with diagonal field of view 95.4°

Resolution

2764 * 3856 px

226

P. Kulkarni et al.

Figure 3 demonstrates that the reprojection errors versus control points graph show the min. reprojection error occurs between the control points 60–90. Hence, the control points between 60 and 90 can be used to calibrate the camera. Figure 4 shows the reprojection errors versus square size graph show the minimum reprojection error is between the size 18–20 mm. Therefore, a checkerboard of size 8 * 10 with square size 18.98 mm with 63 control points is used for the calibration purpose.

3.2 Results With the selected checkerboard the calibration process was again iterated, and the obtained results were compared with the test image with ground truth images (Fig. 6). Figure 5 shows the reprojection of projected points on the actual image points. This test image is also passed through the same camera and the pixel points are obtained in order to find the actual 3D coordinates. By applying Euclidian distance Fig. 3 Graph of control point versus reprojection error

Fig. 4 Graph of control point versus reprojection error

Fig. 5 Graph of projected points versus image points

A Practical Approach to Camera Calibration for Part …

227

Fig. 6 Test image with measurements (true values)

formula, the distance between two points is obtained which is then compared to the actual distances for the triangle and rectangle (see red circular marks on the image). These seven red points are used for the 2D–3D conversion. The distance between two points is calculated by distance formula. The obtained values are then matched with the ground truth. Percent error is calculated.

3.3 Result Table Table 2 shows the error in finding the actual distances with the obtained 3D point. The minimum error determined is 69 µ and the maximum error obtained is 2500 µ. The error completely varies from point to point which is well observed in the result table. Table 2 Results for calibration for checkerboard of size 8 * 10 Pixels Ground truth

x

y

(Control points: 63, square size: 19.98) Obtained 3D points

OpenCV results

X

Y

Distance

Difference

% error

Triangle 86.185

1147

879

0.017840768

0.155969189

85.58429

0.6007149

86.5175

1801

925

0.01710161

0.070388096

84.36582

2.1516824

0.697006 2.486991

86.335

1539

461

−0.055296558

0.113702026

84.47237

1.8626283

2.157443

63.6175

1122

1617

0.118064245

0.140722541

63.53347

78.5025

1665

1663

0.117983589

0.077189123

78.67311

63.645

1583

2379

0.196656602

0.077311849

63.57526

983

2329

0.197280947

0.140884048

79.21687

Rectangle

78.6

0.0840308 −0.170608 0.0697355 −0.616866

0.132088 −0.21733 0.109569 −0.78482

228

P. Kulkarni et al.

4 Conclusion Camera calibration process and determination of its accuracy are a crucial factor in precision applications. The work presented here shows that the error completely varies from point to point which is well observed in the result table (Table 2). The error is minimum for the image coordinates which appear closer to the camera lens (rectangle), i.e., 69 microns. But, the error is maximum in case of the image points which appear away from the lens (triangle), i.e., 2500 µ.

5 Future Work The projection of images and 3D–2D conversions contribute to a possible loss of data. For 2D camera calibration, the setup is simple and the third coordinate is termed as zero. The calibration results for different focal lengths can be obtained by moving the selected 6 * 8 checkerboard pattern in different orientations toward or away from the camera lens till the “sweet spot” of accurate calibration, i.e., point with lowest reprojection error is detected.

References 1. Harun, W.S.W., Kamariah, M.S.I.N., Muhamad, N., Ghani, S.A.C., Ahmad, F., Mohamed, Z.: A Review of Powder Additive Manufacturing Processes for Metallic Biomaterials, pp. 128–151 (2018) 2. Majid, S.: Accurate camera calibration algorithm for laser applications: the first international conference on Laser applications and advanced materials, university of technology. 3. Sonka, M., et al.: Image Processing Analysis, and Machine Vision, 3rd edn (2013) 4. Zhang, Z.: A flexible new technique for camera calibration. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 22, no. 11, pp. 1330–1334. Nov (2000) 5. The OpenCV Reference Manual Release 3.0.0-dev, June 25, 2014 6. Chatterjee, C., Roychowdhury, V.P.: Algorithms for coplanar camera calibration. Mach Vision Appl. 12, 84–97 7. Weng, J.: Camera calibration with distortion models and accuracy evaluation. 14 (1992) 8. Shang, Y., Yu, Q., Zhang, X.: Analytical method for camera calibration from a single image with four coplanar control lines (2004) 9. Mottaa, J.M.S.T., de Carvalhob,G.C., Mc Masterc, R.S.: Robot calibration using a 3D visionbased measurement system with a single camera. Robot. Comput-Integr. Manufact. 1(6), 87– 497 (2001) 10. Park, M.-S., Kwon, J.-W., Park, M.H., Kin, J.S., Hong, S.-K., Han, S.W.: Experimental study on camera calibration and pose estimation for the application to vehicle’s wheel alignment. In: SICE-ICASE International Joint Conference, pp. 2952–2957. IEEE, Busan (2006) 11. Song, L.M., et al.: High precision camera calibration in vision measurement. Opt. Laser Technol. 39, 1413–1420 (2007) 12. Tsai, R.: A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses 323–344 (1987) 13. Malay Pakhira, K.: Digital Image Processing And Pattern Recognition (2011)

A Practical Approach to Camera Calibration for Part …

229

14. Heikkila, J., Silven, O.: A Four-step Camera Calibration Procedure with Implicit Image Correction. IEEE, pp. 1106–1112 (1997) 15. Kaehler, A., Bradski, G.: Learning OpenCV 3 (2016)

Development of a Colour and Orientation Detection System for Small Part Feeding Aditya Sachdeva, Rohan Kapoor, Awwal Singh and Pradeep Khanna

Abstract With the growing industrial conscience for obtaining maximum possible output in minimum time, automation has become an indispensable tool. The objective of the present research is to develop a low-cost electronic detection system based on Arduino microcontroller. The present work has been carried out by making the small parts move on the tracks of a vibratory bowl feeder which is extensively used in industry for small part feeding on assembly lines. The work can, therefore, be seen as important as it will make conventional vibratory bowl feeding systems work in a smarter manner. The present system consists of two sensing elements namely IR and colour recognition sensor. The former is used to check the orientation of the components, whereas the latter senses the colour. A servo-based actuating plunger discards the components of undesired orientation and colour. It is expected that the system will prove to be beneficial for industrial applications.

1 Introduction In manufacturing industries, high speed and accurate feeding of parts are essential for an efficient assembly system. For small- and medium-sized production units, classical automation with dedicated equipment may be feasible but is not flexible enough. With the help of robots in assembly lines, higher flexibility can be achieved but getting the robot to pick parts from bulk is not possible [1]. Part feeding is therefore of critical concern in automated assembly lines, where parts are required to be correctly oriented for accurate feeding into an assembly line at the desired feed rate [2]. In mass production, part feeding is essential and majorly relies on A. Sachdeva (B) · R. Kapoor · A. Singh · P. Khanna MPAE Division, NSUT, New Delhi, India e-mail: [email protected] R. Kapoor e-mail: [email protected] A. Singh e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_22

231

232

A. Sachdeva et al.

Fig. 1 Vibratory bowl feeder [5]

vibratory bowl feeders, which provide a robust, reliable and versatile solution to the part feeding requirements. They are used when a randomly sorted bulk package of small components must be fed into another machine one-by-one, oriented in a particular direction [3]. Vibratory bowl feeders broadly have two major components: a platform (bowl) and a base. The bowl is filled with sufficient small parts surrounded by a helical metal track. The bowl and track undergo an asymmetric helical vibration that causes parts to move up the track [4] (Fig. 1). The vibratory bowl feeders are usually used to feed one kind of product at a time on an assembly line, whereas the authors believe that the developed system has a novelty to provide the following features: 1. The developed system has the ability to feed a pre-selected type of product out of different types placed in the same, thereby making the system flexible. 2. The system has been developed with a compact, low-cost sensing and sorting unit directly coupled to the feeder, with no elaborated external mechanism accessories required. In the present work, a compact and intelligent microcontroller-based sorting and identification system have been integrated with the vibratory bowl feeder. The system consists of sensing devices such as IR and colour recognition sensors and a servobased actuating mechanism which ensures that only the parts of the selected colour and orientation are fed to the delivery chute. The complete setup of the system is shown in Fig. 2. As far as the developed system is concerned, it is a modified vibratory bowl feeder with in-built smart sensing and sorting system making it flexible. To the best of knowledge of the authors, no similar types of systems are available to feed small parts discretely with flexibility. Although a few sensing and sorting systems

Development of a Colour and Orientation …

233

Fig. 2 Experimental setup

are available, but they are quite elaborated with many linkages and moving parts which probably puts a question mark on their speed of operation and reliability.

2 Experimental Setup The experimental setup consists of a vibratory bowl feeder whose track has been slightly modified near the delivery chute to accommodate the developed identification and sorting system. The role of the major components in the system is explained under the following headings.

2.1 Colour Sensor The colour sensor constantly emits red, green and blue colour intensity signals to the Arduino microcontroller. As per the coloured part kept in front of it, the intensity of the above three basic colours keeps changing. By applying an appropriate algorithm on the data received from the sensor through an Arduino microcontroller, different colours combinations can be detected. In the present work, the colour sensor was used to sort out the red coloured parts from others.

234

A. Sachdeva et al.

2.2 Infrared Sensor The IR sensor has an infrared LED acting as an emitter which emits infrared light. This infrared light after reflecting from a surface gets detected by the receiver which further transmits the signal to the microcontroller. The intensity of the signal depends upon the closeness of the reflecting surface from the receiver. The detection proximity of the sensor can be adjusted using a potentiometer. In present work, the inverted caps were separated from the straight caps as the intensity of the signal received in case of inverted caps is weaker than those of straight caps, as the distance between the reflecting surface of the cap and the receiver increases in case of inverted cap.

2.3 Servo-Based Actuator The servo-based actuator has been used because of its high accuracy and precision which enables it to sort out even the small parts without causing disturbance to the neighbouring parts. The servo receives PWM signals from the microcontroller according to which it aligns its rotor at the specific angles. In present work, the rotating head was modified by attaching a thin metal strip placed adjacent to the wall of the feeder in its normal position, thus not creating hindrance in the smooth movement of the desirable parts.

2.4 Microcontroller Arduino Uno board which consists of Atmega328P microcontroller has been used as the processing unit of the present work. It constantly monitors the readings from the colour sensor and infrared sensor, and further processes the procured readings to take the final decision whether to kick the part or not and accordingly sends the signal to the actuator (Fig. 3).

3 Description of Algorithm Used in Code 3.1 Mean of a Set of 10 Readings of Colour Sensor to Reduce the Error for(q = 0;q < 10;q ++) { // Setting red filtered photodiodes to be read digitalWrite(S2,LOW);

Development of a Colour and Orientation …

Fig. 3 Circuit diagram

digitalWrite(S3,LOW); red = pulseIn(sensorOut, LOW); delay(10); // Setting Green filtered photodiodes to be read digitalWrite(S2,HIGH); digitalWrite(S3,HIGH); green = pulseIn(sensorOut, LOW); delay(10); // Setting Blue filtered photodiodes to be read digitalWrite(S2,LOW); digitalWrite(S3,HIGH); blue = pulseIn(sensorOut, LOW); delay(10); ravg = ravg + red; gavg = gavg + green; bavg = bavg + blue; } ravg = ravg/10; bavg = bavg/10; gavg = gavg/10;

3.2 Identification of the Colour from the Mean RGB Values if(((ravg >= 90)&&(ravg =118)&&(gavg = 99)&&(bavg = 111)&&(ravg =112)&&(gavg = 107)&&(bavg = 72)&&(ravg =86)&&(gavg = 66)&&(bavg 120) {c = 1; Serial.println(“the cap is straight”);} y = 0; }

3.4 Condition for Rejection of the Undesirable Part by the Actuator if((col ==1)||(col ==2)||(c ==1)) // col ==1 indicates color is blue { Serial.println(“the cap must be kicked”); // col ==1 indicates color is green // c ==1 indicates cap is straight for (pos = 85; pos = 85; pos - = 1) { //servo backward movement // in steps of 1 degree myservo.write(pos); delay(5); }

Development of a Colour and Orientation …

237

4 Working The parts fed into the bowl feeder rotate around the helical pathway by the virtue of vibration in the feeder, caused by the electromagnets. The parts are randomly oriented and are of different colours. As the parts reach the detection platform with the sensors installed, every single part is checked for its current orientation and colour. This information is then sent to the microcontroller, which computes the input data and sends the appropriate signal to the actuator as per the program. In present work, different coloured caps were used as the subject specimens. Only the red coloured inverted caps were allowed to pass through the inspection system. Rest all other caps were rejected successfully.

5 Conclusion The developed system has successfully been tested over a multiple runs and has proved to be accurate with 2% error. Arduino has proved to be a useful tool in making such low-cost identification systems, which can find immense utility in industry with minor certain modifications.

References 1. Deshmukh, T.: Performance analysis of a vibratory bowl feeder. Int. J. Curr. Eng. Technol. 7(5), 1741–1744 (2017) 2. Dai, J.: Force analysis of a vibratory bowl feeder for automatic assembly. J. Mech. Des. 127, 637–645 (2005) 3. Groover, M.P.: Automation, Production Systems, and Computer-Integrated Manufacturing, 2nd edn. Prentice Hall of India Pvt. Ltd., New Delhi 4. Berretty, R., Goldberg, K., Cheung, L., Overmars, M., Smith, G., Stappen, F.: Trap design for vibratory bowl feeders. In: IEEE International Conference on Robotics and Automation, May 1999 5. Pandey, T., Garg, V., Bhagat, S., Khanna, P.: Mathematical analysis of vibratory bowl feeder. Int. J. Latest Trends Eng. Technol. (IJLTET) 4, 315–324

Comparative Analysis on Battery Used in Solar Refrigerated E-Rickshaw in India Surender Kumar and Rabinder Singh Bharj

Abstract Solar refrigerated hybrid E-rickshaws have gained more popularity due to zero carbon emission, less energy consumption, comfortable and light transport in the last few years which used in congested Indian cities for urban transport. Most of these E-rickshaws were used four to six lead-acid batteries. The weight of these batteries was lies between 110 and 200 kg which was highly effects on the driving range of E-rickshaws. Then, the driving range of E-rickshaw was not more than 110 km and which required to replace in 1.5 years to maintaining driving range. Erickshaw used lithium-ion battery which was facing difficulties in terms of accurate health due to various internal and external factors. Lithium-ion battery bank used for E-rickshaw was lighter in weight and longer driving range provided as compared to the lead-acid battery bank. This research paper focuses on all issues and challenges to using lithium-ion battery bank for E-rickshaw with possible solutions which include technological development in the state of health and remaining useful life for current and future.

1 Introduction India’s automobile sector is fourth-ranked in worldwide. More than 80% of the loads are distributed by road transportation in India. Road transport plays a significant role in the GDP of India. India has nearly 1.5 million (15 lakh) E-rickshaws and nearly 11 thousand new E-rickshaws are entering each month. Total electric vehicle sale in the market worldwide is shown in Fig. 1. Indian mobility systems are facing sustainabil-

S. Kumar (B) · R. S. Bharj Mechanical Engineering Department, NIT, Jalandhar, Punjab, India e-mail: [email protected] R. S. Bharj e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_23

239

240

S. Kumar and R. S. Bharj

Fig. 1 Total electric vehicle sales in the last few years [4]

Fig. 2 Recent investment announcements for EV infrastructure development in top EV sales countries [5]

ity problems in the smart city. Recent investment announcements for electric vehicle (EV) infrastructure development in top EV sales countries are shown in Fig. 2 [1]. Current problems are a large amount of fuel consumption, greenhouse gas emissions and fuel price continuous rising have attracted attention on energy storage devices for present and future E-rickshaw in the automobile sector. Today the research community engaging in profuse efforts to achieve effective energy storage method for solar hybrid refrigerated E-rickshaws (SHRERs) [2]. These sources of renewable energies can be useful for next-generation SHRERs. Currently, 96% of Indian E-rickshaws are operated by lead-acid battery bank having four to six lead-acid batteries. These lead-acid batteries are preferred in E-rickshaws because of its lower cost and easily available in the local market. These batteries having small life cycle (500–800) and higher in weight, in this regard, are directly effects on driving range of SHRERs. These batteries required 6–8 h in one time complete charging and consume 5–7 unit electricity. These lead-acid batteries are not fit for long run application in SHRER as well as E-rickshaws. Li-ion batteries have a fast charging rate, long cycle life (1000–5000), high power density and low self-discharge rate as compared to leadacid batteries. Battery management system (BMS) plays an important role in health monitoring Li-ion batteries but it not required in the lead-acid battery bank. BMS

Comparative Analysis on Battery Used in Solar …

241

Fig. 3 BMS function in SHRERs and E-rickshaw [3]

controls overcharging, overheating, battery uniformity and battery fault diagnosis. BMS functions in SHRERs are shown in Fig. 3. BMS provides more depth of discharge of Li-ion battery as compared to the lead-acid battery bank. The purpose of the present analysis is to focus on the advantages of Li-ion battery use for SHRERs and E-rickshaw and study its economic and environmental impact.

2 Smart Mobility System for Indian Megacities Megacity population in India growing sharply in the last few decades which provides huge strains on urban transport systems. Megacity citizens are faced many problems during his trip and goods transportation which are related to the security of his family or goods [2]. Most of the time urban citizens utilizes three-wheelers and erickshaws for traveling short distances. E-rickshaws and SHRERs provide door to door services in megacity at a lower cost of transportation. E-rickshaw uses a GPS tracking system and electronic road pricing system to show customer goods delivery time, location and its delivery charges [6]. E-rickshaws and SHRERs need charging station which is situated near the metro stations. These battery-operated electric vehicles are ecofriendly working and help to control pollution in urban areas. Smart mobility system for Indian megacities is shown in Fig. 4.

242

S. Kumar and R. S. Bharj

Fig. 4 Smart mobility system for Indian megacities [2, 6]

3 BMS Duty for Li-Ion Battery Use in E-Rickshaw and SHRERs • To maintain the battery operation as per requirements of E-rickshaw and SHRERs. • To protect and control the damages inside battery bank and cells. • To provide the proper voltage and temperature interval which helps to increase battery service life as long as possible. • To stop overcharging, antipole and over discharging in the battery bank. • To help in smoke detection, insulation detection and impedance detection in battery working time. • To estimation of battery states such as includes state of charge (SOC), depth of discharge (DOD), state of health (SOH) and state of function (SOF). • To facilities on-board diagnosis (OBD) which checks sensor, actuator, network, battery, overvoltage, under voltage, over current, ultra high or low temperature, loose connection and insulation fault. • To indicate safety and beep sound alarm in any kind of fault deduction. • It helps to control charge for equalizing charging among all cells of the battery. • BMS maintains thermal management of battery to start and stop heating or cooling according to charge or discharge of battery bank requirement. • Controller area network (CAN) uses in BMS to show vehicle data online. • BMS provides facility to store vehicle data such as SOC, SOH, charge, discharge current and voltage, fault code, the uniformity which is important for past vehicle performance check. The relation among the state of charge (SOC), state of health (SOH) and state of function (SOF) are shown in Fig. 5.

Comparative Analysis on Battery Used in Solar …

243

Fig. 5 Working process of a battery management system for SHRERs and E-rickshaw [7]

4 The Relation Between State of Function with SOC and SOH State of charge (SOC) is explained in such a way that how the battery differs from a fully charged battery. State of health (SOH) is explained in such a way that how the battery differs from a fresh battery which is determined by service life prediction and fault diagnosis output together. State of function (SOF) describes in such a way how the battery performance meets the real demands. SOF defines as: Remaining available energy in the battery Maximum possible energy could be stored in the battery P − Pdemands = Pmax − Pdemands

SOF =

where P Means of possible power the battery could supply. Pdemands Means the demands of power. Means the maximum possible supplied power of the battery. Pmax

(1)

244

S. Kumar and R. S. Bharj

Fig. 6 Relationship between the state of function (SOF) with SOC and SOH [3]

SOF is determined with the help of SOC, SOH, operating temperature, fault level and the charge or discharge history. The relationship among the SOF and SOC, SOH are shown in Fig. 6 [7, 8]. SOC/DOD measurement has a big concern for all battery derived E-rickshaw and SHERs. It gives information about the remaining useful energy and usable time which is directly estimated with working current, temperature and voltage. The selfbalancing property of the connection of the parallel cells was found in Li-ion battery module [9]. Thus, its SOC could be just estimated just like the single cell. SOH is estimated directly with performance degradation and misuse of batteries. When SOF equals to zero its means battery does not completely power demand for vehicle motors. All motors load are considered (includes vehicle motor as well as air conditioners) during the estimation of SOF [10] (Tables 1, 2 and 3).

5 Conclusion E-rickshaws and SHRERs have played an important role in passengers as well as freight transportation in an Indian megacity. Most of them operated on 4–6 lead-acid battery which required to replace after 1.5 years (500 cycles). Lead-acid battery bank weight was lies between 120 and 180 kg which was highly effects on vehicle driving range. These electric vehicles do not cover a distance of more than 95–120 km. Liion one or two batteries were provided suitable power for 1000–1500 W motor in

Comparative Analysis on Battery Used in Solar …

245

Table 1 SOC estimation techniques comparison use for E-rickshaw and SHRERs [1, 7, 11] Name of technique

Parameter required

Benefit of technique

The weakness of the technique

Discharge test technique

• Remaining charge • Capacity

Easy to use and accurate

The long time required, offline, a lot of energy loss

Ampere-hour integral technique

• • • •

Practical, fast, accurate if the initial SOC value. Current measurement and efficiency is precise. Easy to implement

Depends on the initial SOC value and open loop Requires precise value of the self-discharge rate and coulomb efficiency. Sensitive to the current sensor precision Not suitable for batteries under very dynamic conditions

Open-circuit voltage technique

• Rest time • Voltage

Simple, easy to implement, accurate

Requires certain rest time. Open loop Only suitable when SOC is very high or low Sensitive for voltage sensor precision

Battery model-based SOC estimation technique

• Current • Voltage • Battery model

Needs no rest time, insensitive of the initial, SOC value

Its measure noise. Only suitable when SOC is very high or low

Neural network model

• • • •

Fit for all types of batteries, generic, accuracy depends on the type of model use Enormous computational difficulty

Requires a large amount and quality of training data High accuracy Models are becoming simpler

Fuzzy logic

• Current • Voltage

Generic The capability of self-learning

Not accurate. Sensitive to the amount and quality of training data. Accuracy depends on the type of model use

Coulomb efficiency Self-discharge rate Current capacity Initial SOC value

Current Voltage Cumulative charge Initial SOC

(continued)

246

S. Kumar and R. S. Bharj

Table 1 (continued) Name of technique

Parameter required

Benefit of technique

The weakness of the technique

Resistance-based techniques

• Resistance



AC impedance—hard and cost. DC resistance—not so accurate

Weighted fusion algorithm

• Current and voltage • Capacity and coulomb efficiency • Battery model

Ampere-hour integral technique, battery model-based SOC estimation technique

Lot of computation data required. Unpredictability if the weight factor is not suitable

Kalman Filter

• Current and voltage • Capacity and coulomb efficiency • Battery model

Closed loop, online, accurate, dynamic. Error in bounds

Percentage of computation complicated instability if the gain is undesirable. Highly depends on model accuracy

Sliding mode observer

• Current and voltage • Capacity and coulomb efficiency • Battery model

Accurate, robustness, closed loop, online, insensitive of the noise, model error and the initial SOC value error

Nonlinear More computationally expensive than non-feedback techniques, and highly dependent on the model accuracy Not easy to implement

E-rickshaws and shares. Li-ion battery life cycle found six to eight times more as compared to lead-acid batteries bank. Li-ion was performed well with the SOC, SOH and SOF value. Li-ion battery weight was 4–5 times less for same power output in E-rickshaw or SHRERs. The initial cost of Li-ion batteries was 3–4 times more but the cost was overcome due to his long life. Charging time of Li-ion battery was 3 times less which consumed 7–9 units in one time full charging. BMS was imparted a significant role for long life in the Li-ion battery.

Comparative Analysis on Battery Used in Solar …

247

Table 2 Existing SOH estimation techniques use in E-rickshaw and SHRERs [1, 7, 11] Technique name

Main parameter

Benefit of technique

The weakness of the technique

Durability model with open-loop technique

Durability

Comprehensive understanding

Correct input requires

Durability

Easy to predict, capacity diminishes and internal resistance rise

A large number of experiments required

DC resistance

Simple and easy to use

Accuracy level is low. Sensitive to disturbances

Battery model based on parameter identification with closed-loop technique

AC impedance

Precision

Difficult

Extend Kalman filter

Accurate

Depend on model accuracy level

Fuzzy logic

Accurate and simple

Slow convergence

Sample entropy

Simple

Experimental data required more

Discharge voltage

Easy to use

Less accurate

Adaptive control system

Online use

More sensitive to modeling precision

Table 3 Comparison between Li-ion versus lead-acid battery for E-rickshaw and SHRERs [12] Battery specification

Lead-acid type

Li-ion phosphate type

No. battery used

4 battery connected in series

One or two

Capacity (Ah)

100

100

Normal voltage

12 V * 4 = 48 V

51.2 V

Dimension (L * W * H)

41 * 17.6 * 24.5 cm3 for single piece

57.8 * 41 * 21 cm3

Weight

30 * 4 = 120 kg

62 kg

Cycle life

600 @ 50% DOD

3000 @ 80% DOD

State of charge (%)

50

80

Energy density (Wh/L)

90

250

Specific energy (Wh/kg)

40

150

Initial cost ($/kWh)

120

600

Efficiency

70% @ 10 h rate

99% @ 10 h rate

BMS status

No

Yes

Self-discharge

less than 3% per month

Less than 3% per month

Internal resistance

Approx 4.9 m

50 m @ 50% SOC

Total battery bank cost

25,000–30,000/- Rs. INR

Near 100,000/- Rs. INR

248

S. Kumar and R. S. Bharj

References 1. Berecibar, M., Gandiaga, I., Villarreal, I., Omar, N., Van Mierlo, J., Van den Bossche, P.: Critical review of the state of health estimation methods of Li-ion batteries for real applications. Renew. Sustain. Energy Rev. 56, 572–587 (2016) 2. Kumar, S., Bharj, R.S.: Energy consumption of solar hybrid 48 V operated mini mobile cold storage. IOP Conf. Ser. Mater. Sci. Eng. 455, 1–10 (2018). ISSN: 1757–8981 3. Li, W., Zeng, L., Wu, Y., Yu, Y.: Nanostructured electrode materials for lithium-ion and sodiumion batteries via electrospinning. Sci. China Mater. 59(4), 287–321 (2016) 4. https://auto.economictimes.indiatimes.com/news/passenger-vehicle/cars/indias-electricitydemand-from-evs-may-reach-69-6-twh-by-2030-study/64584576. Retrieved on 21/02/2019 5. https://www.hdfcbank.com/assets/pdf/privatebanking/ThematicNote-AutoElectricVehicle240118.pdf. Retrieved on 23/02/2019 6. Kumar, S., Bharj, R.S.: Emerging composite material used in a current electric vehicle. Mater Today Proc 5(14P2), 27946–27954 (2018). ISSN: 2214–7853 7. Hannan, M.A., Lipu, M.H., Hussain, A., Mohamed, A.: A review of lithium-ion battery state of charge estimation and management system in electric vehicle applications: Challenges and recommendations. Renew. Sustain. Energy Rev. 78, 834–854 (2017) 8. Harks, P.P.R.M.L., Mulder, F.M., Notten, P.H.L.: In situ methods for Li-ion battery research: a review of recent developments. J Power Sour. 288, 92–105 (2015) 9. Schipper, F., Auerbach, D.: A brief review: past, present, and future of lithium-ion batteries. Russ J Electrochem 52(12), 1095–1121 (2016) 10. https://ceb8596f236225acd0078e95328c173a04ed694af83ee4e24c15.ssl.cf5.rackcdn.com/ docs/product/Relion-Data-Sheet-RB48V100.pdf. Retrieved on 26/09/2017 11. Jaguemont, J., Boulon, L., Dubé, Y.: A comprehensive review of lithium-ion batteries used in hybrid and electric vehicles at cold temperatures. Appl. Energy 164, 99–114 (2016) 12. https://docs-emea.rs-online.com/webdocs/0f59/0900766b80f5939b.pdf. Retrieved on 26/09/2017 13. Place, T., Kloepsch, R., Duehnen, S., Winter, M.: Lithium-ion, lithium metal, and alternative rechargeable battery technologies: the odyssey for high energy density. J. Solid State Electrochem. 21(7), 1939–1964 (2017) 14. Zuo, X., Zhu, J., Müller-Buschbaum, P., Cheng, Y.J.: Silicon-based lithium-ion battery anodes: a chronicle perspective review. Nano Energy 31, 113–143 (2017) 15. Abada, S., Marlair, G., Lecocq, A., Petit, M., Sauvant-Moynot, V., Huet, F.: Safety focused modeling of lithium-ion batteries: a review. J. Power Sources 306, 178–192 (2016)

Application of Value Stream Mapping (VSM) in a Sewing Line for Improving Overall Equipment Effectiveness (OEE): A Case Study Shawkat Imam Shakil and Mahmud Parvez

Abstract Value stream mapping (VSM) has been successfully applied to improve OEE and other performance parameters in a sewing line of VIP Industries Ltd. Improvement scopes have been identified from the current state map, and rough set theory has been adopted to identify focused areas and improvement strategy regarding where and how lean control should be approached. After that, two new layouts have been designed. Different performance parameters are calculated for both of the layouts and compared with the existing layout. New layout 2 exhibited the most promising outcome and is selected as the proposed layout. Production lead time, inventory and processing time are visualized in the future state VSM based on the proposed layout. Finally, OEE improved from 45 to 53.75%. This methodology is suitable for manufacturing environment having process layout and achieves the goals of reducing waste and improving productivity and performance.

1 Introduction 1.1 General Nowadays, lean manufacturing is a quite familiar term in many organizations in Bangladesh and the organizations consider lean manufacturing as an important initiative in order to secure their competitive market positions. Lean has many tools and philosophy, and the suitability of their application depends on the specific operations and/or processes performed by the organizations. This research aimed at improving the overall equipment effectiveness (OEE) of a sewing line in VIP Industries BD Pvt. Ltd. by applying VSM and Rough Set Theory. Value stream mapping assists S. I. Shakil (B) · M. Parvez Khulna University of Engineering and Technology, Khulna 9203, Bangladesh e-mail: [email protected] M. Parvez e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_24

249

250

S. I. Shakil and M. Parvez

in thinking out the whole production method, illustrating both flow of material and information which provides a means of improving an enterprise where it is applied. Value stream refers to a collection of activities that are necessary to bring a product/service or a group of product/service from input resource to the end customer [1]. In addition to VSM, Rough Set Theory is adopted in the sewing line in order to identify the core attribute, eliminate redundant attributes and generate reduct rules. Outcome of the rough set theory is in the form of some if-then rules which will direct in the development of the future state VSM.

1.2 Objectives • • • •

To draw and analyze current state map for the sewing line. To apply rough set theory in order to get specific directions. To identify and propose potential avenues for improving current level of VSM. To determine the extent of OEE as an assessment of improvement after the application of proposed VSM.

2 Literature Review The available literature on VSM, which is published in international journals, conferences and books, is classified according to the classification scheme given by Garg and Deshmukh [2].

2.1 Case Studies A case study was carried out with the help of value stream management in a computer manufacturing company, and the problems were solved. To find the reasons for the problems, current status of the company was shown in a map and an improved status was proposed in future state map eliminating the problems [3]. Using current state map, the procedure of workflows was explained throughout the different stages of production. Analyzing the existing workflows, some changes were proposed in the future state map [4]. Another case study in a steel industry shows that using VSM the cost of production can be reduced. Through the removal of inventory, a significant amount of cost was proposed to be reduced in the future state [5]. The result of a case study in a small-volume manufacturing industry shows that using VSM technique cycle time reduced by 33.18%, changeover time reduced by 81.5%, production lead time reduced by 81.4% and value-added time reduced by 1.41% [6]. From the conclusion of many case studies, it is found that VSM can be used as a redesign tool [7].

Application of Value Stream Mapping (VSM) in a Sewing Line …

251

2.2 Conceptual Work The activities which add value to the product or service are mapped and shown with their flow which is termed as value stream [8]. The rate of value addition from the raw material to the final product and the delivery to the final customer is mentioned for a single product [9]. A classification scheme is given specifying three types of operations: value adding, non-value adding and necessary non-value adding [10]. VSM has to be performed as the first step toward lean approach among the five theories of lean manufacturing [11]. For mapping a specific product family, practical issues like necessary calculation, placement of supermarket, types of flow and types of operations are needed to be explained [12].

2.3 Modeling Work VSM was applied to British Telecommunication PLC, and a simulation model was developed based on the proposed state for showing the improvements [13]. Using only static view given by the VSM, a simulation model was developed and results obtained for the current and future state were explained [14]. In a factory of Taiwan, rough set theory is applied after the development of current state map which helped identifying the focus areas to go for improvement. After that, a simulation model was developed using Arena in order to show the efficacy of the proposed scenario [15]. In Orient Handbag Limited, a simulation model was developed and visualized its production system in a simulated environment which is subjected to high variety in batch [16].

2.4 Survey Articles In a study, VSM was applied to provider network for the distribution of different mechanical elements, electrical parts and components and represented the effectiveness. For the betterment of the program drawing specifying existing operations are made and improvement opportunities were identified [17]. The network improvement team visited nine major US aerospace organizations to collect data on the tools used for product development, process development, lean approach, improvement of the efforts by various survey events. The study concluded that good tools and associated lean approaches are significant factors of success. The outcome of the study provides a VSM methodology for product and process development. The methodology provides a modified way of process mapping through some tools which facilitates examining and improving the process [18].

252

S. I. Shakil and M. Parvez

3 Methodology 3.1 Current State VSM Current state map is prepared according to the ongoing status of the sewing line. The current state map collects information at a specific instance. For the sake of analysis, the variation in shift and operator performance is not considered. Also, the data for inventory between workstations are taken as snapshot in time. Figure 1 shows the current state map specifying all the processes, their processing time, inventory, production lead time and total processing time. The current state map shows that the variations in the processing time result in huge amount of inventory between some stations and that consequently result in longer production lead time.

3.2 Rough Set Theory Data recorded from the production floor, for example, processing time, transportation time, number of workstations, distance, work in process (WIP), are aggregated and considered as features of production. Then, rough set theory is used to analyze the reduct rules to identify the improvement scope. In rough set theory, a reduct refers to a subset which is sufficient and has the same ability to discern concepts as when the full set of attributes is used [19] (Tables 1, 2, 3 and 4). To find the core, each attribute is removed one at a time and tested to see if this creates indiscernibility. F 1 is the only attribute which is part of the core. Considerations for rule generation: • First, try to make single attribute rules by combining an input attribute with the output attribute. Here, one thing should be kept in mind that the input attribute must be part of the core while we are trying to make single attribute rules. • When it is not possible to make consistent rules from single attribute, combine two or more attributes to make rules. In this case, at least one input attribute must be part of the core. • Try to make consistent rules with 100 percent confidence and higher support. Rule 1: If the level of workstation is 2 and WIP is 0, then the level of waste is 2. Rule 2: If the level of workstation is 1 and WIP is 0, then the level of waste is 2. Rule 3: If the level of workstation is 0 and WIP is 2, then the level of waste is 0. Rule 4: If the level of workstation is 1 and WIP is 1, then the level of waste is 1. Rule 5: If the level of workstation is 0 and WIP is 0, then the level of waste is 0. Hence, rule 5 shows most favorable outcome with lower processing time (PT) and non-value-added time (NVAT) as well as higher machine utilization (MU) and line efficiency (E) (Table 5).

Application of Value Stream Mapping (VSM) in a Sewing Line …

Fig. 1 Current state VSM

253

254

S. I. Shakil and M. Parvez

Table 1 Data set for objects [15] Object No.

Object

No. of workstations used

Process time (s)

Changeover time (s)

Percentage of changeover

Distance (m)

1

Front part

9

434

60

13.84

18

2

Top part

7

323

120

37.15

14

3

U part

7

353.34

120

33.96

13.5

4

Bottom part

3

163

60

36.8

16

5

Side Toton part

3

282.33

60

21.25

21.5

6

Handle to back part

4

213.67

60

28.08

27

7

Assembly

3

174

120

68.96

5.5

Table 2 Input and output features with level definition [15] Level

F 1 works stations

F2 changeover percentage

F 3 distance (m)

F 4 WIP

O output (waste)

0

25%

>20

th3

Simulation of Underground Cable Defects with the Detection …

Fig. 6 Electric field norm component for external cut defect

Fig. 7 Electric field at cable termination

505

506

H. Delvadiya et al.

Fig. 8 Electric field of cable termination and left out completely

E fd = ∫0x Edx where E fd = electric field of defected cable th1 = 1.5 × σ (E) × μ(E) th2 = 2 × σ (E) × μ(E) th3 = 3 × σ (E) × μ(E) where μ(E) = average of electric field over the distance in normal cable, and σ (E) = standard deviation electric field over the distance in normal cable.

4 Conclusion Electrical properties of the power cables at the moment of the various defects were studied and presented in the paper. The defects such as void, external cut or any air particle present in the insulation part of the power cable causes partial discharge that can be damaging for the insulation. In future, analysis will be performed on the acoustic waveform generated from the partial discharge. Because the main advantage of acoustic detection over the conventional detection is that, in acoustics there is no any type of magnetic interference involved in the detection of the defects (i.e. cable defects detection experiment). And the results of that experiment would be analysed with the help of signal processing techniques. This analysis will give us the complete data about the defects.

Simulation of Underground Cable Defects with the Detection …

507

References 1. Cselkó, R., Tamus, Z.Á., Szabó, A., Berta, I.: Comparison of acoustic and electrical partial discharge measurements on cable terminations. In: 2010 IEEE International Symposium on Electrical Insulation pp. 1–5. IEEE) (2010) 2. Ahmed, N.H., Srinivas, N.N.: On-line partial discharge detection in cables. IEEE Trans. Dielectr. Electr. Insul. 5(2), 181–188 (1998) 3. Ye, H., Fechner, T., Lei, X., Luo, Y., Zhou, M., Han, Z., Wang, H., Zhuang, Q., Xu, R., Li, D.: Review on HVDC cable terminations. High Voltage 3(2), 79–89 (2018) 4. Mahdipour, M., Akbari, A., Werle, P.: Charge concept in partial discharge in power cables. IEEE Trans. Dielectr. Electr. Insul. 24(2), 817–825 (2017)

Comparison of C.I Engine Performance Parameters and Emissions by Varying Designs of Intake Manifolds N. Balaji Ganesh and P. V. Srihari

Abstract Fluid motion in compression ignition engine is induced during the induction process and later modified during the compression process. The main problem in compression ignition engine is improper combustion which is due to improper mixing of air and fuel due to shorter delay periods. In conventional engines, air motion is linear and there is no rotational flow of air which leads to improper mixing of air and fuel within the shorter delay periods. In order to enchance proper mixing a secondary motion is to be provided to the air so that it properly mixes with fuel for shorter delay periods which increases engine performance and reduce emissions due to complete combustion, in order to provide secondary motion to the air the design of intake manifold is changed which provides rotational movement to the air instead of linear motion which enhances proper mixing of air and fuel leading to the variation of many engine performance parameters along with emissions. In this work, different designs of intake manifolds are considered and the performance parameters and emissions are calculated and compared with conventional engine.

1 Introduction Heat engine is a device which converts chemical energy of fuel to mechanical work output; in this conversion, efficiency plays key role, and compression ignition is one of the classifications of heat engine in which working fluid is diesel. Compression ignition engines are widely used in transportation because of its high efficiency when compared to with that of spark-ignition engine, but compression ignition engine is also having some disadvantages: one of the main drawbacks is improper combustion which is due to shorter ignition delay period and in order to overcome it, secondary motion to the air is provided which enhances proper mixing of air and fuel. Air N. Balaji Ganesh (B) · P. V. Srihari Department of Mechanical Engineering, RV College of Engineering, Bengaluru, India e-mail: [email protected] P. V. Srihari e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_50

509

510

N. Balaji Ganesh and P. V. Srihari

swirl is created by directing the air flow in intake manifold with different types of internal threads, viz. acme, buttress and knuckle of constant pitch [1], various shaped threaded manifold of different pitches provides swirl and they have proved that the performance characteristics are improved along with reduction in emissions [2]. Solid Works, Gambit and Fluent with optimal geometry design of the inlet manifold assembly have redesigned the inlet port of a small I.C engine in order to have better turbulence by swirl [3]. The swirl effect on the engine is tested by using swirl adapter by making a blade angle of 30° which provides effective swirl at part load conditions [4]. A ring-type generator with four curvilinear blades is fitted in the intake air duct, and the comparison is carried out and found out that with 1500 rpm effective swirl is generated and with reduced emissions [5]. The comparisons of volumetric efficiency with three different configurations, viz. helical, spiral and helical–spiral combination, are done on single cylinder 4-stroke engine, and all the three dimensional models of the manifolds and the cylinder are created and meshed using the pre-processor GAMBIT. The flow characteristics of these engine manifolds are examined under transient conditions using computational fluid dynamics (CFD) code STAR-CD. By comparing all the results, swirl inside the cylinder is more in case of helical–spiral combined manifold then spiral and normal manifold [6]. The orientation of the intake manifold was changed by inclining it at different angles, viz. normal intake manifold and intake manifold at 25° inclination, 50° inclination and 75° inclination, and experimentation is done, from the results manifold inclined at 50° is giving best results when compared to that of remaining inclinations [7]. 3D model of three manifolds with helical, spiral and helical–spiral shapes has been prepared and observations are recorded and all the three are giving better performance than that of normal manifold [8]. Air swirl generated by directing the air flow in intake manifold by grooving the inlet manifold with a helical groove of size of 1 mm width and 2 mm depth of different pitches to direct the airflow and test is carried out among which 8 mm pitch groove gives best results [9]. The flow characteristics inside the engine cylinder equipped with different piston configurations were compared, the piston geometry had little influence on the in-cylinder flow during the intake stroke and the first part of the compression stroke, and the bowl shape plays a significant role near TDC and in the early stage of the expansion stroke by controlling both the ensemble-averaged mean and the turbulence velocity fields [10]. There is a need to change the design of intake manifolds to enhance proper mixing of air and fuel because of shorter delay periods.

2 Experimental Set-Up Figure 1 represents the engine along with various parts incorporated in it and the specifications are listed below.

Comparison of C.I Engine Performance Parameters …

511

Fig. 1 Experimental set-up of engine

2.1 Engine Specifications See Table 1.

3 Designs of Manifolds In this section, different designs of inlet manifolds are discussed by varying the design of inlet manifolds and secondary motion is provided to incoming air which enhances proper mixing of air and fuel. Table 1 Engine specifications

Make

Kirloskar AV1

Ignition system

Compression ignition

Cooling

Water cooled

Bore

88 mm

Stroke

116 mm

Compression ratio

16:1

Speed

1500 rpm

Rated power

5 HP

Fuel

Diesel

Lubricant

SAE 20/SAE 40

512

N. Balaji Ganesh and P. V. Srihari

Fig. 2 Normal manifold

3.1 Normal Manifold It is the normal manifold, in which the core diameter of the manifold is 30 mm and the length of the manifold is 430 mm, and it is manufactured by stainless steel. Initially, the experiment is conducted on the engine with normal manifold and all the performance parameters are calculated along with the emissions (Fig. 2).

3.2 Nozzle Inside the Inlet Manifold The design of intake manifold is modified by placing a nozzle inside the manifold where the outer diameter of nozzle is 29 mm and inner diameter of nozzle is about 15 mm and is manufactured by using stainless steel as material. In this design, some sort of restriction is provided to air passage due to shape of the nozzle but increases the velocity of air (Fig. 3).

3.3 Internal Threaded Manifold The intake manifold of the CI engine was modified with helical threads of pitch 3 mm and the core diameter of the manifold is about 30 mm, by considering the thread outer diameter as 30 mm and inner diameter as about 26 mm provided inside the manifold which is manufactured by stainless steel. In this design also, there is restriction to the passage of air but it provides secondary motion to the air inside the manifold with enhances mixing of air and fuel (Fig. 4).

Comparison of C.I Engine Performance Parameters …

513

Fig. 3 Nozzle inside the inlet manifold

Fig. 4 Internal threaded manifold

4 Tabular Columns for Experimental Results The recorded values obtained from experimentation are further calculated by using standard mechanical formulae and all the values are listed below tabular columns along with emissions (Tables 2, 3 and 4).

5 Results and Discussions The results obtained by conducting performance test on engine by varying inlet manifold design are compared with normal manifold along with emissions. Graphs

514

N. Balaji Ganesh and P. V. Srihari

Table 2 Experimental results with normal manifold S. No.

Description

Units

Trail-I

Trail-II

Trail-III

Trail-IV

Trail-V

1

Spring balance reading 1

kg

0

0

0

0

0

2

Spring balance reading 2

kg

1

3

5

7

9

3

Load

kg

1

3

5

7

9

4

Speed

rpm

1500

1500

1500

1500

1500

5

Manometer reading

mm

45

45

45

45

45

6

Time taken for 10 cc fuel

s

87

81

74

64

56

7

Brake power

kW

0.231

0.693

1.155

1.617

2.08

8

Total fuel consumption

kg/h

0.351

0.371

0.413

0.478

0.546

9

Specific fuel consumption

kg/kWh

1.529

0.535

0.357

0.295

0.2625

10

Heat input

kW

4.1

4.407

4.818

5.578

6.375

11

Friction power

kW

1.9

1.9

1.9

1.9

1.9

12

Indicated power

kW

2.13

2.593

3.054

3.51

3.98

13

Mechanical efficiency

%

10.79

26.72

37.8

46.06

52.26

14

Brake thermal efficiency

%

5.69

15.72

23.97

28.98

32.62

15

Indicated thermal efficiency

%

51.95

58.83

63.4

63.01

62.43

16

Actual air intake

m3 /s

0.00508

0.00508

0.00508

0.00508

0.00508

17

Theoretical air intake

m3 /s

0.00691

0.00691

0.00691

0.00691

0.007

18

Volumetric efficiency

%

73.51

73.51

73.51

73.51

73.51

19

CO

%

0.21

0.24

0.27

0.29

0.35

20

CO2

%

1.4

1.47

1.53

1.57

1.63

21

HC

ppm

20

24

29

32

35

22

O2

%

19.01

18.55

18.4

18.2

18.01

The variation of fuel consumption with load is shown in Fig. 5 in which fuel consumption is less for manifold with internal threads when compared with that of remaining manifold designs which is due to proper mixing of air and fuel, whereas in case of manifold with nozzle the fuel consumption is more due to less quantity of air availability inside the engine due to restriction for passage of air. The variation of brake power with load is shown in Fig. 6 in which brake power remains constant for all the manifold designs because the engine is operated at

Comparison of C.I Engine Performance Parameters …

515

Table 3 Experimental results with internal threads S. No.

Description

Units

Trail-I

Trail-II

Trail-III

Trail-IV

Trail-V

1

Spring balance reading 1

kg

0

0

0

0

0

2

Spring balance reading 2

kg

1

3

5

7

9

3

Load

kg

1

3

5

7

9

4

Speed

Rpm

1500

1500

1500

1500

1500

5

Manometer reading

Mm

47

47

47

47

47

6

Time taken for 10 cc fuel

s

93

88

80

69

64

7

Brake power

kW

0.231

0.693

1.155

1.61

2.08

8

Total fuel consumption

kg/h

0.329

0.3472

0.3825

0.443

0.478

9

Specific fuel consumption

kg/kWh

1.42

0.5

0.331

0.273

0.23

10

Heat input

kW

3.83

4.05

4.46

5.17

5.578

11

Friction power

kW

1.8

1.8

1.8

1.8

1.8

12

Indicated power

kW

2.039

2.493

2.955

3.41

3.88

13

Mechanical efficiency

%

11.72

27.7

39.08

47.2

53.6

14

Brake thermal efficiency

%

6.03

17.18

25.78

31.27

37.16

15

Indicated thermal efficiency

%

53.12

61.5

66.2

65.97

69.57

16

Actual air intake

m3 /s

0.00519

0.00519

0.00519

0.00519

0.00519

17

Theoretical air intake

m3 /s

0.0069

0.0069

0.0069

0.0069

0.0069

18

Volumetric efficiency

%

75.22

75.22

75.22

75.22

75.22

19

CO

%

0.2

0.22

0.25

0.27

0.33

20

CO2

%

1.38

1.42

1.49

1.54

1.59

21

HC

ppm

19

22

27

30

32

22

O2

%

19.1

18.9

18.75

18.5

18.4

constant speed so that fuel consumption will vary which is already discussed in Graph 1. The variation of brake thermal efficiency with load is shown in Fig. 7 in which brake thermal efficiency for manifold with internal threads is more when compared to that of remaining designs because the engine is operated at constant speed by varying the load, and if the load varies, the fuel consumption varies and remaining

516

N. Balaji Ganesh and P. V. Srihari

Table 4 Experimental results with nozzle inside the manifold S. No.

Description

Units

Trail-I

Trail-II

Trail-III

Trail-IV

Trail-V

1

Spring balance reading 1

kg

0

0

0

0

0

2

Spring balance reading 2

kg

1

3

5

7

9

3

Load

kg

1

3

5

7

9

4

Speed

Rpm

1500

1500

1500

1500

1500

5

Manometer reading

Mm

46

46

46

46

46

6

Time taken for 10 cc fuel

s

84

79

72

61

54

7

Brake power

kW

0.231

0.693

1.155

1.61

2.07

8

Total fuel consumption

kg/h

0.364

0.387

0.425

0.501

0.566

9

Specific fuel consumption

kg/kWh

1.575

0.558

0.367

0.311

0.273

10

Heat input

kW

4.242

4.515

4.956

5.838

6.594

11

Friction power

kW

1.8

1.8

1.8

1.8

1.8

12

Indicated power

kW

2.03

2.49

2.95

3.41

3.87

13

Mechanical efficiency

%

11.72

27.7

39.08

47.2

53.6

14

Brake thermal efficiency

%

5.44

15.34

23.3

27.5

31.39

15

Indicated thermal efficiency

%

47.85

55.14

59.72

62.45

58.68

16

Actual air intake

m3 /s

0.00514

0.00514

0.00514

0.00514

0.00514

17

Theoretical air intake

m3 /s

0.00691

0.00691

0.00691

0.00691

0.00691

18

Volumetric efficiency

%

74.38

74.38

74.38

74.38

74.38

19

CO

%

0.23

0.25

0.28

0.32

0.37

20

CO2

%

1.47

1.5

1.57

1.62

1.66

21

HC

ppm

22

25

30

33

36

22

O2

%

18.8

18.4

18.1

18

17.9

parameters remain constant so brake thermal efficiency completely depends on mass of the fuel consumed which is less in manifold with internal threads. The variation of mechanical efficiency with load is shown in Fig. 8 in which mechanical efficiency remains almost constant for all the manifolds at various loads due to constant brake power for all the loads. The variation of volumetric efficiency with load is shown in Fig. 9 in which volumetric efficiency remains almost same which is in the range of 75% because the

Comparison of C.I Engine Performance Parameters …

517

0.60

Variable

Normal Manifold Manifold with Internal Threads Manifold with Nozzle

TFC in kg/hr

0.55 0.50 0.45 0.40 0.35 0.30 0

1

2

3

4

5

6

7

8

9

LOAD in kg

Fig. 5 Load versus total fuel consumption

LOAD VS BRAKE POWER Variable Normal Manifold

2.0

Manifold with Internal Threads Manifold with Nozzle

B.P in kW

1.5

1.0

0.5

0.0 0

1

2

3

4

5

6

7

8

9

LOAD in kg Fig. 6 Load versus brake power

engine is operated at atmospheric conditions but in microscopic approach slight increase in volumetric efficiency for the manifold with internal threads due to proper mixing of air and fuel which creates slightly more suction when compared to remaining cases. The variation of CO2 in the engine exhaust gases is shown in Fig. 10 in which concentration of CO2 almost remains same but there slight reduction is observed in

518

N. Balaji Ganesh and P. V. Srihari

LOAD VS Brake Thermal Efficiency 40

Variable Normal Manifold

Brake thermal Efiiciency in %

35

Manifold With Internal Threads Manifold with Nozzle

30 25 20 15 10 5 0

1

2

3

4

5

6

7

8

9

LOAD in kg

Fig. 7 Load versus brake thermal efficiency

LOAD VS Mechanical Efficiency 60

Variable

Mechanical Efficiency in%

NormalManifold Manifold with Internal Threads

50

Manifold with Nozzle

40

30

20

10 0

1

2

3

4

5

LOAD in kg

6

Fig. 8 Load versus mechanical efficiency

7

8

9

Comparison of C.I Engine Performance Parameters …

519

LOAD VS Volumetric Efficiency 75.25

Variable

Normal Manifold Manifold with Internal Threads

Volumetric Efficiency in %

75.00

Manifold with Nozzle

74.75 74.50 74.25 74.00 73.75 73.50 0

1

2

3

4

5

6

7

8

9

LOAD in kg Fig. 9 Load versus volumetric efficiency LOAD VS CO2 Variable

Normal Manifold

1.65

Manifold with Internal Threads Manifold with Nozzle

CO2 IN %

1.60

1.55

1.50

1.45

1.40

0

1

2

3

4

5

6

LOAD in kg Fig. 10 Load versus CO2

7

8

9

520

N. Balaji Ganesh and P. V. Srihari LOAD VS CO 0.38

Variable Normal Manifold

0.36

Manifold with Internal Threads Manifold with Nozzle

0.34

C O in %

0.32 0.30 0.28 0.26 0.24 0.22 0.20 0

1

2

3

4

5

6

7

8

9

Load in kg

Fig. 11 Load versus CO

case of manifold with internal threads which is due to proper mixing of air and fuel, where is in manifold with nozzle due to lack of air formation of CO2 is more when compared to that of remaining two manifold designs. The variation of CO in the engine exhaust gases is shown in Fig. 11 in which concentration of CO is less in case of manifold with internal threads which is due to proper combustion where is in manifold with nozzle due to lack of air improper combustion takes place due to which more concentration of CO is formed. The variations of HC in the engine exhaust gases is shown in Fig. 12 in which concentration of HC is less in case of manifold with internal threads due to proper mixing of air and fuel which reduces formation of unburnt hydrocarbons, whereas in manifold with nozzle due to improper combustion, unburnt fuel is present inside the engine due to which HC is formed. The amount of oxygen content present in the exhaust gases is shown in Fig. 13 in which the amount of oxygen present is more in case of manifold with internal threads which indicates combustion is more effective, whereas in manifold with nozzle less amount of oxygen is present due to improper combustion.

6 Conclusions The performance parameters of an engine are calculated by varying designs of intake manifold and compared with normal manifold. By comparing all the results, the following conclusions are made

Comparison of C.I Engine Performance Parameters …

521

37.5

Variable

35.0

Manifold with Internal Threads

Normal Manifold Manifold with Nozzle

HC in ppm

32.5 30.0 27.5 25.0 22.5 20.0

0

1

2

3

4

5

6

7

8

9

LOAD in kg

Fig. 12 Load versus HC

19.2

Variable

19.0

Normal Manifold Manifold with Internal Threads Manifold with Nozzle

O2 in %

18.8 18.6 18.4 18.2 18.0

0

1

2

3

4

5

6

7

8

9

LOAD in kg

Fig. 13 Load versus O2

• Fuel consumption is less in case of manifolds with internal threads when compared with that of normal manifold and manifold with nozzle, where are as fuel consumption for manifold with nozzle is more due to restricted passage of air. • Brake power remains constant for all the designs of manifold because the engine is operated at constant speed. • Mechanical efficiency and volumetric efficiency remain constant for all the designs of manifolds due to constant friction power and inlet pressure remains constant. • Brake thermal efficiency is more for manifold with internal threads when compared to that of remaining designs due to lesser fuel consumption.

522

N. Balaji Ganesh and P. V. Srihari

• Emissions such as HC, CO2 and CO are slightly lower for manifold with internal threads which provides some secondary motion to the air inside the manifold which enhances proper mixing proper mixing of air and fuel so that wastage of fuel is slightly reduces in turn reduces the emissions and increases oxygen content in emissions which is an indication of better combustion. • Emissions such as HC, CO2 and CO are slightly higher for manifold with nozzle which is due to restriction of air passage at the inlet which lacks proper mixing of air and fuel increases emissions and reduces oxygen content at the exhaust.

References 1. Shrirao, P.N., Sambhe, R.U.: Effect of swirl induction by internally threaded inlet manifolds on exhaust emissions of single cylinder (DI) diesel engine. Int. J. Sci. Res. (2014) 2. Phaneendra, V., Pandurangadu, V., Chandramouli, M.: Performance evaluation of a four stroke compression ignition engine with various helical threaded intake manifolds. Int. J. Appl. Res. Mech. Eng. (2012) 3. Martins, J., Teixeira, S., Coene, S.: Design of an inlet track of a small I. C. engine for swirl enhancement. In: 20th International Congress of Mechanical Engineering, Gramado, RS, Brazil (2009) 4. Mohiuddin, A.K.M.: Investigation of the swirl effect on engine using designed swirl adapter. IIUM Eng. J. (2011) 5. Shenghua, L.: Development of new swirl system and its effect on DI diesel engine economy. SAE International (1999) 6. Paul, B., Ganesan, V.: Flow field development in a direct injection diesel engine with different manifolds. Int. J. Eng. Sci. Technol. (2010) 7. Shah, J.V., Patel, P.D.: Experimental analysis of single cylinder 4-stroke diesel engine for the performance and emission characteristics at different inclinations of the intake manifold. Int. J. Sci. Res. Dev. (2014) 8. Ramakrishna Reddy, P.R., Rajulu, K.G., Naidu, T.V.S.: Experimental investigation on diesel engines by swirl induction with different manifolds. Int. J. Curr. Eng. Technol. (2014) 9. Prasad, S.L.V., Pandurangadu, V.: Reduction of emissions by intensifying air swirl in a single cylinder di diesel engine with modified inlet manifold. Int. J. Appl. Eng. Technol. (2013) 10. Payri, F., Benajes, J., Margot, X., Gil, A.: CFD modeling of the in-cylinder flow in direct injection diesel engines. Elsevier Comput. Fluids 33 (2004)

Variation of Time Lag, Decrement Factor and Inside Surface Temperature with Solar Optical Properties of Building Envelope in Different Climatic Zones of India Debasish Mahapatra and T. P. Ashok Babu Abstract Maintenance of thermal comfort inside buildings requires a significant amount of energy. According to the Centre for Science and Environment, the energy spent on achieving thermal comfort in commercial buildings and in residential buildings are 31% and 7%, respectively. Considering the energy crisis that world has been suffering; every individual should try to save energy by some means. The increase in urbanization over the last few years gave rise to boom in constructions. Choosing the materials used for the construction of buildings wisely can contribute towards energy saving. The solar optical properties of building envelope affect the surface temperature up to a great extent which in turn affects the energy used for thermal comfort. In this paper, the effect of solar optical properties on time lag decrement factor and inside surface temperature is studied in different Indian climatic zones.

1 Introduction Better employment opportunities, health and education facilities, the standard of living and social status attract people to migrate towards urban areas. Urbanization helps in social, economic and political progress but on the other hand it leads to socioeconomic and environmental problems because of the lack of unplanned growth in urban population. The heat released from various human activities, vehicular emission, air conditioner, industries get accumulated and create an island of heat called Urban Heat Island (UHI). The city center temperature in USA was found to be 5.60 °C higher because of UHI [1]. In India, urbanization is happening very rapidly. In 1901, the percentage of people residing in urban areas was only 11.4%. It increased to 28.53% in 2001 and the number reached 31.16% in 2011 [2]. So, D. Mahapatra (B) · T. P. Ashok Babu Department of Mechanical Engineering, National Institute of Technology Surathkal, Surathkal, Karnataka 575025, India e-mail: [email protected] T. P. Ashok Babu e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_51

523

524

D. Mahapatra and T. P. Ashok Babu

taking into account the rate at which urbanization is taking place, in India utmost care should be taken to avoid the creation of UHI. Use of cool coatings on the building envelope is one of the practices which can be adapted to avoid creation of UHI. Cool coatings are basically characterized by their high reflectivity. The high reflectivity of the coating keeps the surface temperature of the envelope low and doesn’t allow the temperature gradient to raise much thereby preventing the heat transfer into the building. The study of Synfa et al. [3] found that an energy-saving up to 75% can be achieved by increasing the solar reflectance of the roof from 0.2 to 0.6. The inside and outside wall temperature was measured by Hernandez-Perez et al. [4] for the reflectance of 0.80, 0.84 and 0.33. The surface temperature and heat flux were found to be decreased with increase in solar reflectance. The outside surface temperatures were 30.5, 23.70 °C, and 38.30–400 °C for coatings having solar reflectance of 0.80, 0.84, and 0.33, respectively. Solar reflectance of 0.80, 0.84, and 0.33 resulted in inside surface temperature of 250 °C, 240 °C, 310 °C to 330 °C, respectively. Shen et al. [5] found the effect of reflectance coating on surface temperature and energy saving on both conditioned and non-conditioned building by experimentation. The maximum decrease in exterior surface temperature was 19.90 °C for non-conditioned buildings. In west direction, the maximum surface temperature reduction was found.

1.1 Time Lag and Decrement Factor Thermal mass of the wall doesn’t allow the maximum outside and inside temperature to happen at the same time [6]. The amplitude of the heatwave outside and inside also varies because of the thermal mass. So, the time difference between maximum outside and inside temperature is called time lag and the ratio of difference of maximum and minimum temperature inside to outside is called decrement factor [7] (Fig. 1). Mathematically, ϕ = tTomax − tTemax

(1)

And f =

Tomax − Tomin Ao = Ae Temax − Temin

(2)

The concept of time lag and decrement factor should be taken into consideration while designing buildings in the area where there is a large variation of temperature throughout the day. If we use a material having such a time lag that it allows the

Variation of Time Lag, Decrement Factor …

525

Fig. 1 Schematic presentation of time lag and decrement factor. Source Asan [7]

maximum temperature to occur at a time when the building is not in use then we can able to cool the building passively thereby saving energy.

2 Methodology 2.1 Sole Air Temperature The equation governing heat transfer from ambient to the surface of the wall is Eq. 3. q0 = f 0 (t0 − tso ) + a I (3) An equivalent temperature is obtained by combining the effect of solar radiation and ambient temperature and is called sole air temperature. q0 = f 0 (te − tso )

(4)

where, te = t0 +

aI f0

(5)

526

D. Mahapatra and T. P. Ashok Babu

Fig. 2 Heat transfer from atmosphere to outside surface of a building wall

From the above equation, it can be noticed that absorptivity plays an important role in heat transfer into the building as it affects the surface temperature of the wall and reduced absorptivity can help in reducing the heat transfer (Fig. 2).

2.2 Climatic Zones of India Hot and dry, warm and humid, composite, moderate, cold and sunny and cold and cloudy are the five major categories of climate of India. Time lag and decrement factor were calculated for Pune (moderate), Mangalore (warm and humid), Ahmadabad (hot and dry), Hyderabad (composite) on peak summer day.

2.3 Time Lag and Decrement Factor Calculation The one-dimensional heat conduction equation through wall is given by 1 ∂t ∂ 2t = ∂ 2x α ∂τ

(6)

where ‘α’ is the thermal diffusivity defined as α=

k ρCp

(7)

Variation of Time Lag, Decrement Factor …

527

And ‘k’ is the thermal conductivity and ρCp is the heat capacity of the wall material. The one-dimensional heat transfer equation was solved by finite difference method by considering nodes at the surface. Hourly outside temperature, inside temperature, i.e. room temperature, incident solar radiation, absorptivity of the surface, outside heat transfer coefficient, inside heat transfer coefficient, thickness of the wall and thermal conductivity of the wall are the parameters required for solving one-dimensional heat transfer equation. The outside air temperature was taken from the book ISHRAE Indian weather data 2017 [8]. Room temperature was maintained at 25 °C to get thermal comfort. The incident solar radiation which is the sum of direct and diffuse radiation was calculated by using the necessary formula. Different colours such as red, yellow, green, brown and black were considered for this study. The reflectance value for these colours was taken from literature [4, 9–11]. The absorptivity was calculated by considering the transmittance as zero. The absorptivities obtained were 0.2, 0.25 and 0.65 for white, yellow and brown, respectively. The absorptivity for black and green was 0.73.

3 Results and Discussion 3.1 Time Lag and Decrement Factor Calculation 3.1.1

Time Lag and Decrement Factor Calculation for Hot and Dry Climatic Zone (Ahmadabad)

Time lag, Decrement factor and inside surface temperature of the wall were calculated for Ahmadabad on peak summer day, i.e. on May 15th by varying the absorptivity. Time lag varied randomly with absorptivity. Decrement factor and inside surface temperature increased with increase in absorptivity. Decrement factor and inside surface temperature were the highest in west direction and the least in south direction (Fig. 3).

3.1.2

Time Lag and Decrement Factor Calculation for Composite Climatic Zone (Hyderabad)

Time lag, decrement factor and inside surface temperature for composite climatic zone (Hyderabad) was calculated on peak summer day, i.e. on May 15th. Variation of time lag and decrement factor with absorptivity of the wall was analyzed and it was found that time lag varied randomly. The decrement factor increased with increase in absorptivity. The decrement factor was the highest in west direction and lowest

528

D. Mahapatra and T. P. Ashok Babu

Fig. 3 a Decrement factor b excess inside surface temperature c time lag for different absorptivity in hot and dry climatic zone (Ahmedabad)

in south direction, for all absorptivity value of wall. Inside surface temperature of the wall, which majorly affects the cooling load was also found. It was noticed that inside surface temperature varied same as decrement factor. For all the cases, the inside surface temperature was the highest for wall facing west and the lowest for wall facing south (Fig. 4).

3.1.3

Time Lag and Decrement Factor Calculation for Warm and Humid Climatic Zone (Mangalore)

On peak summer day, i.e. on April 21st the time lag, decrement factor and inside surface temperature for warm and humid climatic zone (Mangalore) were calculated by varying the absorptivity of the envelope. Time lag varied randomly with absorptivity. Decrement factor and inside surface temperature increased with increase in absorptivity. Wall facing west was having the highest decrement factor and inside surface temperature and wall facing south was having the least decrement factor and inside surface temperature (Fig. 5).

Variation of Time Lag, Decrement Factor …

529

Fig. 4 a Decrement factor b excess inside surface temperature c time lag for different absorptivity in composite climatic zone (Hyderabad)

Fig. 5 a Decrement factor b excess inside surface temperature c time lag for different absorptivity in warm and humid climatic zone (Mangalore)

530

D. Mahapatra and T. P. Ashok Babu

Fig. 6 a Decrement factor b excess inside surface temperature c time lag for different absorptivity in moderate climatic zone (Pune)

3.1.4

Time Lag and Decrement Factor Calculation for Moderate Climatic Zone (Pune)

Variation of time lag, decrement factor, and inside surface temperature with absorptivity was analyzed for moderate climatic zone on peak summer day i.e. on May 15th. Time lag varied randomly with absorptivity. The decrement factor and inside surface temperature were found to be the highest on wall facing west and the lowest on wall facing south (Fig. 6).

3.2 Electricity Saving For the calculation of electricity saving the average load was calculated in west direction for 0.2 and 0.73 absorptivity. The wall area is considered as 100 m2 . The working hour of air-conditioner was assumed as 6 h per day. The electricity bill for each city was taken from [12] (Table 1).

Variation of Time Lag, Decrement Factor …

531

Table 1 Cost saving in percentage for different cities Total unit (kW-h) = (average load*6*30)/1000

Cost (per 100 m2 of wall area)

City

Average load (Watt)

% Saving

0.2 Abs

0.73 Abs

0.2 Abs

0.73 Abs

0.2 Abs

0.73 Abs

Ahmadabad

2463

3213

443.34

578.34

2353

3146

25.2

Hyderabad

1527

2380

274.86

428.4

1597

2888

44.7

Mangalore

1074

1983

193.32

356.94

1161

2449

52.59

Pune

1442

2287

259.56

411.66

2224

4114

45.94

4 Conclusions From the analysis of variation of time lag and decrement factor with absorptivity of the wall the following conclusions can be drawn. • Time lag varies randomly with absorptivity, this may be because of the dependence of time lag on thermal mass of the wall. • Decrement factor increases with increase in absorptivity. In all the climatic zones studied decrement factor was the highest for wall facing west and the lowest for the wall facing south. In the south direction a minimum of 8% change (Hyderabad) and maximum of 28% change (Mangalore) in decrement factor was observed. In west direction the minimum change was observed in Pune, i.e. 52% and maximum change were observed in Mangalore, i.e. 79%. All the changes are calculated by changing the absorptivity from 0.2 to 0.73. • Inside surface temperature showed the same variation as decrement factor. The value of inside surface temperature was the highest in west-facing wall and the lowest for south-facing wall in all the climatic zones studied. Change of absorptivity from 0.2 to 0.73 results in a minimum of 17% (Ahmadabad) and maximum of 28% (Mangalore) change in inside surface temperature in south direction. Similarly the same change in absorptivity results in minimum of 50% (Ahmadabad) and maximum of 75% (Mangalore) change inside surface temperature in west direction. • Cooling load of a building is significantly affected by decrement factor and inside surface temperature of the building. So, the absorptivity should always be kept low in order to reduce the cooling load thereby saving energy and cost. Cost-saving up to 53% can be achieved by changing the absorptivity from 0.73 to 0.2. Wall facing west should be taken care a lot as the decrement factor and inside temperature are the highest for it.

532

D. Mahapatra and T. P. Ashok Babu

References 1. Valsson, S., Bharat, A.: Urban heat island: cause for micro climate variation. Arch.—Time Space People, 20–25 (2009) 2. Urbanization in India. https://en.wikipedia.org/wiki/Urbanisation_in_India. Last accessed 26 Mar 2019 3. Synnefa, A., Santamouris, M., Akbari, H.: Estimating the effect of using cool coatings on energy loads and thermal comfort in residential buildings in various climatic conditions. Energy Build. 39(11), 1167–1174 (2007) 4. Hernandez-Perez, I., Xaman, J., Macias-Melo, E.V., Aguilar-Castro, K.M., Zavala-Guillen, I., Hernandez-Lopez, I., Sima, E.: Experimental thermal evaluation of building roofs with conventional and reflective coatings. Energy Build. 158, 569–579 (2018) 5. Shen, H., Tan, H., Tzempelikos, A.: The effect of reflective coatings on building surface temperatures, indoor environment and energy consumption. Energy Build. 43(2–3), 573–580 (2011) 6. Arora, C.P.: Refrigeration and air conditioning, 3rd edn. Tata McGraw-Hill, New Delhi (2009) 7. Asan, H.: Investigation of wall’s optimum insulation position from maximum time lag and minimum decrement factor point of view. Energy Build. 32, 197–203 (2000) 8. ISHRAE: Indian weather data society of heating, refrigerating and air conditioning engineers. India (2017) 9. Sameera, S., Rao, P.P., Divya, S., Raj, A.K.V., Thara, T.R.A.: High IR reflecting BiVO4 – CaMoO4 based yellow pigments for cool roof applications. Energy and Build. 154, 491–498 (2017) 10. Synnefa, A., Santamouris, M., Apostolakis, K.: On the development optical properties of cool coloured coatings for urban environment. Sol. Energy 81(4), 488–497 (2007) 11. Thongkanluang, T., Chirakanphaisarn, N., Limsuwan, P.: Preparation of NIR reflective brown pigment. Procedia Eng. 32, 895–901 (2012) 12. Online electricity bill calculator—for all states in India. https://www.bijlibachao.com/ electricity-bill/online-electricity-bill-calculator-for-all-states-in-india.html. Last accessed 6 May 2019

Study on Tensile and Hardness Properties of Aluminium 7075 Alloy Reinforced with Graphite, Mica and E-Glass T. G. Gangadhar, D. P. Girish, A. C. Prapul Chandra, Gangadhar Angadi and K. V. Karthik Raj

Abstract The development of metal matrix composites is of major interest in industrial applications due to some important advantages like strength-to-weight ratio, heat resistance and chemical stability. In the present experimental investigation work, aluminium-based hybrid metal matrix composites of various percentage compositions of graphite (1, 3, 5 and 7 wt%), E-glass (1,2,3 and 4 wt%) and mica (2 and 4 wt%). Al7075/graphite/E-glass/mica metal matrix composite fabricated by stir-casting process, and specimens are prepared as per ASTM standards. Experiments were conducted as per L16 orthogonal array. Mechanical properties like tensile strength, hardness and microstructure of metal matrix optimized by S/N ratio and ANOVA. The regression model is executed for both responses, and the most influencing factor is determined by S/N ratio. The results indicate that the graphite filler enhances the tensile strength and hardness properties. The properties of tensile strength and hardness showed improvement for 5 wt% graphite, 4 wt% E-glass and 2 wt% mica.

T. G. Gangadhar (B) · K. V. K. Raj Department of Mechanical Engineering, SJBIT, Bengaluru, India e-mail: [email protected] K. V. K. Raj e-mail: [email protected] D. P. Girish Department of Mechanical Engineering, GEC, Ramanagaram, India e-mail: [email protected] A. C. P. Chandra · G. Angadi Department of Mechanical Engineering, R. V. College of Engineering, Bengaluru, India e-mail: [email protected] G. Angadi e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_52

533

534

T. G. Gangadhar et al.

1 Introduction Composites are the materials made from two or more constituents which are insoluble to each other with different properties in both physical manner and chemical manner and due to combination which produces a material with characteristics different from the individual components. The two constituents are matrix and reinforcement; matrix is having binding nature which binds or holds the fibre materials/reinforced material. The obtained new materials may be preferred for the all fields for many reasons like strength, low cost, lighter, etc., when compared to traditional base materials. Nowadays, composite materials are commonly used for constructing buildings, structures of bridges, automotive, locomotives, aircrafts, etc. Effect of addition of Al2 O3 and graphite in base aluminium material on hardness and tensile properties of the hybrid composites increases with the increase in filler material percentage up to some level [1]. The use of graphite reinforcement in a metal matrix has a potential to create a material with a high thermal conductivity, excellent mechanical properties, and attractive damping behaviour at elevated temperatures [2, 3]. However, lack of wettability between aluminium and the reinforcement, and oxidation of the graphite lead to manufacturing difficulties and cavitation of the material at high temperatures [4]. Although it is concluded that Al7075-SiC composites exhibit higher tensile strength and hardness values to the margin of 10% as compared to Al6061–Al2 O3 composite system [5]. The elongation and strength decrease with enhancing percentage of silicon carbide and graphite particles added to the matrix of aluminium. The graphite has a high negative impact on the elongation of the composites than the silicon carbide particles due to weaker graphite/matrix interface [6]. Anwar Khanet et al. discussed that Al7075–bagasse ash–graphite hybrid composite specimens were prepared using stir-casting technique. Hardness of composites was increased due to enhancing filler in the base alloy. The 95% base metal with 5% filler material has shown high BHN, and it can be seen that ductility of the composites material decreased with increasing filler material in the matrix alloy [7]. The main objectives of the present work is to synthesize Al7075/graphite/Eglass/mica composites by stir-casting technique and study the influence of filler (graphite, E-glass and mica) particulates reinforcement on mechanical (hardness and tensile strength) and microstructure (optical) of Al7075 metal matrix composite.

2 Materials and Process 2.1 Materials Al7075, graphite, E-glass and mica were purchased from M/s Fen Fee Metallurgical, Bangalore, Karnataka, India. The stir-casting technique facility was used in R. V. college of Engineering, Bengaluru. The hardness and tensile tests were measured in

Study on Tensile and Hardness Properties …

535

Raghavendra Spectro Metallurgical Laboratory, Peenya Industrial Area, Bangalore, as per ASTM standard.

2.2 Process The Al7075 hybrid composite is prepared by using stir-casting technique. This approach involves mixing of the reinforcement (graphite, E-glass and mica) particulate into a molten metal bath and transferring the mixture directly into a cylindershaped die to complete solidification. In this technique, aluminium alloy 7075 ingot pieces are to be heated in the furnace to its molten state. When the temperature is maintained between 800 and 8500 °C, a vortex will be created using a mechanical stirrer. Different wt% of graphite, E-glass and mica particles are added into the molten Al7075 furnace when stirring is in process. Stirring is continued for about 15 min/500 rpm after addition of filler particles for uniform distribution in the Al7075 melt. Castings are prepared by pouring the melt into preheated moulds of cylindrical shapes. The experiments are conducted based on the design of experiment (DOE) as shown in Table 1. Table 1 Percentage composition of reinforcements in aluminium 7075 alloy

Exp. no.

Graphite (wt%)

E-glass (wt%)

Mica (wt%)

L1

1

1

2

L2

1

2

2

L3

1

3

4

L4

1

4

4

L5

3

1

2

L6

3

2

2

L7

3

3

4

L8

3

4

4

L9

5

1

4

L10

5

2

4

L11

5

3

2

L12

5

4

2

L13

7

1

4

L14

7

2

4

L15

7

3

2

L16

7

4

2

536

T. G. Gangadhar et al.

Fig. 1 ASTM E10-15 hardness specimen

2.3 Hardness Test Hardness tests were carried out according to ASTM E10-15 standards (Fig. 1) using a 10 mm ball diameter indenter and an applied load of 60 kg for 30 s at room temperature. The diameter of the indentation left in the test material is measured with a low powered microscope. The Brinell’s hardness number (BHN) is calculated by dividing the load applied by the surface area of the indentation. The BHN is calculated according to the following formula: BHN =

F   π 2 − D2 D ∗ D − D i 2 

where BHN F D Di

Brinell’s hardness number Imposed load in kg Diameter of the spherical indenter in mm Diameter of the resulting indenter impression in mm.

2.4 Tensile Test Tensile tests were conducted according to ASTM A370 standards by using a computerized universal testing machine (model: TUE-CN-600). For the test, the specimen’s dimensions are 20 mm grip diameter, 50 mm grip length, 36 mm gauge length, radius of arc of 8 mm, inner diameter of 9 mm, and total length 161 mm. Ultimate tensile strength (UTS) was evaluated by the average value of three tests conducted on each composition (Fig. 2).

Study on Tensile and Hardness Properties …

537

Fig. 2 Tensile test specimen

3 Result and Discussion 3.1 Tensile Strength and Hardness UTS and BHN values for Al7075/graphite/E-glass/mica metal matrix hybrid composites with varying graphite, E-glass and mica content are shown in Fig. 3a, b. The mechanical properties of Al7075 metal matrix composites mainly depend on the proper dispersion and proper bonding between the reinforcement with matrix material. From Fig. 3a, b experiment L12 showed that highest ultimate tensile strength of 342 MPa and Brinell’s hardness number of 98. The properties increases upto 5% wt of graphite and increase in garphite above 5% wt the properties decreases due to

UTS (MPa)

(a) 350 250 150

-50

L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 L11 L12 L13 L14 L15 L16

50

Exp. No.

(b) 100 80 60 40 20 0

L1 L2 L3 L4 L5 L6 L7 L8 L9 L10 L11 L12 L13 L14 L15 L16

BHN

Fig. 3 a Ultimate tensile strength and b Brinell’s hardness number for different composition

Exp. No.

538

T. G. Gangadhar et al.

higher percentage of reinforcement of graphite, uneven distribution and poor bonding between the matrixes of reinforced phase. The properties of hybrid composites increases with increase in E-glass but decreses with increase in mica filler.

3.2 Signal-to-Noise Ratio Experimental values of ultimate tensile strength and hardness number of aluminium/graphite/E-glass/mica composite are shown in Table 2. The mean values of S/N ratio for UTS and BHN are 49.24 and 38.36 db. The main effects of the UTS and BHN were shown in Fig. 5 for mean value of S/N ratio. This plot clearly indicated that how the different wt% of fillers such as graphite, E-glass and mica have effect on the response. From the main effect plot (Fig. 4), the HBN and UTS properties increase up to 5 wt% of graphite, and beyond that, the properties drop due to the agglomeration of the filler particle in the aluminium. The addition of E-glass increases the both BHN and UTS, but in case of mica addition, both the properties reduce. From the S/N plot, conclusion is made that the combination of independent factors, 5 wt% of graphite, 4 wt% of E-glass and 2 wt% of mica, produces maximum properties of BHN and UTS. Table 2 Experimental values of UTM and BHN of aluminium/graphite/E-glass/mica composite Exp

Graphite (wt%)

E-glass (wt%)

Mica (wt%)

Hardness number (BHN)

Ultimate tensile strength (MPa)

S/N ratio of BHN

S/N ratio of UTS

L1

1

1

2

67

234.5

36.52

47.40

L2

1

2

2

72

252

37.14

48.02

L3

1

3

4

73

255.5

37.26

48.14

L4

1

4

4

76

266

37.61

48.49

L5

3

1

2

80

280

38.06

48.94

L6

3

2

2

81

283.5

38.16

49.05

L7

3

3

4

82

287

38.27

49.15

L8

3

4

4

83

290.5

38.38

49.26

L9

5

1

4

90

315

39.08

49.96

L10

5

2

4

91

318.5

39.18

50.06

L11

5

3

2

94

329

39.46

50.34

L12

5

4

2

98

343

39.82

50.70

L13

7

1

4

84

294

38.48

49.36

L14

7

2

4

85

297.5

38.58

49.46

L15

7

3

2

87

304.5

38.79

49.67

L16

7

4

2

89

311.5

38.98

49.86

Study on Tensile and Hardness Properties …

539 Main Effects Plot for Ultimate Tensile Strength (MPa) Data Means

Main Effects Plot for Hardness Number (BHN) Data Means E-Glass (wt.%)

Graphite (wt.%)

95

E-Glass (wt.%)

Graphite (wt.%)

330

Mica (wt.%)

Mica (wt.%)

320 310 Mean

Mean

90 85

300 290 280

80

270 75

260 250

70 1

3

5

7

1

2

3

4

2

1

4

3

5

7

1

2

3

4

2

4

Fig. 4 Mean value of S/N ratio for UTS and BHN

3.3 Analysis of Variance (ANOVA) Taguchi analysis cannot justice and decide effects of specific contents of the filler in composites while percentage contribution of specific filler can be determined by ANOVA. ANOVA module was used to study the effects of fillers (graphite, E-glass and mica) using GRG. P-value (0.000) of graphite filler indicates the addition of graphite contributing towards UTS and BHN response. From Table 3, according to ANOVA results the graphite (90.47%) has the most significant filler followed by E-glass (8.01%) and mica (0.09%). Table 3 ANOVA results for signal-to-noise ratio for UTS and BHN Response

Source

DF

SS

Contribution (%)

MS

F

P

BHN

Graphite (wt%)

3

954.50

90.47

318.167

169.69

0.000

UTS

E-glass (wt%)

3

84.50

8.01

28.167

15.02

0.001

Mica (wt%)

1

1.00

0.09

1.000

0.53

0.486

Error

8

15.00

1.42

1.875

Total

15

1055.00

100.00

Graphite (wt%)

3

11,692.6

90.47

3897.54

169.69

0.000

E-glass (wt%)

3

1035.1

8.01

345.04

15.02

0.001

0.53

0.486

Mica (wt%)

1

12.3

0.09

12.25

Error

8

183.7

1.42

22.97

Total

15

12,923.7

100.00

DF—degree of freedom, SS—sum of squared deviation, MS—mean squared deviation, F—Fisher’s F ratio, P—probability of significance

540

T. G. Gangadhar et al.

3.4 Microstructure Studies Optical microstructural images of selected Al7075-based metal matrix composites for different wt% of graphite, mica and E-glass casted component are shown in Fig. 5, and Fig. 5a indicated that the filler materials are invisible due to low wt% and mica is distributed uniformly throughout the matrix. The graphite, mica and Eglass are uniformly distributed with Al matrix and no cluster of reinforcement in the microstructure. The fillers of glass fibres are evenly distributed on the grain boundary of matrix (Fig. 5b). Generally, metal matrix composites have a high tendency to agglomerate at grain boundaries during preparing metal matrix composites by stircasting methods [8]. From Fig. 5c, it is clearly visible from the micrographs that there exists good bond between Al7075 matrix alloy and E-glass fibres, graphite and mica particles. Because graphite and mica are present more percentage than E-glass, these two materials are clearly visible than E-glass and are bit randomly distributed. As reported by many researchers, surface tension and wettability are the two important aspects which play vital role in distributing E-glass fibres and mica particles. Adopting best conditions during composite preparations helps in homogeneous dispersion and minimizes clustering of E-glass fibres and mica particles [9]. From

(a) Al 7075+1% Graphite+1% E – Glass+2%Mica

(b) Al 7075+5% Graphite+4% E – Glass+2%Mica

(c) Al 7075+5% Graphite+1% E – Glass+4%Mica

(d) Al 7075+3% Graphite+2%Mica+2% E –Glass

Fig. 5 Optical micrographs of hybrid composites with multiple reinforcement

Study on Tensile and Hardness Properties …

541

Fig. 5d, it is shown that there are no visible defects or porosities associated with reinforced particles indicating good quality of composites.

4 Conclusion Among all the combinations of hybrid composites studied, the highest hardness was recorded for 5 wt% graphite, 2 wt% mica and 4 wt% E-glass fibre of Al7075 metal matrix composites. This combination has exhibited a maximum improvement of 98 BHN. The fundamental reason behind the improvement in the hardness of hybrid composites was the presence of hard reinforcing phase underneath the indenter during Brinell’s hardness test. The presence of hard secondary phase in soft and ductile aluminium matrix generally contributes to significant improvement in the hardness of hybrid composites.

References 1. Baradeswaran, A., Perumal, A.E.: Study on mechanical and wear properties of Al7075/Al2 O3 /graphite hybrid composites. Composites 56, 464–471 (2014) 2. Basavarajappa, S., Chandramohan, G., Paulo Davim, J.: Application of Taguchi techniques to study dry sliding wear behaviour of metal matrix composites. Mater. Des. 28, 1393–1398 (2007) 3. Baskaran, S., Anandakrishnan, V., Duraiselvam, M.: Investigations on dry sliding wear behavior of in situ casted AA7075–TiC metal matrix composites by using Taguchi technique. Mater. Des. 60, 184–192 (2014) 4. Natarajan, N., AnandhaMoorthy, A., Sivakumar, R., Manojkumar, M., Suresh, M.: Dry sliding wear and mechanical behavior of aluminium/fly ash/graphite hybrid metal matrix composite using Taguchi method. Int. J. Mod. Eng. Res. 2, 1224–1230 (2012) 5. Balaji, V., Sateesh, N., ManzoorHussain, M.: Manufacture of aluminum metal matrix composite (Al7075-SiC) by Stir Casting technique. Mater. Today: Proc. 2, 3403–3408 (2015) 6. Cheng-jin, H.U., Hong-ge, Y.A.N., Ji-hua, C.H.E.N., Bin, S.U.: Microstructures and mechanical properties of 2024Al/Gr/SiC hybrid composites fabricated by vacuum hot pressing. Trans. Nonferrous Metals Soc. China 26, 1259–1268 (2016) 7. Imran, M., Khan, ARA., Megeri, S., Sadik, S.: Study of hardness and tensile strength of Aluminium-7075 percentage varying reinforced with graphite and bagasse-ash composites. Resour. Effic. Technol. 4, 2405 (2016) 8. Du, X.M., Zheng, K.F., Zhao, T., Liu, F.G.: Fabrication and characterization of al 7075 hybrid composite reinforced with graphene and sic nanoparticles by powder metallurgy. Dig. J. Nanomater. Biostructures 13, 1133–1140 (2018) 9. Wieczorek, J.: Surface tension of Cu–Bi alloys and wettability in a liquid alloy—refractory material—gaseous phase system. Arch. Metall. Mater. 29, 59–63 (2014)

ANN-Based Fault Classification and Section Identification Technique Using Superimposed Currents for Three-Terminal Transmission Line Shivani Vaghela, Nishant Kothari and Dinesh Kumar

Abstract This paper presents artificial neural network (ANN)-based faulty section identification and fault classification technique on three-terminal power transmission lines. Tapped point of electrical power transmission lines has more complex network with protective circuits. Faulty section identification and fault classification are challenging task on a three-terminal transmission lines. The superimposed currents which are difference of similar phase currents of different terminals is utilized for the classification and section identification on a 400 kV Indian three-terminal power transmission system simulated in PSCAD/EMTDC software. The superimposed current based feature is given to the ANN-based classifiers. The performance of proposed technique is studied by changing various fault and system parameters. The accuracy achieved with the proposed method is more than 99% for section identification and fault classification.

1 Introduction The three-terminal transmission lines have economic, technical, and environmental advantages over two-terminal power transmission lines. The protection of threeterminal transmission lines present more complex challenges compared to single and double circuit line protection. In this context, fault detection and fault classification are two major parts of power transmission line protection. Over many years, researchers have been investigating fault detection and classification methods, thus the faulted system can be protected from possible destructive effects caused by the fault. S. Vaghela (B) · N. Kothari · D. Kumar Marwadi Education Foundation’s Group of Institutions, Rajkot, India e-mail: [email protected] N. Kothari e-mail: [email protected] D. Kumar e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_53

543

544

S. Vaghela et al.

The problem of fault identification and classification is largely dependent on feature extraction from the voltage and current waveforms and the significance of the feature with the physical events [1]. In this direction, signals transformation from time domain to frequency domain such as discrete Fourier transform (DFT), fast Fourier transform (FFT) are frequently used [2, 3]. Discrete wavelet transform (DWT) is advanced tool for characterization of the fault signals. It gives information about frequency as well as time of specific frequency transients generated due to fault [4]. Number of features extracted for fault classification using S-transform which reveals local spectral characteristics are used in [5, 6]. Later to feature extraction, the classification of the fault is performed. A thresholdbased approach on the extracted feature is found in [7]. In these methods, a threshold or a series of if-then conditions with present threshold are set [7]. Other methods include fuzzy logic-based methods such as fuzzy-neuro and adaptive-network-based fuzzy inference systems in [8, 9]. Artificial neural networks (ANNs) including feed forward neural network (FNN), radial basis function network (RBFN) are extensively used to classify different types of faults [10, 11]. ANN-based approaches have demonstrated higher accuracy compared to the other soft computing methods because of supervised way of learning at the time of model building. This paper introduces a superimposed current based feature, terminal current difference from the same phases, which is used as an input to ANN classifier for fault detection and fault classification on three-terminal lines. One cycle post-fault data including one hundred samples are given to ANN classifier for section identification and subsequently fault classification process. Various combinations of training and testing data sets are experimented in order to obtain higher classification accuracy. The proposed feature has been tested using instantaneous and RMS currents. The comparative analysis has been presented in the result section. The structure of the rest of the paper is as follows. The system modelling and generation of fault cases is given in Sect. 2. Feature description for section identification and fault classification is given in Sect. 3. The results of section identification and fault classification are presented in Sect. 4. Conclusion of this research work is given in Sect. 5.

2 System Modelling A 400 kV, three-terminal transmission line with source at each end is modelled using PSCAD/EMTDC 4.2 [12]. The single-line diagram of the system under study is shown in Fig. 1. The line parameters and source parameters are taken from [13]. Table 1 shows line length for each line section. Three terminals are marked as Sending (S), Receiving (R) and Tapped (T ) in the single-line diagram of three-terminal transmission line. Various fault cases are generated by varying fault resistance, fault type, load angle, fault inception angle and fault location as mentioned in Table 2. For each line section, total 6480 fault cases are generated. Considering three terminals S, R and T, total of 19,440 fault cases are generated for fault classification

ANN-Based Fault Classification and Section …

545

Fig. 1 400 kV three-terminal transmission line diagram

Table 1 Transmission line data

Terminal

Line length

400 kV transmission line parameters

Table 2 Variable parameters

S-Terminal

160 km

R-Terminal

280 km

T-Terminal

250 km

No.

Parameters

Value

1

Fault resistance in  (Rf)

1, 5, 10, 20, 30, 40

2

Fault type (ftype)

AG, BG, CG, ABG, ACG, BCG, ABC, AB AC, BC

3

Load angle delta (δ) (°)

20°, 25°, 30°

4

Fault inception angle (FIA) (°)

0°, 30°, 45°, 60°, 90°, 115°

5

Fault location (% of total line length) (F km)

15, 30, 45, 60, 90

Total cases

6480 cases

For each line 6480 cases; for three line 6480 × 3 = 19,440 cases

and section identification task. As the proposed feature is tested using instantaneous as well as RMS currents, fundamental frequency components of RMS currents are obtained using Fast Fourier Transform (FFT).

546

S. Vaghela et al.

3 Feature Selection for Section Identification and Fault Classification The artificial neural network consists of input layer, hidden layer and output layer. The performance of the ANN-based classifier is tested with different hidden layers and no. of neurons. A typical feed forward back propagation algorithm is used to construct ANN [14]. The superimposed RMS current based feature given to ANN classifier is shown in Fig. 2 for healthy system and faulty system for AG fault on S, R and T-terminals.

Fig. 2 Superimposed RMS currents for healthy system and for A–G fault on a S, b R and c Tterminal

ANN-Based Fault Classification and Section …

547

RMS and instantaneous currents of three terminals are obtained using synchronized measurements from each terminal for section identification and fault classification. Superimposed current based feature is obtained using RMS and instantaneous current samples as shown in Eq. (1). IAsr = Ias − Iar IBsr = Ibs − Ibr ICsr = Ics − Icr IArt = Iar − Iat IBrt = Ibr − Ibt ICrt = Icr − Ict IAst = Ias − Iat IBst = Ibs − Ibt ICst = Ics − Ict

(1)

where, I as , I bs , I cs , I ar , I br , I cr , I at , I bt , I ct are RMS/instantaneous currents of threeterminal lines. a, b and c indicate phases while s, r and t indicate terminals. I Asr , I Ast and I Art are phase-A superimposed currents of s and r, s and t, r and t terminals, respectively. Similarly, superimposed currents are obtained for B and C phases also. Superimposed currents are obtained using sample to sample differential method. Zero sequence current is used to detect presence of ground in the fault. Figure 2a, b and c shows proposed feature for A–G fault on S, R and T terminal and for healthy system, respectively. Flowchart depicts section identification and fault classification of three-terminal lines using ANN classifier (Fig. 3). Superimposed current based feature applied on instantaneous and RMS currents is given to ANN classifier as explained in flowchart.

4 Results and Discussion In this section, results of the proposed fault classification and section identification technique are presented. The percentage accuracy of the section identification for the faults on S-terminal, R-terminal and T-terminal with different training and testing data sets are as follow. Table 3 shows results with 20% training, 5% validation and 75% testing data sets. Similarly, Table 4 shows 30% training, 5% validation and 65% testing data results. After section identification, fault classification over faulted terminal is done using proposed feature. Percentage accuracy is computed with different combination of hidden layers and neurons. Their results are specified in the below charts. Figures 4, 5 and 6 show fault classification results for S, R and T terminal, respectively, with 20% training, 5% validation and 75% testing data. Figures 7, 8 and 9 reveals fault classification results for S, R and T terminals, respectively, with 30% training, 5% validation and 65% testing data. With higher training data, the accuracy for section and fault classification achieved is higher. Comparing instantaneous and RMS current based feature, it is observed that the instantaneous current based input feature gives better accuracy. With an increase in the hidden layer and no. of neurons, the percentage accuracy has been increased in both terminal identification and fault classification. The increment in computation time with higher hidden layers is comparatively small. Also,

548

S. Vaghela et al.

Fig. 3 Flowchart of section identification and fault classification process

ANN classifier yields better accuracy in section identification and fault classification for superimposed currents obtained using instantaneous current compared to RMS currents. Also, similar analysis was carried out using raw current input in which percentage accuracy obtained for section identification and fault classification is 78%. By applying this feature, overall accuracy achieved is more than 99% for section identification and fault classification.

5 Conclusion In this paper, section identification and fault classification are performed using ANNbased classifier with superimposed currents as inputs. The proposed techniques are tested by varying system and fault parameters like fault resistance, fault type, load

ANN-Based Fault Classification and Section …

549

Table 3 Percentage accuracy of section identification using 20% training, 5% validation and 75% testing data Hidden layer

2 Hidden layer [10 10]

3 Hidden layer [10 15 10]

Terminal

Test accuracy in (%), train and test time in seconds Superimposed current obtained using RMS current

Superimposed current obtained using instantaneous current

Phase A

Phase B

Phase C

Phase A

Phase B

Phase C

S Terminal

99.6

100

99.4

99.9

99.1

99.4

R Terminal

99.2

99.3

99.4

98.2

99

99.4

T Terminal

99.9

99.1

99.9

99.5

99.4

99.9

Over All

99.5

99.5

99.8

99.8

99.1

99.5

Train time

5.415

5.853

5.928

6.913

4.992

6.724

Test time

0.462

0.458

0.466

0.61

0.46

0.465

S Terminal

99.4

99.6

99.9

100

99.8

100

R Terminal

99.4

99.2

99.8

99.4

99.3

99.3

T Terminal

99.9

99.9

99.3

99.3

99.7

99.1

Over All

99.5

99.5

99.8

99.9

99.7

99.5

Train time

5.651

5.754

5.913

6.206

7.555

8.39

Test time

0.522

0.516

0.517

0.504

0.501

0.483

Table 4 Percentage accuracy of section identification using 30% training, 5% validation and 65% testing data Hidden layer

2 Hidden layer [10 10]

3 Hidden layer [10 15 10]

Terminal

Percentage accuracy in (%), train and test time in seconds Superimposed current obtained using RMS current

Superimposed current obtained using instantaneous current

Phase A

Phase B

Phase C

Phase A

Phase B

Phase C 98.8

S Terminal

100

99.8

99.2

99.9

99.3

R Terminal

99.2

99

98.1

98.5

98.1

98.3

T Terminal

99.5

99.5

99.7

99.2

99.7

98.7

Over All

99.6

99.4

99.8

99.8

99

98.7

Train time

5.453

6.164

4.63

8.516

7.517

9.831

Test time

0.476

0.464

0.44

0.464

0.472

0.462

S Terminal

99.6

100

99.4

99.9

99.1

99.4

R Terminal

99.2

99.3

99.4

98.2

99

99.4

T Terminal

99.9

99.1

99.9

99.5

99.4

99.9

Over All

99.5

99.5

99.8

99.8

99.1

99.5

Train time

5.945

8.151

4.891

11.99

11.947

10.337

Test time

0.516

0.505

0.508

0.488

0.49

0.564

550

S. Vaghela et al.

100% 100% 100%

100% 100% 100%

100% 100% 100%

99.70% 99.80% 100%

100% 100% 100%

3 Hidden layer [10 15 10]

GROUND

PHASE A

PHASE B

PHASE C

GROUND

95.70%

99.80% 100%

2 Hidden layer [10 10]

99.80% 100% 100%

99.00% 100% 100%

PERCENTAGE ACCURACY

1 Hidden layer [10]

PHASE A

PHASE B

PHASE C

RMS CURRENT

INSTANTANEOUS CURRENT

Fig. 4 Fault classification using 20% training, 5% validation and 75% testing data on S terminal

100% 100% 100%

99.70% 99.80% 99.90%

99.90% 100% 100%

99.80% 100%

3 Hidden layer [10 15 10]

98%

99.50% 99.50% 99.80%

100% 100% 100%

2 Hidden layer [10 10]

99.80% 99.70% 99.90%

99.40% 99.40% 99.90%

PERCENTAGE ACCURACY

1 Hidden layer [10]

PHASE A

PHASE B

PHASE C

GROUND

RMS CURRENT

PHASE A

PHASE B

PHASE C

GROUND

INSTANTANEOUS CURRENT

Fig. 5 Fault classification using 20% training, 5% validation and 75% testing data on R terminal

RMS CURRENT

PHASE A

100% 100% 100%

GROUND

99.70% 99.80% 100%

PHASE C

3 Hidden layer [10 15 10] 100% 100% 100%

100% 100% 100%

PHASE B

99.90% 99.90% 100%

100% 100% 99.90%

PHASE A

2 Hidden layer [10 10]

100% 99.90% 100%

100% 100% 100%

PERCENTAGE ACCURACY

1 Hidden layer [10]

PHASE B

PHASE C

GROUND

INSTANTANEOUS CURRENT

Fig. 6 Fault classification using 20% training, 5% validation and 75% testing data on T terminal

ANN-Based Fault Classification and Section …

100% 100% 100%

99.70% 99.80% 100%

3 Hidden layer [10 15 10] 100% 100% 100%

100% 100% 100%

100% 100% 100%

99.90% 100%

2 Hidden layer [10 10]

95.70%

100% 100% 100%

99.70% 99.60% 99.90%

PERCENTAGE ACCURACY

1 Hidden layer [10]

551

PHASE A

PHASE B

PHASE C

GROUND

RMS CURRENT

PHASE A

PHASE B

PHASE C

GROUND

INSTANTANEOUS CURRENT

Fig. 7 Fault classification using 30% training, 5% validation and 65% testing data on S terminal

PHASE A

100% 100% 100%

100% 100% 100%

3 Hidden layer [10 15 10] 99.90% 100% 100%

99.90% 99.80% 100%

100% 100% 100%

99.90% 100% 100%

99.80% 99.80%

2 Hidden layer [10 10]

99.00%

99.60% 99.80% 99.90%

PERCENTAGE ACCURACY

1 Hidden layer [10]

PHASE B

PHASE C

GROUND

RMS CURRENT

PHASE A

PHASE B

PHASE C

GROUND

INSTANTANEOUS CURRENT

Fig. 8 Fault classification using 30% training, 5% validation and 65% testing data on R terminal

100% 100% 100%

GROUND

100% 100% 100%

100% 100% 100%

PHASE C

100% 100% 100%

100% 100% 100%

PHASE B

3 Hidden layer [10 15 10]

100% 100% 100%

100% 100% 100%

100%

2 Hidden layer [10 10]

PHASE A

PHASE B

PHASE C

GROUND

99.40% 99.40%

PERCENTAGE ACCURACY

1 Hidden layer [10]

PHASE A

RMS CURRENT

INSTANTANEOUS CURRENT

Fig. 9 Fault classification using 30% training, 5% validation and 65% testing data on T terminal

552

S. Vaghela et al.

angle, fault inception angle and distance of fault. The analysis of the proposed technique is carried out using different ANN classifier structures. The accuracy achieved with superimposed current obtained using instantaneous currents is comparatively higher than superimposed current obtained using RMS currents. The overall accuracy achieved with the proposed technique for section identification and fault classification is more than 99%.

References 1. Chen, K., Huang, C., He, J.: Fault detection, classification and location for transmission lines and distribution systems: a review on the methods. High Volt. 1(1), 25–33 (2016) 2. Yu, Sun-Li, Jyh-Cherng, Gu: Removal of decaying DC in current and voltage signals using a modified Fourier Filter Algorithm. IEEE Trans. Power Deliv. 16(3), 372–379 (2001) 3. Hagh, M.T., Razi, K., Taghizadeh, H.: Fault classification and location of power transmission lines using artificial neural network. International Power Engineering Conference (IPEC 2007). IEEE (2007) 4. Bhowmik, P.S., Purkait, P., Bhattacharya, K.: A novel wavelet transform aided neural network based transmission line fault analysis method. Int. J. Electr. Power Energy Syst. 31(5), 213–219 (2009) 5. Samantaray, S.R., Dash, P.K.: Pattern recognition based digital relaying for advanced series compensated line. Int. J. Electr. Power Energy Syst. 30(2), 102–112 (2008) 6. Krishnanand, K.R., Dash, P.K.: A new real-time fast discrete S-transform for cross-differential protection of shunt-compensated power systems. IEEE Trans. Power Deliv. 28(1), 402–410 (2013) 7. Girgis, A.A., Sallam, A.A., El-Din, A.K.: An adaptive protection scheme for advanced series compensated (ASC) transmission lines. IEEE Trans. Power Deliv. 13(2), 414–420 (1998) 8. Das, B., Reddy, J.V.: Fuzzy-logic-based fault classification scheme for digital distance protection. IEEE Trans. Power Deliv. 20(2), 609–616 (2005) 9. Reddy, M.J., Mohanta, D.K.: Adaptive-neuro-fuzzy inference system approach for transmission line fault classification and location incorporating effects of power swings. IET Gener. Transm. Distrib. 2(2), 235–244 (2008) 10. Silva, K.M., Souza, B.A., Brito, N.S.: Fault detection and classification in transmission lines based on wavelet transform and ANN. IEEE Trans. Power Deliv. 21(4), 2058–2063 (2006) 11. Mahanty, R.N., Gupta, P.D.: Application of RBF neural network to fault classification and location in transmission lines. IEE Proc. Gener. Transm. Distrib. 151(2), 201–212 (2004) 12. Muller, C.: User’s Guide on the Use of PSCAD, Manitoba. Manitoba HVDC Research Centre, Canada (2010) 13. Saber, A., Emam, A., Elghazaly, H.: A backup protection technique for three-terminal multisection compound transmission lines. IEEE Trans. Smart Grid 9(6), 5653–5663 (2018) 14. Haykin, S.: Neural networks: a comprehensive foundation. Prentice Hall PTR, 1 Oct 1994

An Analytical Investigation for Combined Pressure-Driven and Electroosmotic Flow Without the Debye–Huckel Approximation Avisankha Dutta and Sudip Simlandi

Abstract In the present work, an analytical solution is presented for a combined pressure-driven electroosmotic flow of a Newtonian liquid within a microchannel between two parallel plates. The electroosmotic flow is considered to be induced by an externally applied electrostatic potential field and a pressure gradient. The no-slip boundary conditions are considered. The electric potential distribution is represented by the Poisson–Boltzmann equation. The Debye–Huckel linear approximation is ignored in the present work to minimize error in results. The reduced form of the Navier–Stokes and the energy equations are considered, respectively, to determine velocity and temperature distributions. Homotopy perturbation method (HPM) is adopted as an analytical tool to solve the nonlinear Poisson–Boltzmann equation for electrical potential distribution without the Debye–Huckel linear approximation. The Navier–Stokes and the energy equations subjected to respective boundary conditions are solved analytically. An expression of C f Re product is obtained solving the Navier–Stokes equation. The results obtained are validated with existing literature and show good agreement. The zeta potential is varied for a particular electrokinetic length, and proposed results are presented graphically. Finally, the Nusselt number is presented varying electrokinetic length for different values of zeta potential. The results demonstrate the influence of the zeta potential on the potential, velocity, temperature distributions, and Nusselt number.

1 Introduction In recent days, the study on microfluidic systems has become an important area of research for various potential applications in biomedical and chemical industry. Biomedical microelectromechanical systems or bioMEMS can accomplish sample A. Dutta · S. Simlandi (B) Department of Mechanical Engineering, Jadavpur University, Kolkata 700032, India e-mail: [email protected] A. Dutta e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_54

553

554

A. Dutta and S. Simlandi

injection, chemical reaction, separation, and detection in a single integrated microfluidic circuit [1]. Various techniques such as thermopneumatic, magnetohydrodynamic, piezoelectric, electrostatic, and electroosmotic pumping have been proposed for fluid delivery. Among these, the electroosmotic pumping is preferred because pumping a liquid through a very small channel requires a very large pressure difference, whereas it does not require any external pump, but needs electrodes to control the flow field. Due to the absence of moving parts and pulsating flows, ease of microfabrication, and a great degree of flow control, the electroosmotic pumping is very popular [2–6]. Therefore, it is important to understand the fundamental characteristics of electroosmotic flow (EOF) through microchannels for optimal design and efficient control and to ensure the reliability and the stability of the microfluidic devices of electroosmotic pumps. When an electrolytic solution is under no flow condition, the ions dissociate. Those ions having charge opposite to that of the surface are attracted by the surface. Thus, two layers of positively and negatively charged ions are formed near the surface which are called electric double layer (EDL). If a pressure gradient and an electric field are applied tangentially along such a charged surface, the ions in the diffuse layer will start moving under the action of the pressure field and body force exerted by the electric field, resulting in pressure-driven EOF [3]. In this connection, few literatures are studied. Masood Khan et al. [5] discussed the dynamics of an EOF in cylindrical domain. The linearized Poisson–Boltzmann equation and the Cauchy momentum equation were solved using the temporal Fourier and finite Hankel transforms. Ngoma and Erchiqui [7] investigated the liquid flow with the slip boundary condition in a microchannel between two parallel plates with imposed heat flux. They considered the combined effect of pressure-driven flow and electroosmosis. Jain and Jensen [8] presented an analytical investigation on the effects of electrostatic potential in microchannels. The energy equation was solved with the Nusselt number for constant wall heat flux and constant wall temperature boundary conditions and presented with analytic expressions over a wide range of operating conditions. Min et al. [9] analytically solved the fundamental characteristics of electroosmotic flow through rectangular pumping channels without the Debye–Huckel approximation. The Poisson–Boltzmann equation for the electric potential distribution and the momentum equation for the velocity profile are solved by averaging method. The zeta potential is experimentally measured by the streaming potential technique. They found that their method is applicable when the ratio of a half of the channel width to the EDL thickness is larger than 3. It is observed from the above literature that conventionally, the electric potential distribution inside a microchannel is determined by a simplified analysis based on the Debye–Huckel linear approximation which is valid for small value of wall zeta potential (usually 0.93) except for n value in Page model which had a good fit with polynomial curve.

4 Conclusions The drying process of sago starch took place during the falling period, which needs about 4–10 h for drying temperature of 50–80 °C in order to reach the industrial standard for moisture content (13%). The Page model was the best model to describe the drying behavior for all the drying temperature tested.

580

M. M. Kamal et al.

Table 2 Results of modeling criteria; drying constants and standard error, (SE) Model

Drying temperature 50 °C

60 °C

70 °C

80 °C

Lewis newton

SE = 0.052 k = 0.153

SE = 0.045 k = 0.171

SE = 0.031 k = 0.237

SE = 0.034 k = 0.390

Page

SE = 0.014 k = 0.046 n = 1.609

SE = 0.015 k = 0.072 n = 1.449

SE = 0.015 k = 0.140 n = 1.317

SE = 0.007 k = 0.227 n = 1.479

Modified page II

SE = 0.064 k = 6.87 × 10−6 n = 0.943

SE = 0.058 k = 8.44 × 10−6 n = 0.937

SE = 0.044 k = 8.40 × 10−6 n = 0.962

SE = 0.037 k = 1.49 × 10−5 n = 0.962

Henderson and Pabis

SE = 0.044 k = 0.170 a = 1.138

SE = 0.039 k = 0.186 a = 1.091

SE = 0.029 k = 0.251 a = 1.066

SE = 0.031 k = 0.410 a = 1.063

Modified Henderson and Pabis

SE = 0.044 k = 0.170 a = 0.546 b = 0.273 c = 0.295 g = 0.170 p = 0.170

SE = 0.039 k = 0.186 a = 0.521 b = 0.280 c = 0.290 g = 0.186 p = 0.186

SE = 0.029 k = 0.251 a = 0.558 b = 0.300 c = 0.208 g = 0.251 p = 0.251

SE = 0.031 k = 0.410 a = 0.648 b = 0.362 c = 0.053 g = 0.410 p = 0.410

Wang and Singh

SE = 0.016 a = −0.106 b = 0.002

SE = 0.009 a = −0.122 b =0.004

SE = 0.011 a = −0.165 b =0.007

SE = 0.066 a = −0.214 b =0.010

Logarithmic

SE = 0.015 k = 0.077 a = 1.597 c =−0.553

SE = 0.013 k = 0.107 a = 1.342 c =−0.308

SE = 0.016 k = 0.199 a = 1.129 c =−0.093

SE = 0.029 k = 0.385 a = 1.079 c =−0.022

Two term

SE = 0.044 k =0.170 a = 0.557 b = 0.557 g = 0.170

SE = 0.039 k = 0.186 a = 0.545 b =0.545 g = 0.186

SE = 0.029 k = 0.251 a = 0.533 b = 0.533 g = 0.251

SE = 0.031 k = 0.410 a = 0.531 b = 0.531 g = 0.410

Approximation of diffusion

SE = 0.021 k = 0.337 a = −13.616 b =0.930

SE = 0.018 k = 0.349 a = −12.575 b =0.934

SE = 0.013 k = 0.443 a = −11.018 b =0.936

SE = 0.008 k = 0.822 a = −10.076 b =0.913

Verma et al.

SE = 0.052 k = 0.153 a = 0.099 g =0.153

SE = 0.018 k = 0.349 a = −12.575 g =0.934

SE = 0.033 k = 0.237 a = 0.139 g =0.237

SE = 0.028 k = 0.237 a = −6.039 g =0.253

Effect of Drying Temperature …

581

80.00

Experiment at 50°C Experiment at 60°C Experiment at 70°C Experiment at 80°C Predicted at 50°C Predicted at 60°C Predicted at 70°C Predicted at 80°C

Moisture Content (%)

70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 0

2

4

6

8

10

12

14

16

Time (hour) Fig. 2 Predicted and observed moisture content of sago starch using page model for T = 50, 60, 70 and 80

Experimented - Predicted Moisture Content (%)

2.000 1.500 1.000 0.500 0.000 -0.500

0.00

20.00

-1.000

40.00

60.00

80.00

Experimental moisture content (%)

-1.500 -2.000 -2.500 -3.000

Fig. 3 Plot of residual between experimental moisture content (%) and predicted moisture content (%) Table 3 Equation of drying parameter Model Page, MR =

Equation of drying parameter exp(−kt n )

k = 0.0061T + 0.2759 n = 0.0008T 2 − 0.1099T + 5.1033

Approximation of diffusion, MR = aexp(−kt) + (1−a) exp(−kbt)

k = 0.0155T − 0.5191 a = 0.1218T − 0.5191 b = −0.0005T + 0.9601

Logarithmic, MR = aexp(−kt) + c

k = 0.0102T + 0.4684 a = −0.0177T + 2.4353 c = 0.0181T − 1.4192

582

M. M. Kamal et al.

References 1. Senik, G.: Small-Scale Food Processing Enterprise in Malaysia, pp. 1–9. Food Technology Research Center MARDI, Malaysia (2010) 2. AKarim, A.A., Tie, A.P.L., Manan, D.M.A., Zaidul, I.S.M.: Starch from the Sago (Metroxylon sagu) palm tree—properties, prospects, and challenges as a new industrial source for food and other uses. 7, 215–228 (2008) 3. Greenhill, A.R., Shipton, W.A., Blaney, B.J., Warner, J.M.: Fungal colonization of sago starch in Papua New Guinea. Int. J. Food Microbiol. 119, 284–290 (2007) 4. Hammond, S.T., Brown, J.H., Burger, J.R., Flanagan, T.P., Fristoe, T.S., Mercado-Silva, N., et al.: Food spoilage, storage, and transport: implications for a sustainable future. Bioscience 65, 758–768 (2015) 5. Liu, Z.Q., Yi, X.S., Yi, F.: Effect of bound water on thermal behaviors of native starch, amylose and amylopectin. Starch-Starke 51, 406–410 (1999) 6. Ke, T., Sun, X.: Effects of moisture content and heat treatment on the physical properties of starch and poly (lactic acid) blends. J. Appl. Polym. Sci. 81, 3069–3082 (2001) 7. Liu, Q., Thompson, D.B.: Effects of moisture content and different gelatinization heating temperatures on retrogradation of waxy-type maize starches. Carbo. Res. 314, 221–235 (1998) 8. Mustafa Kamal, M., Baini, R., Mohamaddan, S., Selaman, O.S., Ahmad Zauzi, N., Rahman, M.R., et al.: Effect of temperature to the properties of sago starch. IOP Conf. Ser. Mater. Sci. Eng. 206 (2017) 9. O’Callaghan, J.R., Menzies, D.J., Bailey, P.H.: Digital simulation of agricultural dryer performance. J. Agric. Eng. Res. 16(3), 223–244 (1971) 10. Page, G.: Factors influencing the maximum rates of air-drying shelled corn in thin layers. M.S. Thesis, Purdue University, Lafayatte (1949) 11. Henderson, S.M., Pabis, S.: Grain drying theory – temperature effect on drying coefficient. J. Agric. Eng. Res. 6(3), 169–174 (1969) 12. Yagcioglu, A., Degirmencioglu, A., Cagatay, F.: Drying characteristics of laurel leaves under different conditions. In: Proceedings of the 7th international congress on agricultural mechanization and energy, 26–27 May, pp. 565–569, Adana, Turkey (1999) 13. Sharaf-Eldeen, Y.I., Blaisdell, J.L., Hamdy, M.Y.: Model for ear corn drying. Trans. Am. Soc. Agric. Eng. (1980) 14. Wang, C.Y., Singh, R.P.: A single layer drying equation for rough rice. ASAE Paper No: 78–3001. St. Joseph, MI: ASAE (1978) 15. Sharaf-Eldeen, Y.I., Hamdy, M.Y., Blaisdell, J.L.: Mathematical description of drying fully exposed grains. ASAE Paper No: 79–3034, St. Joseph, MI: ASAE (1979) 16. Verma, L.R., Bucklin, R.A., Endan, J.B., Wratten, F.T.: Effects of drying air parameters on rice drying models. Trans. ASAE 28(1), 296–301 (1985) 17. Karathanos, V.T.: Determination of water content of dried fruits by drying kinetics. J. Food Eng. 39(4), 337–344 (1999) 18. Diamante, L. M., Munro, P.A.: Mathematical modelling of hot air drying of sweet potato slices. Int. J. Food Sci. Technol. 26(1), 99–109 (1991) 19. da Silva, W.P., e Silva, C.M.D., Lins, M.A., Gomes, P.: Empirical models to describe thin layer drying of lima bean (Phaseolus lunatus L.). African J. Agric. Res. 7, 6376–6382 (2012) 20. Sobukola, O.P., Dairo, O.U., Odunewu, A.V.: Convective hot air drying of blanched yam slices. Int. J. Food Sci. Technol. 43, 1233–1238 (2008) 21. Doymaz, I.: Thin-layer drying characteristics of sweet potato slices and mathematical modelling. Heat Mass Transf. 47, 277–285 (2011)

Evaluating Mechanical Properties of Egg Shell, and Coco Peat Reinforced Epoxy Composite Vijay Kumar Girisala, D. Mangeelal and Sunkara Jaya Kishore

Abstract Materials place a foremost role in the automobile department. But the materials available from the Earth in the method of alloys are not fulfilling the requirement, demands in the automobile fields. Conventional materials will have good mechanical, thermal properties but weight is more. In some cases will have less weight but have very less mechanical, thermal properties. The cost of the automobiles completely depends upon materials. So everyone is looking for new material which has less weight, high strength, good mechanical, thermal properties. So the above requirements will be fulfilled by composite materials. Composite materials are the combination of two or more dissimilar materials. Previously the composite materials are prepared by using synthetic fibers (carbon, glass, aramid). Synthetic materials cause more damage to environment. And also they are non-biodegradable, non-recyclable materials and are very costly. Hence, most of the scientists and technologists increased their focus on natural fibers. Natural fibers are best suitable for the replacement of synthetic fibers. Natural fibers have unique properties compared to synthetic fibers and have many advantages such as low cost, low density, recyclable, biodegradable. This research paper discusses about mechanical properties of Egg shell and Coco peat reinforced epoxy composite material (CPE) and its mechanical properties and hardness are discussed in detail. This study provides the basis of mechanical properties for CPE hybrid composites.

V. K. Girisala (B) · D. Mangeelal Malla Reddy College of Engineering and Technology, Hyderabad, TS, India e-mail: [email protected] D. Mangeelal e-mail: [email protected] S. Jaya Kishore S.V College of Engineering, Tirupathi, AP, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_57

583

584

V. K. Girisala et al.

1 Introduction Nowadays, the designers of automobile components are looking for the materials having different properties based upon their usage and assembly functionality to decrease the cost. The materials which are available in the market are not fulfilling the designer requirements. Hence, they are moving toward new materials called composite materials. Composite materials having a dissimilar properties when compared with individual material. The composite materials have good weight to strength ratio, mechanical, thermal, and wear properties. A composite material comprises of two phases. One is discontinuous phase and other is continuous phase. The irregular stage is typically harder and stronger than the continuous phase and is called the “reinforcement material”, whereas the continuous segment is termed as the “Matrix” [1–3]. The reinforcement material may be metals, ceramics, and fiber. Matrix is either thermosetting and thermoplastics. Matrix provides many purposes such as retains the fiber in proper position, transfer of stress to fibers and protective fiber’s surface from abrasion. Previously composite materials are prepared by using synthetic fibers like glass, carbon, aramid, and kevlar. Recently scientists and technologies shifted toward natural fibers in the place of synthetic fibers due to high cost, non-biodegradable [4, 5]. Natural fibers will have unique properties when compared with synthetic fiber and have many advantages such as low cost, low density, recyclable and biodegradable [6]. Recently, the use of natural fibers like Coco peat, banana, hemp, flax, sisal, bamboo, etc. as a reinforcement of the polymer composite material has improved and is being better replaced of synthetic fibers [7]. Fiber placed a major role in composite material due to high strength and high volume [8]. Shankar et al. [9] reported that flax fiber content from 10 to 30% by mass mixed up with high-density polyethylene (HDPE) by extrusion and injection molding process to produce composites. The results showed that by increasing fiber content in material increases the mechanical properties up to 20% by volume and then start decreasing. Chicken eggshell is a waste material from domestic sources such as poultries, hatcheries, homes and fast food restaurants. Egg shell has many advantages such as eco-friendly, good compressive strength, and good thermal properties. Mueller White et al. [10] have reported that USA alone disposed about 150,000 ton of this material in landfills. Eggshell contains about 95% calcium carbonate in the form of calcite and 5% organic materials such as type X collagen, sulfated polysaccharides, and other proteins. According to Avatar Singh Saroya et al [11], addition of eggshell to reinforcement leads to reduce in the tensile strength and young’s modulus and other hands increases the %of elongation at break and impact strength. Water absorption capacity of composite also increased. So, with the addition of eggshell powder to Coco peat powder the compression and impact strength of composites increases. The novelty of this research work we have used Coco peat and Egg shell as a reinforced and resin and hardener as a matrix. Coconuts are abundantly available in the world. Coco peat is a by-product of coconut. The traditional use of Coco peat is for agriculture. Coco peats have many advantages such as low cost, eco-friendly, good water storage capacity and moderate

Evaluating Mechanical Properties …

585

mechanical properties, which make them better alternative of synthetic fiber in many automobile applications such as friction clutch plates lining.

2 Methods For the preparation of Coco peat and Egg shell composite material, we have to use Coco peat as a reinforced and resin and hardener (LY556 and HY951) as a matrix. The strength of the composite depends on the composition of the ingredients. Different compositions of materials are prepared.

2.1 Preparation of Composite Material The Coco peat is taken as shown in Fig. 1 and dried in sunlight for about 12–16 h. After drying, the Coco peat is filtered to uniform grain size. For uniform mixture to get uniform strength while making composites. The Egg shells are collected from the food canteens. The Egg shells are cleaned in hot water and dried in sunlight for about 12–16 h. Further, Egg shells are grinded into fine size powder by using grinding machine and filtered into uniform size shown in Fig. 2. For making composite materials, we have chosen die as PVC pipe with 20 mm diameter and 40 mm length. The PVC pipe was cut into two halves as shown in Fig. 3 as mold for easy filling and removal of material. Before going to fill the material in die make sure to remove the entire dirty and dust particles and finally apply grease/vaseline/oil to die for easy removal of die. The Coco peat and Egg shell are taken in three different compositions in ratio of 80:20/70:30/80:20 as shown in Fig. 4 and add matrix (resin and hardener) to prepare

Fig. 1 Coco peat mixture

586

Fig. 2 Egg shell with uniform grain size

Fig. 3 PVC pipe

Fig. 4 Coco peat and egg shell mixture

V. K. Girisala et al.

Evaluating Mechanical Properties …

587

Fig. 5 Coco peat and egg shell with matrix

Fig. 6 Die filled with mixture

natural hybrid composite as shown in Fig. 5. The natural composite was prepared by using hand lay technique. In this hybrid composite, Coco peat processes tensile nature, Egg shell processes compressive nature. The mixture was filled into die as shown in Fig. 6, close the die and tie with thread as shown in Fig. 7. The mixture is made to cure at room temperature for 24–48 h. After curing for 24–48 h, the composite material from the die was removed. The final specimen was shown in Fig. 8. Cares should be taken while removing specimen from the die. Use chisel and hammer to remove material from the die.

2.2 Testing of Composite Material The specimens of different compositions are tested under UTM machine for tensile strength, circular tube universal testing machine for compressive strength, Brinell hardness for hardness and pin-on-disk for wear test.

588

V. K. Girisala et al.

Fig. 7 Mixture inside the die tie with thread

Fig. 8 Final specimen

3 Results and Discussions 3.1 Tensile Test Tensile test is conducted on a universal testing machine (UTM) as per ASTM standards D638 for three different compositions of specimens and the properties are shown in graph 1. This test reveals the amount of energy absorbed by materials during fracture, which refers to the material’s tensile strength. The below bar graph shows the variations of tensile strength for compositions of CPE 80:20/70:30/60:40. From that we can clearly say with increasing Coco peat composition percentage the tensile strength was gradually increasing. This is due to adhesive bonding between two materials. Coco peat (CP) is having tensile nature in it such that as CP is higher it can have capability of withstand for high loads. It is 3.35% for CPE combination of 80:20% composite.

Evaluating Mechanical Properties …

589

Graph 1 Variations of tensile strength for different composition

3.2 Compressive Test Circular tube universal testing machine instrument is used to test compression strength of the specimens, and the properties are shown in graph 2. This test reveals the amount of energy absorbed by a material during fracture, which refers to the materials compression strength. The below bar graph shows the variations of compressive strength for different compositions. From that we can clearly say with increasing Egg shell composition percentage the compressive strength was gradually increasing due to its compressive strength nature. Hence, in graph 2 it is increasing up to 20% for CPE combination of 60:40% composite.

Graph 2 Variations of compressive strength for different composition

590

V. K. Girisala et al.

Graph 3 Variations of hardness for different composition

3.3 Hardness Test Brinell hardness test was used to test the hardness of the composite material, and the properties are shown in graph 3. Hardness is defined as the resistance of a material to permanent deformation such as indentation, wear, abrasion, and scratch. The Brinell hardness test is preferred because it is simple, easy, and relatively non-destructive. The below bar graph shows the variations of hardness for different compositions. As Egg shell percentage improves, the hardness also improves to a percentage of 2.663%.

4 Conclusions The mechanical properties of Coco peat/Egg shell composite materials have been tested in present work. From the final results, conclusions are summarized as follows. The results revealed that the incorporation of Coco peat/Egg shell reinforced materials is superior at tensile strength in composition of 80:20% according to compression and hardness it is superior at 70:30%. Tensile strength of composite material is more for 80:20% composition because coconuts have a good tensile strength nature it improves by 3.35%. According to compressive strength, 60/40 composition has provided better results because Egg shell will have good compressive properties it is increasing up to 20%. As per hardness, 60/40% composition provides a better result because almost equal amount of Egg shell and Coco peat are used and have a good bonding strength between them that improves to a percentage of 2.663%.

Evaluating Mechanical Properties …

591

Acknowledgements We express our heartful gratitude to Jyothi Spectro Analysis Private Limited, Balanagar Hyderabad, India for providing facilities and necessary support in testing specimens. Author declares that there are no conflicts of interest.

References 1. Banana fiber reinforced polymer composites—a review 2. Journal of reinforced plastics and composites 2010 29: 2387 originally published online 24 May 2010 3. Kishore, S.J.: Structural analysis of multi-plate clutch. http://www.ijcttjournal.org 4. Shankar, P.S., Reddy, D.K.T., Sekhar, V.C.: Mechanical performance and analysis of banana fiber reinforced epoxy composites (IJRTME) 1(4), Nov 2013 5. Saroy, A.S.: Study of mechanical properties of hybrid natural fiber composite 6. Maschinenwesen, D.F.: Investigation on jute fibres and their composite based on polypropylene and epoxy matrices. Kuruvilla Joseph Romildo Dias Tolêdo Filho 2, a review on sisal fiber reinforced polymer composites 7. Lu, X., Zhang, M.Q.: The preparation of self-reinforced sisal fiber composites 8. Mueller, D.H., Krobjilowski, A.: New discovery in the properties of composites reinforced with natural fibers. J. Ind. Text. 33(2), 111–129 (2003) 9. Lilholt, H., Lawther, J.M.: Comprehensive composite materials. Chap. 1.10 (2000) 10. Mishra, S.C.: Nadiya bihari nayak and alok satapathy: investigation on bio-waste reinforced epoxy composites. J. Reinf. Plast. Compos. 29(19), 3016–3020 (1999) 11. White, N.M., Ansell, M.P.: Straw reinforced polyester composites. J. Mater. Sci. 18(5), 1549– 1556 (1993)

Design Modification of Rear Axle Housing by Fatigue Failure Analysis Barada P. Baisakh and Anil K. Prasad

Abstract The rear axle housing is one of the key components in heavyweight carrying vehicles. The failure of this component before warranty period during normal working conditions is not acceptable. There is a big difference in practical road condition and boundary condition made in analytical approach. In the present work, Ansys stress life approach is used for better reliable results. Ratio loading type in Ansys fatigue tool helps to analyze the fatigue failure very close to practical load condition. The fatigue failure analysis is carried out considering static structural analysis methods using mechanical APDL solver. The fatigue life is determined as well as location of critical cross-section, maximum equivalent stress before failure initiation. Further, design enhancement solution to increase the minimum fatigue life is proposed. The stress concentration zone near the spindle is reduced by extrude cut technique using Creo parameter 2.0 solid modeling software.

1 Introduction The rear axle housing is the main load-carrying component in the vehicle and due to irregular road conditions, dynamic repeated load is acting on it which may lead to fatigue failure of the component. The component is designed with high safety factor that leads to crack growth when fatigue stress exceeds the average load value. Many researchers have carried out to predict the fatigue failure of the axle housing. Topac et al. [1] analyzed prediction of fatigue failure rear axle housing prototype and tension test was performed on prototype to know the young’s modulus, yield strength, and ultimate strength. Test results compared with Finite element analysis results and solution given by him to increase the fatigue life was to increase the thickness of the prototype which is not economical. In this paper design enhancement B. P. Baisakh (B) · A. K. Prasad National Institute of Technology Jamshedpur, Jamshedpur, Jharkhand 831014, India e-mail: [email protected] A. K. Prasad e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_58

593

594

B. P. Baisakh and A. K. Prasad

solution is provided to increase the fatigue life of the component which is practically cost-effective. Yimin et al. [2] failure in drive axle of a mining dump truck by using dynamic strain measurement technique and this is done by studying the effect of elasticity of axle housing on static stress. Bradley Steven et al. [3] failure of crane truck axle was analyzed with different fracture mechanisms. The cause of failure was field addition of stop block which is welded to axle housing. Jing et al. [4] fatigue life prediction of vehicle driving axle house was analyzed under random loading conditions and different classes of virtual road surfaces. Cong et al. [5] were analyzed driving axle housing fatigue failure in fatigue bench test using FEA. In this method, various strain gauss are used to know the exact load spectrum. According to the acceptance of the axle housing model it has to resist number of load cycles between 104 and 105 without fatigue failure. But during the fatigue analysis it was found that load cycle is less than 4 × 104 load cycle. So, design enhancement solutions should be provided in order to safe functioning of axle housing model. In order to perform design analysis first a detailed model of the housing was formed in Creo parameter 2.0 solid modeling software. By using this finite element model was formed. Then the stress and fatigue analysis were performed in Ansys 16.0 structural analysis in Ansys Apdl solver. Tensile test results were referred from Topac et al. [1] prototype vertical tension test which was to be used in validation purposed. For fatigue analysis, fatigue tool was used in ANSYS 16.0 and load ratio method as per loading condition. In order to perform fatigue analysis, it is very much important to give the value of alternating stress table which will draw S N curve according to the loading condition we have taken by considering strength modifying factor.

2 CAD and FE Model Description The full-scale cad model was prepared in Creo parameter 2.0 solid modeling software shown in Fig. 1. The required thickness of different parts and contact surfaces were given to the CAD model shown in Fig. 2. The axle housing consists of two twin wall shells of thickness of t s = 9.5 mm and welded together. There is one spindle in each side was friction welded with the thin walled shell. There are break mounting flange on each side of spindle to hold the braking assembly and it was welded with thin wall shells. For the sealing purpose dome cover was welded with the thin wall shells and as we can see there are one-step variable cross-section. For the sealing purpose dome cover was welded with thin walled shell. The cross-section x-x is a rectangular hollow section dimension as shown in Fig. 1. Cad model which was prepared in Creo parameter 2.0 imported into Ansys workbench V16.0 to constitute FE model. In meshing, physical preferences are set to mechanical and element size is set to 8.5 mm as the thickness of the thin wall sheet was taken as 9.5 mm and it should be less than that for better finite element results.

Design Modification of Rear Axle …

595

Fig. 1 Geometry of the model (all dimensions are in mm)

Fig. 2 Cad model showing different parts of the model

Figure 3 shows the meshing of housing in finite element method. The number of nodes and element were found to be 438,468 and 254,280, respectively.

3 Axle Housing Material The material assigned to all the parts of housing was structural steel and tension test results of housing material were shown in Table 1. To perform the fatigue analysis, we have to assign the alternating stress value which is shown in Fig. 4 and the log graph between the number of cycles and alternating stress can be seen in Fig. 4. The safe range of number of cycles is between 104 and 106 . The infinite number of

596

B. P. Baisakh and A. K. Prasad

Fig. 3 Meshing of the housing Table 1 Result of tension test Material

Modulus of elasticity (GPA)

Poisson’s ratio

Yield strength (MPA)

Ultimate strength (MPA)

% of elongation

Structural steel

208.5

0.3

497.5

629.9

26.5

Fig. 4 Alternating stress

Design Modification of Rear Axle …

597

Fig. 5 Log log graph

cycle i.e. 106 number of cycles and corresponding stress value is known as endurance strength. Shells were manufactured by stamp welding process and thickness of shell is 9.5 mm throughout made up of structural steel. The mechanical properties can be found in Topac et al. [1] (Fig. 5).

4 Loading Condition The housing model was simply supported with one vertical loading on each side that is nothing but the spring load (load of the vehicle) and for the practical condition significance two moment due to spring back effect are applied. Support was given exactly where the bearing setting position on both the side. Vertical maximum load of P = 89,271 N was applied on each side and moment M = 4.5 Nmm was applied on each side. The actual load ranges from 1786 to 89271 N. This is the alternating load which was continuously applied on housing model in number of cycles. The characteristics of actuator loading used in fatigue test was given in Fig. 6. In fatigue tool of Ansys the ratio method is used in we give the ratio, i.e. ratio between the minimum load to the maximum load were referred from ANSYS theory [7]. So that Ansys applies these two loads in number of cycles. The stress life analysis type was chosen. Goodman mean stress theory was used by solver to calculate the number of cycles. The applied loads, moments and boundary conditions were shown in Fig. 7.

5 Finite Element Analysis of Housing and Results The reason for using finite element analysis is to find the location where the stress concentration is higher after the application of load P and M on the housing model.

598

Fig. 6 Characteristics of actuator load

Fig. 7 Boundary condition

B. P. Baisakh and A. K. Prasad

Design Modification of Rear Axle …

599

Fig. 8 Max von mises stress

In this paper, we will see the stress concentration, total deformation life and factor of safety of housing model and after that design optimization technique is used to make all the parts and long running life. The analysis was carried out using Ansys workbench V16.0 FEA software on a 2 GHz Intel I5 processor Lenovo Ideapad 330. Figure 8 shows the Von Mises stress distribution provided from FE analysis. And it can be clearly seen the location of stress concentration and maximum stress location was found to be in spindle. The maximum von mises stress σ max = 436.57 MPA can be seen near steps in spindle region. If load exerted statically and for a single cycle it can be a safe condition. The spindle part is not safe afterlife of 3.8 × 104 number of cycles. The first initiation of fatigue failure can be seen.

6 Theoretical Validation The rear axle housing is actually loaded with dynamic loading so fatigue analysis of the model was performed. As we know the endurance strength of the component is given by se = 0.5 × sut where sut = ultimate strength of the axle material. The fatigue life prediction of part in the range of 104 –106 cycle, the S N curve can be seen in Fig. 5 which was obtained from simple tensile test data. In order to predict the true fatigue life of the component in laboratory sample, we have to consider the design, manufacturing and environmental influences on fatigue strength [6]. Se = ka ∗ kb ∗ kc ∗ kd ∗ ke ∗ Se

(1)

600

B. P. Baisakh and A. K. Prasad

where • K a = surface factor which depends on surface finish recommended value ultimate stress is 0.959 • K b = shape factor and for non-round depth of cross-section >50 mm • K c = load factor and for bending given by 1 • K d = temperature factor and given by 1 for range of temperature 0–250° • K e = fatigue strength modifying factor. It is used because from FEA analysis it can be seen stress concentration region on the spindle region = 1/K f • K f = fatigue stress concentration factor it is depending on stress concentration as model has shape and dimension complexity. • K t = σpeak /σnominal , • σ nominal = stress induced in the model if there is no stress concentration = MZ where M = max moment corresponding to the max load applied on the model that is 9100 kg. Z is the section modulus of the critical cross-section which is hollow critical cross-section dimension shown in Fig. 1. σ nominal was computed as 329 MPA and σ peak can be seen from maximum stress coming near spindle region found from FEA analysis. Stress life approach was used to determine fatigue life of housing material. Modified Goodman method used to verify the factor of safety of the component at the critical cross-section. As the mean stress is greater than 0 we can use this method. σm σa 1 + = sut se fos

(2)

σmax +σ min . where σm = mean stress = 2 σmax and σmin is corresponding to the maximum load and minimum load applied to the component.

σa = amplitude stress =

σmax − σmin 2

From the above Eq. 2 we can find out the factor of safety of the component to be 0.641 and this value can be verified by the minimum safety factor value found to be 0.622 analysis by stress approach method shown in Fig. 9.

7 Design Optimization Design optimization in the housing model was very much required because it can be seen σ max was found to be 436.59 MPA which was near to σ yield which was 497.5 MPA. So yielding may occur and fatigue analysis results also show that minimum life of the component was 3.8 × 104 number of cycles. And as addition of new

Design Modification of Rear Axle …

601

Fig. 9 Safety factor

material having high rigidity and increase the thickness of the housing metal sheet is not cost-effective we opted the design optimization method. Design optimization technique should be performed because it can be seen in Topac et al. [1] that if we increase the thickness of housing model its rigidity increases hence life increases but at the same time its weight also increases. And now for the same input, we are requiring more power to drive the vehicle. As the region near the spindle sudden change in cross-section takes place shown in Fig. 10, so it can be considered. In design optimization, we have to focus the part where the stress concentration is maximum and in our case is at spindle region. There are many design constraints that we have to keep in mind that we cannot change the design the portion of spindle which

Fig. 10 Existing design

602

B. P. Baisakh and A. K. Prasad

Fig. 11 Proposed design

have dependence with other part assembly. For example, bearing seating location of spindle. Bearing seating assembly dimension are fixed so we cannot change that portion of spindle because if we do so total bearing design have to be changed which is not cost-effective. There is another way we can change the dimension of rectangular beam but that is also not cost-effective because if we do so we have to change the spring rest assembly. So another option left is to increase the number of steps given in spindle and in spindle where there was only one step and stress concentration was high if we increase the number of steps to two we can reduce the stress concentration region due to distributed area shown in Fig. 11.

8 Results and Discussion The fatigue failure during vertical fatigue test can be seen by the finite element model Fig. 8 subjected to stress concentration which can cause premature failure before the predicted minimum life cycle of 4.5 × 104 . The enhancement of fatigue life can easily be achieved by increase in thickness of sheet metal that considerably increases weight of the axle. But at the same time leaving spindle region all other parts show the infinite number of cycles. The experiment shows that in Topac et al. [1] if we increase thickness by 0.5 mm the design life of the critical cross-section increased to 5.8 × 106 number of cycles which is undesirable. By changing the design of transition region of spindle the stress concentration decreases hence increase the design life of the component. The change of maximum equivalent stress and minimum life of the critical component (spindle) before and after the design modification can be seen in Table 2 and FE model analysis can be seen in Figs. 12, 13, 14, and 15.

Design Modification of Rear Axle …

603

Table 2 Finite element analysis results Allowable stress (MPA)

Maximum equivalent(T) induced (MPA)

Minimum life (number of cycle)

497.5

Existing model

Proposed model

Existing model

Proposed model

436

376

38,829

42,540

Maximum deformation (mm)

1.11–1.22

Fig. 12 Equivalent stress (existing model)

9 Conclusions Design optimization of truck rear axle housing is investigated by using finite element analysis. In the analysis, we have compared the maximum tensile stress-induced, minimum life of axle housing, maximum deformation value before and after the design modification. From the analysis of failure of axle housing model, the following conclusions can be drawn. 1. The Fatigue cracks originated near spindle critical cross-sections which are stress concentration regions. 2. The maximum tensile stress induced in axle housing model in both the case is well within the allowable stress. By the help of design optimization, we have

604

Fig. 13 Equivalent stress (proposed model)

Fig. 14 Life (existing model)

B. P. Baisakh and A. K. Prasad

Design Modification of Rear Axle …

605

Fig. 15 Life (proposed model)

decreased the value of maximum tensile stress to 376 MPA and increased the life of the critical cross-section to 42,540 number of cycles. 3. The maximum tensile stress value of the proposed model is less than the yield point stress and the minimum life of the proposed model is within acceptance limit that is 104 –106 for axle housing. So now design is safe. 4. The maximum deformation of the axle housing was found to be 1.11–1.22 mm which is bearable because finite element analysis shows no slips of contacts in the assembly. There can be various other methods in which we can optimize the design there are as follows: 1. Carrier housing thread diameter can be changed which is now at M12–M16 for giving extra strength to the assembly of lower and upper sleeves. 2. The distance from transition zone can be shift towards circular zone in each side of axle housing model. 3. Spindle radius can be changed without undercutting and bell mouth profile can be changed in order to smoothen the profile.

References 1. Topac, M.M., Gunal, H., Kuralay, N.S.: Fatigue failure prediction of a rear axle housing prototype by using finite element analysis. Eng. Fail. Anal. 16, 1474–1482 (2009) 2. Yimin, S., Jing, L., Mechefske Chris, K.: Drive axle housing failure analysis of a mining dump truck based on the load spectrum. Eng. Fail. Anal. 18(3), 1049–1057 (2011)

606

B. P. Baisakh and A. K. Prasad

3. Bradley Steven, W., Bradley, Walter L.: Analysis of failure of axle housing crane truck with fracture mechanics. Eng. Fail. Anal. 2(4), 233–246 (1995) 4. Jing, G., Jian, S., Tao, Z.: Fatigue life prediction of vehicle’s driving axle house under random loading. J. Mech. Strength 30(6), 982–987 (2008) 5. Cong, N., Shang, J., Chen, X., Yang, Z., Liang, K.: Accelerated fatigue bench test of driving axle housing based on FEA. International Conference on Measuring Technology and Mechatronics Automation, pp. 3–6 (2009) 6. Shigley, J.E.: Mechanical Engineering Design, p. 196. McGraw-Hill Kogakusha, Tokyo (1977) 7. ANSYS theory reference. ANSYS release 11.0. ANSYS, Inc. (2007)

Mixture Design Using Low-Cost Adsorbent Materials for Decolourisation of Biomethanated Distillery Spent Wash in Continuous Packed Bed Column Ishwar Chandra, Anima Upadhyay and N. Ramesh

Abstract During ethanol fermentation from molasses, a large quantity of coloured wastewater is generated called spent wash. The spent wash mainly consists of melanoidins which contribute to its colour. Melanoidin is recalcitrant compounds which are toxic and inhibitory to the micro-organisms and make the degradation of spent wash a difficult task. Therefore, conventional anaerobic digestion and aerobic treatment are incapable of bringing the spent wash characteristics to the level set by CPCB. The advanced technology of multiple-effect evaporator and reverse osmosis is unaffordable. None of the existing technology promises to provide foolproof solution with return on investment. Therefore, through this study, we are presenting a novel approach for decolourisation of spent wash with some return on investment through crop cultivation. Our methodology involves initial qualitative study using Soil, Sand, DYS, and Bagasse on decolourisation followed by quantitative study through packed bed, sand, DYS and bagasse showed the maximum decolourisation of 73.33%, 66.086%, 62.958% and 59.646%, respectively. Later, column studies with soil, sand and bagasse were carried out. The efficiency of packed bed column depends on the composition of packed modified soil. Therefore, various combinations were tested, and maximum decolourisation of 98% was achieved with 175 mL

I. Chandra (B) Department of Biotechnology, Sir M Visvesvaraya Institute of Technology, Bangalore, India e-mail: [email protected] Scholar (Ph D) at Department of Biotechnology, School of Applied Sciences, REVA University, Bangalore, India A. Upadhyay Department of Chemistry, Sir M Visvesvaraya Institute of Technology (Affiliated to Visvesvaraya Technological University), Krishnadeveraya Nagar, Hunasamaranahalli, Bangalore, Karnataka 562157, India e-mail: [email protected] I. Chandra · N. Ramesh Department of Biotechnology, School of Applied Sciences, REVA University, Rukmini Knowledge Park, Kattigenahalli, Yelahanka, Bangalore, Karnataka 560064, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_59

607

608

I. Chandra et al.

of soil, 75 mL of sand and 0 mL of bagasse. Nutritional analysis of spent modified soil reveals increased quality.

1 Introduction In the sugar mill, several by-products, like molasses syrup, pressed bagasse and fibre cake, are delivered. Molasses syrup being the most imperative among them since it contains around half saccharides and along these lines, it has high business esteem. It is utilised as a carbon source in many bioprocessing industries like alcohol, amino acids and antibiotics productions. However, when molasses is utilised as substrate for fermentation of ethanol, an oversized quantity of coloured wastewater is released. Quite recently, considerable attention has been paid to molasses-based alcohol distilleries, and they are featured under the “Red Category” list as per the Indian Ministry of Environment and Forest [1]. The wastewater left after the distillation of ethanol is the major liquid pollutant from these distilleries, and it is known as spent wash. They are considered as environmental threat for sustainable development because the spent wash is capable of polluting recipient water body and agricultural land by altering their chemical and biological properties [2–8]. The common features of discharge from these industries are presented in Table 1. For each litre of ethanol produced, 8–15 L of spent wash is generated [9] which is dark brown colour with high chemical oxygen demand (COD) [10]. The colour in spent wash is recalcitrant in nature and toxic to environment [10, 11]. According to the latest reports, there are 397 operating distilleries in India producing more than 2.7 billion litres of ethanol and 40 billion litre of spent wash annually [12, 13]. Therefore, proper handling and disposal of spent wash are necessary to prevent its adverse impact on the environment. Several authors have studied decolourisation of spent wash and showed that physico-chemical methods are highly effective in reducing colour and COD up to 80–90%. However, these methods also pose some serious limitation which is summarised in Table 2 [14–18]. Treatment by biological approach has been attempted by many authors, which results in satisfactory COD reduction but marginal amount of colour degradation. This is most interesting approach to this issue which involves sequential anaerobic–aerobic treatment with the help of various microbes. It recovers methane gas in first step and leaves behind biomethanated distillery spent wash (BDSW), which is later treated in aerated lagoons. When compared to physico-chemical methods, biological methods are less effective in reducing COD, but because of low maintenance cost and easy operation, it is followed in most of the distilleries [19]. A major difficulty in biological treatment of spent wash is due to the presence of a polymer and a natural browning pigment called melanoidins [20, 21]. They are responsible for imparting dark brown colour to spent wash. The melanoidins from BDSW have empirical formula of C17–18 H26–27 O10 N with molecular weight distribution between 5–40 KDa [22]. They have diverse structure, elemental composition [23, 24] and show antioxidant [25], antimicrobial [26] property that is why they are toxic to the microbes used in spent wash treatment leading to failure of biological

Mixture Design Using Low-Cost Adsorbent … Table 1 Physico-chemical characteristics of distillery spent wash (as per AIDA, New Delhi)

Characteristics

609 Value

pH

3.80

EC (dS/m)

43

TDS (mg/L)

91,700

TSS (mg/L)

26,560

TS (mg/L)

118,260

BOD (mg/L)

43,000

COD (mg/L)

128,000

Organic carbon (%)

3.7

Nitrogen (mg/L)

1460

Phosphorus (mg/L)

326

Potassium (mg/L)

14,300

Sodium (mg/L)

356

Calcium (mg/L)

6800

Magnesium (mg/L)

4384

Chloride (mg/L)

10,650

Sulphate (mg/L)

3000

Copper (mg/L)

2.8

Manganese (mg/L)

9.2

Iron (mg/L)

24.6

Zinc (mg/L)

7.8

Bicarbonates (mg/L)

1530

Table 2 Physico-chemical treatment methods and their corresponding limitations Treatment method

Limitations

Adsorption using activated carbon [14], flocculation and coagulation [15]

Generate secondary pollutants

Incineration in multiple-effect evaporator (AIDA and CPCB, India)

Cost intensive

Advanced oxidation techniques, viz. ultrasound and ozone [16]

Scale-up issues

Ultrafiltration and reverse osmosis [17] and membrane technology [18]

Energy intensive

treatment process. Chiefly, the maillard product of sucrose–aspartic acid (SAA-MP) is the major colour imparting melanoidin in BDSW [27, 28]. Due to the absence of suitable technique of spent wash disposable and strict instructions from CPCB India for zero-discharge, the distilleries are choosing to dispose the spent wash by dilution, gardening, blending with concrete and controlled use in fertigation [29, 30]. All these solutions to the problem are concession, and they

610

I. Chandra et al.

do not bring down the spent wash characteristics below the standards set by CPCB. Looking at these shortfalls, the paper presents a novel work aimed to increase the overall efficiency of the BDSW treatment with some return on investment (RoI). The approach involves preparation of modified soil packed in a cylindrical column. When the BDSW is passed through the column, the colour is adsorbed up to the tune of 95–98%. The left spent soil could be added back to the field which will increase the crop yield.

2 Materials and Methods The methods of biosorbent preparation, analytical methods for characterisation, batch and continuous decolourisation studies are presented in this section in detail.

2.1 Collection and Bio-chemical Analysis of BDSW Biomethanated distillery spent wash was collected from a reputed distillery unit in Karnataka, India. Physico-chemical and biological parameters were estimated as per standard methods prescribed by CPCB (Govt. of India Organisation). BDSW was characterised for colour using UV visible double-beam spectrophotometer (Shimadzu Analytical, India) and later calibrated against COD (open reflux method). BDSW having optical density 0.069 (after 50X dilution, since the colour was very dark) at 475 nm (λmax of Melanoidin) served as stock solution.

2.2 Preparation and Characterisation of Biosorbent Carrier Materials Soil, sand, bagasse and distillery yeast sludge (DYS) were used as biosorbent carrier materials. Soil, DYS and bagasse were collected from the distillery unit, and sand was obtained locally. All the biosorbents were cleaned thoroughly and shade dried. Bagasse was subjected to size reduction into a fine powder using a mixer cum grinder (Panasonic, India). Size characterisation of all the carrier material was done using sieve equipment (Universal engineering, Bangalore) of known mesh cutoffs. Soil, sand and bagasse were characterised for nutritional content like carbon, nitrogen, phosphorus, potassium, calcium, magnesium, sulphur, iron, manganese, zinc, copper as per standard methods. DYS was analysed for free amino acid (Ninhydrin method), protein concentration (Lowry’s method), water-soluble and fat-soluble vitamins. The quantification of water-soluble vitamin content was done through UPLC-MS/MS system. The analytical column 2.1 × 50 mm UPLC BEH-C18 (at IIHR, Bangalore)

Mixture Design Using Low-Cost Adsorbent …

611

was used for the water-soluble and fat-soluble vitamins analysis. The carrier materials used for the study were taken on volume basis.

2.3 Batch Experiments Preliminary qualitative studies using individual biosorbent were carried out under batch mode to check their possible effect on decolourisation of BDSW. The batch adsorption studies were performed in 50 mL test tubes, with 5 mL of biosorbent dispensed in 40 mL of BDSW (pH of 7.0) at varying colour concentrations obtained by diluting the stock with distilled water. All the samples were incubated at constant temperature of 30 °C ± 2 °C under static conditions with intermittent mixing. After 17 days, percentage of decolourisation (% DC) was calculated. Biosorbent which showed maximum colour reduction was used in continuous studies.

2.4 Continuous Column Flow Studies 2.4.1

Design and Principle

Packed bed reactor was fabricated for the study which works on the principle of capillary seepage under gravity feed. The system consists of transparent cylindrical plastic column with defined height and diameter, packed with mixture of biosorbent carrier materials known as modified soil. BDSW is allowed to pass through the packing under gravity feed. The input flow rate is controlled, and output samples were collected from the exit of the column. Figure 1 illustrates the schematic diagram of the packed bed with modified soil. The detail specifications of the column are presented in Table 3. Fig. 1 Schematic design of packed bed column with modified soil

612

I. Chandra et al.

Table 3 Column design for decolourisation studies

2.4.2

Property

Value

Column makeup

Plastic

Aspect ratio H:D (outer)

29:4.6 cm

Inner diameter

4.1 cm

Total height of column

300 mL ≈ 29 cm

Working height

250 mL ≈ 21.7 cm

Residence time

12 h

Experimental Setup

Although, batch studies provide proof of concept of biosorption, and for practical application, the continuous studies are equally essential at large scale. In continuous column systems, the biosorbent can be used in multiple cycles of biosorption/desorption, and cost of operation can be reduced. The main disadvantage of using soil in continuous systems is the clogging of column, due to its small size; therefore, its porosity is altered by adding other biosorbent carrier materials. Blending of soil with biosorbents like sand and bagasse forms modified soil. Continuous column studies on decolourisation of BDSW were conducted with the modified soil packed in the column to yield the desired fixed bed height of 250 mL (equivalent to 21.7 cm). Undiluted BDSW (100% strength) was fed to the packed column by gravity feed method at a particular flow rate to maintain a residence time of 12 h. Samples were collected from the exit of the column, and colour intensity was measured using spectrophotometer. All experiments were carried out at 30 °C ± 2 °C since it would practically resemble the industrial situation.

2.4.3

Preparation of Modified Soil

For column experiments, modified soil was prepared by mixing dry biosorbents, viz. sand, soil and bagasse in a proportion mentioned in Table 4. Mixture design with three factors (simplex centroid design) was chosen in the study. Each ml of bagasse, sand and soil corresponds to 0.0616 g, 1.6318 g and 1.1426 g, respectively. These values of mixing ratio have been obtained statistically through mixture design of Minitab software. To find the best blend proportion of each component, all the biosorbent in the modified soil were considered as independent variable and response measured, i.e. % DC was dependent variable, estimated after passing the BDSW through the column of packed modified soil. Each independent variable was evaluated at upper and lower bounds with their corresponding proportions, as indicated in Table 5. Factor levels were chosen by considering the operating limits of the experimental apparatus.

Mixture Design Using Low-Cost Adsorbent …

613

Table 4 Experimental mixture design of three components (v/v) in packed bed column studies during continuous mode operation Reactor no.

Mix composition in reactor (mL) Soil

Sand

Bagasse

1

125

25

100

2

200

25

25

3

250

0

0

4

100

0

150

5

125

100

25

6

100

75

75

7

100

150

0

8

150

50

50

9

175

0

75

10

175

75

0

Table 5 Factors used in the preparation of modified soil at their corresponding lower and upper bounds, with proportions Components

Amount

Proportion

Pseudo-component

Lower

Upper

Lower

Upper

Lower

Upper

A

100.00

250.00

0.400

1.00

0

1.00

B

0.00

150.00

0.00

0.600

0

1.00

C

0.00

150.00

0.00

0.600

0

1.00

2.5 Decolourisation Assay The output sample from the column was analysed for percentage of colour removal against that of original BDSW. The melanoidin degradation efficiency was expressed as the degree of decrease in the absorbance at 475 nm (λmax ) against the initial absorbance (dark brown colour). Following Eq. (1) was used to calculate the % DC. Percent Decolourisation =

Initial OD − Final OD × 100. Initial OD

(1)

2.6 Nutritional Analysis of Spent Modified Soil The residual soil left behind after running the column was analysed for nitrogen, potassium, phosphorus and other micronutrients through standard methods.

614 Table 6 Nutritional properties of BDSW and bagasse

I. Chandra et al. Nutrient

Value in BDSW (mg/L)

Value in bagasse

Total nitrogen

2520

0.28%

Total phosphorus

14.9

0.10%

Total potassium

2600

0.38%

Total calcium

34.3

0.19%

Total magnesium

91.8

0.16%

Total sulphur

21

0.04%

Total iron

306.6

145.2 ppm

Total manganese

6.2

31.4 ppm

Total zinc

87

18.54 ppm

Total copper

4.5

5.20 ppm

3 Results and Discussion 3.1 Characteristics of BDSW Anaerobically, digested distillery spent wash collected after biomethanation process was dark brown in colour. COD of BDSW by open reflux method was 22,000 mg/L (σ = 50 mg/L), total dissolved solids were 0.225 g/mL, detail properties of BDSW are summarised in Table 6. The microbial count in BDSW showed a total of 32 × 108 CFU/mL. Colour of BDSW exhibited perfect positive correlation with COD, which means colour of the spent wash is responsible for high COD.

3.2 Properties of Biosorbent The volume surface mean diameter of soil, sand, bagasse and DYS is 0.184524 mm, 0.200155 mm, 0.200152 mm, 0.174224 mm, respectively. The microbial count in soil, sand, bagasse and DYS is 51 × 106 CFU/g, 238 × 106 CFU/g, 63 × 103 CFU/g and 20 × 103 CFU/g. Soil pH is slightly alkaline, and salt content of the soil is within the safe limits. Organic carbon content of the soil is medium, nitrogen content is low, phosphorus and sulphur content is optimum, potassium content is very low, whereas exchangeable calcium and magnesium contents are low. Sand pH is slightly alkaline, and salt content is within the safe limits. Organic carbon content of the sand is high, nitrogen content is optimum, phosphorus content is very high, potassium content is high, exchangeable calcium, magnesium and sulphur content is high/optimum. Micronutrients such as manganese and copper contents are high/optimum, whereas iron and zinc contents are low. Micronutrients such as manganese content of the sand are optimum and iron, zinc, copper contents are low. Table 7 represents the

Mixture Design Using Low-Cost Adsorbent … Table 7 Values of nutritional content in soil and sand

615

Parameter

Soil values

Sand value

pH

7.33

7.8

EC (dSm−1 )

1.46

0.28

Organic carbon (%)

0.48

Available nitrogen (ppm)

77.16

Available phosphorus (ppm)

33.77

Available potassium (ppm)

300.0

Exchangeable calcium (ppm)

1.05 170.1 63.84 338

6247

5707

Exchangeable magnesium (ppm)

482.5

435

Available sulphur (ppm)

209.13

14.35

Available iron (ppm)

3.58

0.20

Available manganese (ppm)

3.80

4.23

Available zinc (ppm)

1.30

0.55

Available copper (ppm)

1.58

1.35

nutritional value of soil and sand. Bagasse contains good amount macronutrient and micronutrients like nitrogen, phosphorus, potassium, calcium, magnesium, sulphur, iron, manganese, zinc and copper. The detail composition is summarised in Table 6. Vitamins and fats were extracted from DYS, quantification was performed through UPLC-MS/MS system with analytical column 2.1 × 50 mm UPLC BEH-C18, and it was found to be very rich in nutritional content; it showed very high percentage of water-soluble vitamins, viz. niacine, pyridoxine, panthothenic acid and fat-soluble vitamins like vitamin E, and details are summarised in Table 8. Table 8 Values of nutritional content in DYS Protein and amino acids

Water-soluble vitamins

Fat-soluble vitamins

39.98 mg/g and 51.56 mg/GMC

Thiamine

43.60 µg/100 g

Vitamin K2

Niacine

224.12 µg/100 g

Vitamin D2

7.15 µg/100 g

Pyridoxine

158.61 µg/100 g

Vitamin D1

3.43 µg/100 g

Panthothenic

594.62 µg/100 g

Vitamin E

1677.83 µg/100 g

Biotin

3.17 µg/100 g

Vitamin K1

3.52 µg/100 g

Riboflavin

3.04 µg/100 g

Folic acid

0.22 µg/100 g

0.30 µg/100 g

616

I. Chandra et al.

Table 9 Percentage decolourisation with sand, soil, bagasse and DYS Sl. no.

BDSW dilution % v/v

Percentage decolourisation (% DC) Sand

Soil

Bagasse

DYS

1

10

66.087

73.3333

3.18841

41.4493

2

20

62.1302

60.355

9.70414

62.9586

3

30

48.0909

17.3636

4

40

33.09

5

50

34.0226

6

60

35.4808

7

70

8

80

9 10

19.4545

45.9091

17.5182

42.1898

38.6861

16.6541

47.8195

40.1504

19.1667

53.5256

39.5192

34.5304

22.3481

57.1271

42.8729

37.5442

24.7788

59.646

43.1858

90

40.7187

28.1314

59.4661

44.1889

100

10.6673

15.9974

20.8866

10.5534

3.3 Colour Degradation 3.3.1

Batch System Studies

All the biosorbents were found to be effective in decolourisation of BDSW. Soil showed the maximum decolourisation of 73.33% at 10% v/v BDSW. Sand showed maximum decolourisation of 66.086%. On increasing the strength of BDSW, there was a significant decrease in percent decolourisation in the case of sand and soil. Studies with DYS indicated initial increase in percent decolourisation followed by a decline phase of decolourisation. DYS was capable of reducing the colour to the tune of 62.958% for 20% BDSW. Studies with bagasse showed increase in decolourisation with increase in strength of BDSW. A maximum decolourisation of 59.646% for 80% BDSW was achieved. Table 9 summarises the percentage decolourisation with different types of carrier materials. Additionally, all the carrier materials were able to remove the odour except DYS. Bagasse gave a sweet smell to the BDSW after treatment. Since sand, soil and bagasse were effective in reducing colour and odour, they were selected for the continuous studies in the column flow experiments. Different combinations were tried in decolourisation study which is explained in the next section.

3.3.2

Continuous System Studies

Based on the results obtained from the batch studies, the continuous experiments through packed bed column were designed for biosorption of colour from BDSW. The column packed with modified soil was prepared, and BDSW was fed from the top at a particular flow rate. Flow rate is an important experimental parameter in the evaluation of biosorption process performance for any continuous treatment of

Mixture Design Using Low-Cost Adsorbent …

617

industrial wastewater; therefore, the flow rate was regulated based on the calculation of void volume taking into the account of 12 h residence time. The highest percentage of decolourisation 98% was obtained when soil was 175 mL, sand was 75 mL, and bagasse was 0 mL. The percentage decolourisation with ten deferent types of modified soil packed in a column is presented in detail in Table 10. Figure 2 (A–J) shows the discharge output of BDSW from ten different columns. It can be seen that seventh reactor with soil 100 mL, sand 150 mL and bagasse 0 mL, the percentage of colour removal was 97%. Therefore, it can be confirmed that soil and sand are two most important components in formulation of modified soil.

3.4 Nutritional Quality of Spent Modified Soil In most of the effluent treatment plant, it is difficult to dispose the biosorbent material, which later becomes a secondary pollutant and harm the nature. But in our proposed technology, the leftover spent biosorbent can be added back to the field. It is noted that many properties of the modified soil are enhanced after the completion of the column run. The details are given in Table 11. After the column running is completed, the spent modified soil was taken out, and nutritional analysis was performed. Nutritional analysis of the modified soil shows that soil salinity is getting very high, and it is suggested to avoid growing salt-sensitive crops (like grapes, etc.). Soil is getting rich in organic carbon and available nitrogen contents. Available phosphorus and potassium are also increased. It can be suggested to reduce the recommended dose of phosphorus and potassium fertilisers by 50%. Exchangeable calcium, exchangeable magnesium and available sulphur are influenced positively, and there is no need to apply these fertilisers. Available micronutrients, viz. iron, manganese, zinc and copper, are adequately increased, and its application is not required. Organic carbon is increased by 196.87%, available nitrogen increased by 199.18%, available phosphorus increased by 143.41%, available potassium increased by 475.42%, exchangeable calcium decreased by −35.38%, exchangeable magnesium increased by 16.79%, available sulphur increased by 25.35%, available iron increased by 201.68%, available manganese increased by 42.11%, available zinc increased by 54.23%, and available copper increased by 70.25%.

4 Conclusion Molasses-based alcohol distillery releases spent wash consisting of melanoidin which is not easily biodegraded/adsorbed due to complex structure and properties and leads to a environmental damage. Batch adsorption studies show that soil, sand and bagasse could be excellent carrier materials because they can remove considerable amount of colour from the spent wash under static batch mode. Soil showed the maximum

1

92.43

Reactors

% DC

95.33

2 97.36

3 74.61

4

Table 10 Percentage decolourisation achieved through column studies 96.05

5 83.46

6 97.68

7

87.13

8

87.96

9

98.43

10

618 I. Chandra et al.

Mixture Design Using Low-Cost Adsorbent …

619

Fig. 2 Discharged BDSW from the ten columns packed with modified soil

Table 11 Nutritional value of the spent modified soil after running of the column

Parameter

Unit

pH

Spent soil trail 1 7.1

Spent soil trail 2 7.57

EC

dSm−1

2.36

2.9

Organic carbon

%

1.35

1.50

Available nitrogen

ppm

218.7

Available phosphorus

ppm

85.2

Available potassium

ppm

1157.5

2295

Exchangeable calcium

ppm

5039.5

3034.5

Exchangeable magnesium

ppm

548.5

578.5

Available sulphur

ppm

260.0

264.3

Available iron

ppm

6.2

15.4

Available manganese

ppm

7.6

3.2

Available zinc

ppm

1.28

2.73

Available copper

ppm

2.20

3.18

243 79.2

decolourisation of 73.33%. Sand showed maximum decolourisation of 66.086%. On increasing the strength of BDSW, there was a significant decrease in percent decolourisation in the case of sand and soil. DYS was capable of reducing the colour to the tune of 62.958% for 20% BDSW. DYS indicated initial increase in percent decolourisation followed by a decline phase of decolourisation. Studies with bagasse showed increase in decolourisation with increase in strength of BDSW. A maximum decolourisation of 59.646% for 80% BDSW was achieved. In continuous studies, the highest percentage of decolourisation was 98%, obtained when soil was 175 mL, sand was 75 mL, and bagasse was 0 mL. It can also be seen that seventh reactor with soil 100 mL, sand 150 mL and bagasse 0 mL, the percentage of colour removal was 97%. In both the cases, bagasse is 0 mL. So, for a good mixture design, the

620

I. Chandra et al.

bagasse can be eliminated, and percentage of soil and sand can be increased. Further to this, the adsorbents are cost effective, readily available, very rich in macro and micronutrients and easy to dispose because the nutritional profile of spent modified soil reflects improved quality.

References 1. Guide for treatment of distillery effluents, IS: 8032–1976. Panel for food and fermentation industry wastes, BIS. New Delhi, fourth reprint (1998) 2. Kumar, S., Viswanathan, L.: Production of biomass, carbon dioxide, volatile acids, and their interrelationship with decrease in chemical oxygen demand, during distillery waste treatment by bacterial strains. Enzym. Microb. Technol. 13, 179–187 (1991) 3. FitzGibbon, F.J., Nigam, P., Singh, D., Marchant, R.: Biological treatment of distillery waste for pollution-remediation. J. Basic Microbio. 35, 293–301 (1995) 4. Bernardo, E.C., Egashira, R., Kawasaki, J.: Decolorization of molasses’ wastewater using activated carbon prepared from cane bagasse. Carbon 35, 1217–1221 (1997) 5. Pazouki, M., Shayegan, J., Afshari, A.: Screening of microorganisms for decolorization of treated distillery wastewater. Fab 1, Iran. J. Sci. Technol. 32, 53 (2008) 6. Agrawal, C.S., Pandey, G.S.: Soil pollution by spent wash discharge: depletion of manganese (II) and impairment of its oxidation. J. Environ. Biol. 15, 49–53 (1994) 7. Ramana, S., Biswas, A., Kundu, S., Saha, J., Yadava, R.: Effect of distillery effluent on seed germination in some vegetable crops. Biores. Technol. 82(3), 273–275 (2002) 8. Kannabiran, B., Pragasam, A.: Effect of distillery effluent on seed germination, seedling growth and pigment content of Vigna mungo (L.) Hepper (CVT 9). Geobios-Jodhpur 20, 108–112 (1993) 9. Guruswami, R.: Pollution Control in Distillery Industry. National Seminar on Pollution Control in Sugar and Allied Industries, Bombay (1988) 10. Vimala, Dahiya: Utilization of distillery effluents. Chem. Age India 35, 535–537 (1984) 11. Maiorella, B., Blanch, H., Wilke, C.: Distillery effluent treatment and by-product recovery. Process Biochem. 18, 5–8 (1983) 12. Yadav, S., Chandra, R.: Environmental health hazards of post-methanated distillery effluent and its biodegradation and decolorization. Sl.: Singapore, 73–101 (2019) 13. Jogdand, S.N.: Biotechnology for Waste Treatment of Food and Allied Industries, pp. 189–193. Environmental Biotechnology, Himalya Publishing House, Bombay (2003) 14. Satyawali, Balakrishnan, Y.M.: Removal of color from biomethanated distillery spentwash by treatment with activated carbons. Biores. Technol. 98, 2629–2635 (2007) 15. Prasad, R.K.: Color removal from distillery spent wash through coagulation using Moringa oleifera seeds: use of optimum response surface methodology. J. Hazard. Mater. 165, 804–811 (2009) 16. Sangave, P., Gogate, P., Pandit, A.: Ultrasound and ozone assisted biological degradation of thermally pretreated and anaerobically pretreated distillery wastewater. Chemosphere 68, 42– 50 (2007) 17. Murthy, Z.V.P., Chaudhari, L.B.: Treatment of distillery spent wash by combined UF and RO processes, L.B., Global NEST J. 11, 235–240 (2009) 18. Nataraj, S., Hosamani, K., Aminabhavi, T.: Distillery wastewater treatment by the membranebased nanofiltration and reverse osmosis processes. Water Res. 40, 2349–2356 (2006) 19. Shivayogimath, C.B., Ramanujam, T.K.: Treatment of distillery spent wash by hybrid UASB reactor. Bioprocess. Eng. 21, 255–259 (1999) 20. Wedzicha, B.L., Kaputo, M.T.: Melanoidins from glucose and glycine: composition, characteristics and reactivity towards sulphite ion. Food Chem. 43, 359–367 (1992)

Mixture Design Using Low-Cost Adsorbent …

621

21. Sirianuntapiboon, S., Somchai, P., Ohmomo, S., Atthasampunna, P.: Screening of filamentous fungi having the ability to decolorize molasses pigments. Agric. Biol. Chem. 52, 387–392 (1988) 22. Rizzi, G.P.: Chemical structure of colored Maillard reaction products. Food Rev. Int. 13, 1–28 (1997) 23. Cämmerer, B., Kroh, L.W.: Investigation of the influence of reaction conditions on the elementary composition of melanoidins. Food Chem. 53, 55–59 (1995) 24. Yaylayan, V.A., Kaminsky, E.: Isolation and structural analysis of Maillard polymers: caramel and melanoidin formation in glycine/glucose model system. Food Chem. 63, 25–31 (1998) 25. Kitts, D.D., Wu, C.H., Stich, H.F., Powerte, W.D.: Effect of glucose-glycine maillard reaction products on bacterial and mammalian cells mutagenesis. J. Agric. Food Chem. 41, 2353–2358 (1993) 26. Banat, I.M., Nigam, P., Singh, D., Marchant, R.: Microbial decolorization of textiledyecontaining effluents: a review. Biores. Technol. 58, 217–227 (1996) 27. Yadav, S., Chandra, R., Rai, V.: Characterization of potential MnP producing bacteria and its metabolic products during decolourisation of synthetic melanoidins due to biostimulatory effect of D-xylose at stationary phase. Process Biochem. 46, 1774–1784 (2011) 28. Hodge, J.E.: Chemistry of browning reactions in models systems. J. Agric. Food Chem. 1, 928–943 (1953) 29. Chidankumar, C.S., Chandraju, S., Nagendraswamy, R.: Impact of distillery spentwash irrigation on the yields of top vegetables (Creepers). World Appl. Sci. J. 6, 1270–1273 (2009) 30. Ramana, S., Biswas, A., Singh, A., Yadava, R.: Relative efficacy of different distillery effluents on growth, nitrogen fixation and yield of groundnut. Biores. Technol. 81, 117–121 (2002)

Prediction of the WPPO Biodiesel-Fuelled HCCI Engine Using Artificial Neural Networks Ramavathu Jyothu Naik and Kota Thirupathi Reddy

Abstract This paper presents experimental study efforts to explore the performance and emission characteristics of an existing single-cylinder, four-stroke, water-cooled, direct injection Kirloskar diesel engine was converted into HCCI engine. From the investigation, it was stated that WPPO with diesel results increased the brake thermal efficiency by 42.12% at 413 K inlet air temperature and full load condition. Formerly NOx were decreased for all blends and later slightly increases but smoke is negligible. However, the CO and UHC emissions are first increased and then decreased for the HCCI operation. The ANN was trained, validated and tested with experimental data sets. The artificial neural network system was created to predict the performance and emission parameters of the engine. A multi-layer discernment network was utilized for non-straight mapping among input and output parameters. Six objectives—BTE, EGT, NOx , Smoke, CO and UHC were considered. The performance of the ANN model is determined also illustrations the efficiency of the model to predict the performance and emission with a determination coefficient of 0.999.

1 Introduction Attaining a non-pollutant environment becomes a big task in present days, the reason is that carcinogenic chemicals were accumulated continuously in air, non-degradable and degradable matters in air, sea and land disposal of toxic substance. The polluted environment is caused by the hazardous substances released from the industrial events and burning of fossil fuels. R. Jyothu Naik (B) Department of Mechanical Engineering, Jawaharlal Nehru Technological University Anantapur, Ananthapuramu, India e-mail: [email protected] K. Thirupathi Reddy Department of Mechanical Engineering, Rajeev Gandhi Memorial College of Engineering and Technology, Nandyal, Kurnool, AP 518501, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_60

623

624

R. Jyothu Naik and K. Thirupathi Reddy

It is stated that the use of alternative fuels such as hydrogen and alcohols and clean combustion technologies may cause to improve the quality of air. At this condition, it can be predicted nitric oxide (NO) and ultra-low smoke emissions will be produced by HCCI compared to conventional engines [1]. In petrol and diesel engines, fuel injection and spark timing initiate the start of combustion. But in the case of HCCI, the chemical kinetics purely governs the auto-ignition of the fuel/air premixed mixture. So, direct control mechanism is lacked by the HCCI combustion [2, 3]. The most efficient method for the preparation of homogeneous charge is by using external mixture formation. Maurya et al. [4] did exploration different tests regarding PFI technique for ready homogeneous mixture. Tests were performed by varying equivalence ratio (2.0–5.0) and the intake charge temperature (120–150 °C) at constant speed 1500 rpm with a specific final result to achieve the stable HCCI ignition. As stated, that at 120 °C, λ = 2.5 given improved combustion efficiency, combustion parameters and fewer NOx emissions. The results utilizing ANN are a transformative methodology which has not to use any critical numerical conditions for clarifying the non-straight and multi-dimensional frameworks. At the time of training the network, the ANN models the output data validate with predict output. Trained ANN model can predict the performance of an engine for unknown data [5, 6]. Predict the performance and emission characteristics have extensively used the ANN models of SI and CI engines [7, 8]. In this, the present investigation is as follows: ANN-based forecast instrument was utilized to process the HCCI engine performance and emission characteristics, i.e. assessed the connection between the engine input parameters and yield parameters. The experimental outcomes and prediction outcomes of the ANN predicted results are listed in this article.

2 Experimental Setup For attaining the objective of the waste plastic pyrolysis oil-diesel-fuelled HCCI mode, tests have conducted on a single-cylinder, water-cooled, four-stroke naturally aspirated direct injection (DI) diesel engine was converted into HCCI mode. The engine specifications are shown in Table 1. To give a specific load to the engine, an electrical dynamometer was attached to the experimental setup. The fuel is introduce Table 1 Engine specifications

Make

Kirloskar

Bore × stroke

87.5 mm × 110 mm

Swept volume

661.45 (cc)

Connecting rod length

234.00 (mm)

Compression ratio

18:1

Rate speed

1500 rpm

Cooling method

water

Prediction of the WPPO Biodiesel Fuelled …

625

into the cylinder with PFI and is restrained by electronic control unit which synchronized with a crank angle sensor. To accomplish auto-ignition temperature preheats the intake charge of waste plastic pyrolysis oil biodiesel.

3 Results and Discussion Figure 1 shows the effect of intake charge temperature on BTE of HCCI engine for selected fuel blends operation at altered loads. The BTE increases with the increase in the engine load. The lower BTE is found for the lower engine loads because of the retarded start of combustion in HCCI engine the heat losses are lower because of LTC, combustion time is less, better mixture homogeneity causes to less soot formation. The higher combustion temperatures and high HRR are affected by the advance start of combustion. Hence, the surfaces of the cylinder and piston are increased by heat loss; therefore, the net work done is decreases. Least combustion efficiency has noticed at the retardation of start of combustion because of increased emissions and lowers LTC. The maximum thermal efficiency for the WPPO-diesel-fuelled HCCI mode is stated to be 45.12% at 413 K with WPPO-5%with diesel. Figure 2 shows the relation between EGT and engine load for the waste plastic pyrolysis oil blend with diesel HCCI operation. As the results for all intake temperatures, the EGT directly related to the engine load. As engine load increases, EGT increases at all intake temperatures, as estimated. But, the EGT decreases with the intake temperature. As the intake temperature increases, the SOC is advanced because of the chemical kinetics and faster reaction rates. High heat transfer caused by an early SOC with little ignition time. EGT reduces caused by the increase of burned hot gases residence time in cylinder. Figure 3 shows the difference of the NOx emissions for the different fuel blends like WPPO-5%, WPPO-10%, WPPO-15% and WPPO-20% with various load conditions

Fig. 1 Effect of intake charge temperature on BTE of fuel blends at different loads

626

R. Jyothu Naik and K. Thirupathi Reddy

Fig. 2 Effect of intake charge temperature on exhaust gas temperature of fuel blends at various loads

Fig. 3 Effect of intake charge temperature on NOx Emissions of fuel blends at different loads

in HCCI engine. From the figure, it is noticed that the NOx emissions are increased with the engine load for the WPPO fuelled with diesel HCCI engine due to high combustion temperature of the burned gases. High NOx emissions are caused by the increasing in the residence time of the burned gases which is affected by the increase intake temperature, because of the early start of combustion. High NOx emissions were observed at full load condition at 413 K WPPO biodiesel-fuelled HCCI mode. Figure 4 represents the influence of intake charge temperature on CO Emissions of HCCI engine for selected fuel blends operation at different loads. LTC is caused by not enough oxidation temperature of the gases, which results in the creation of CO emissions in the HCCI engine. The decrement in the CO emission is found with increasing engine load due to increase in the peak cylinder temperature. The CO

Prediction of the WPPO Biodiesel Fuelled …

627

Fig. 4 Effect of intake charge temperature on CO emissions of fuel blends at different loads

emission drop is because of early SOC meanwhile more CO emission is because of delay in combustion phasing. The CO emission is higher at a lower intake air temperature (353 K) and loads due to misfire, and it can be controlled at lower load conditions. Figure 5 represents the affect of intake charge temperature on UHC of HCCI engine for selected fuel blends operation at various loads. The incomplete combustion of hydrocarbon fuel is indicated by UHC formation emissions in the HCCI engine because of a LTC. The temperature at the walls of combustion chamber is lower because of the heat losses. Superior parts of UHC emissions rise from the combustion chamber regions. From the figure, it is clear that the increasing the intake charge temperature and load conditions decrease the UHC emission because of increase equivalence ratio with the load. From Fig. 6, it is noticed that decrease in nature of smoke emission is found for increase in the engine load and intake charge temperature. At full load, WPPO20% and at 413 K the smoke emission is negligible because of the preparation of high degree homogeneous mixture with the increased ignition delay. The intake temperature and load cause to decrease the smoke emissions, oxidation temperature and high residence time of burned gases, smoke emission noticed to be negligible at 413 K, full load and WPPO-20%. Be that as it may, at higher load ranges as a result of non-availability of adequate air and abnormal combustion there was a noticeable white smoke emission.

628

R. Jyothu Naik and K. Thirupathi Reddy

Fig. 5 Effect of intake charge temperature on UHC of fuel blends at different load

Fig. 6 Impact of intake charge temperature on smoke opacity of fuel blends at different loads

Prediction of the WPPO Biodiesel Fuelled …

629

Fig. 7 Validation performance graph of tested ANN model

3.1 Modelling with ANN The engine performance and emission parameters have been predicted by ANN model was shown in Fig. 7. The created system engineering has a 3-8-6 configuration, with three neurons in the input layer representing intake air temperature, engine load, and EGT. Output is predicted by mapping 6 neurons BTE, EGT, NOx , smoke, CO and UHC, in output and varying neurons of hidden layer. Train of network with 44 values approximately 70% of experimental data has been used and rest of 30% data has been used to testing. The found test data are defined for getting better the performance of the system. To map input and output data, MLP was utilized and BP algorithm used to train the network. The network training is managed by this algorithm, where the network weights and leaning are computed arbitrarily at the creation of the training stage. Using the gradient descent rule, it achieves the error minimization process. The creation functions are selected by the sigmoid and linear functions for hidden and out layers. MATLAB 16 version trained and tested the developed ANN. To frame the best ANN design, it can use distinct neurons in the hidden layer; their activation and correlation coefficient are listed in Table 1. In view of the correlation coefficient esteem (R) picked the best system. It is clear from Table 1 that the R esteem is not expanded past the 18 neurons and one concealed layer arranges engineering. ANN modelling selects the network with one hidden layer and eight neurons the best. The taken network training performance, achieved with an error rate of 131.8787 and 12 epochs, and is exposed in Fig. 8. The values of correlation coefficient have been in between −1 and +1. If R value near to + 1 the relationship is positive, and it is nearer to −1 the relationship is negative. Levenberg Marquardt (LM), correlation coefficient (R), Sigmoid (sig), Linear (ln), RMSE, MRE. For characterizing the system performance uses the mean relative error (MRE) and RMSE.

630

R. Jyothu Naik and K. Thirupathi Reddy

Fig. 8 ANN values verse experimental values for the engine performance data

The error found during the learning method is the RMSE and is defined as:  n RMSE =

i=1 (E i

− Pi )2

1/2

n

(1)

The MRE, which expressions the mean ratio between the experimental values and the error is defined as:  n  (E i − Pi )  1  100 ∗ MRE(%) =  n i=1  Ei

(2)

where n is the quantity of the points in the data collection and E and P are experimental output and anticipated output sets (Fig. 9).

Fig. 9 ANN values verse experimental values for the engine emission data

Prediction of the WPPO Biodiesel Fuelled …

631

4 Conclusion Maximum thermal efficiency for the WPPO-5% blended with diesel HCCI operation observed as 42.12% at 413 K inlet temperature. The NOx emissions were reduced for all blends and later slightly increased. For the HCCI mode, smoke emission was observed below 0.1% that is almost negligible. The evolved ANN tool can expect the performance characteristics BTE and EGT with the ±5 error. Moreover, ANN of the emission characteristics together with NOx , smoke, CO and UHC within ±5 error. The ANN model can also be utilized for the HCCI engine restraint and testing. Future research, the analysis can be carried out by implementing fuzzy logic-GA-FAHP (Fuzzy Analytical hierarchy process) to get more accurate results.

References 1. Turkcan, A., Ozsezen, A.N., Canakci, M.: Effects of second injection timing on combustion characteristics of a two stage direct injection gasoline—alcohol HCCI engine (2013) 2. Saxena, S., Schneider, S., Aceves, S., Dibble, R.: Wet ethanol in HCCI engines with exhaust heat recovery to improve the energy balance of ethanol fuels (2012) 3. Zhen, X., Wang, Y.: Numerical analysis of knock during HCCI in a high compression ratio methanol engine based on LES with detailed chemical kinetics (2015) 4. Murugan, S., Ramaswamy, M.C., Nagarajan, G.: Assessment of pyrolysis oil as an energy source for diesel engines. Fuel Process Technol. 90, 67–74 (2009) 5. Maurya, Rakesh Kumar, Agarwal, Avinash Kumar: Experimental study of combustion and emission characteristics of ethanol fuelled port injected homogeneous charge compression ignition (HCCI) combustion engine. Appl. Energy 88(4), 1169–1180 (2011) 6. Özener, Orkun, Yüksek, Levent, Özkan, Muammer: Artificial neural network approach to predicting engine-out emissions and performance parameters of a turbo charged diesel engine. Therm. Sci. 17(1), 153–166 (2013) 7. O˘guz, Hidayet, Sarıtas, Ismail, Baydan, Hakan Emre: Prediction of diesel engine performance using biofuels with artificial neural network. Expert Syst. Appl. 37(9), 6579–6586 (2010) 8. Kiani, M., Kiani, D., et al.: Application of artificial neural networks for the prediction of performance and exhaust emissions in SI engine using ethanol-gasoline blends. Energy 35(1), 65–69 (2010)

Dimensional Analysis of Form Drilling Parameters by Buckingham Pi Theorem and Optimization of Heat Generation in Form Drilling Process by Taguchi Y. Bhargavi and V. Diwakar Reddy

Abstract Form drilling, a novel method of hole-making process, is performed on aluminum 8011 work material with tungsten carbide conical tool. The thrust force and torque are measured by varying the process parameters such as speed, feed, diameter of tool, thickness of work material, and magnesium powder. Temperature of the workpiece during the operation is measured by infrared thermometer. The objective of this study is to identify the most effective parameters which give a cylindricalshaped bushing without significant radial fracture or petal formation and effect of the heat generated during bushing formation in form drilling. Taguchi method is applied to optimize the influencing parameters in form drilling for heat generation. The influencing parameter affecting heat generated in form drilling is investigated by using ANOVA, and the same is confirmed by confirmation test of Taguchi method of analysis. The experimental results show that the process variable speed influence is more followed by feed. It is observed that Mg powder, workpiece thickness, tool diameter, speed, and feed affect the petal height formation by 2.98, 0.77. 85.85, 1.83, and 1.68%, respectively. Among all the process variables, tool diameter influence is more, and after the tool diameter, the influencing parameter is Mg powder.

1 Introduction Form drilling is also known as friction drilling, flow drilling, thermal drilling, and friction stir drilling. Form drilling is a non-traditional hole-making process. Heat generated from friction between rapid rotating conical tool and work material is used to soften the work material and penetrate a hole [1–4]. It forms a bushing in situ from the thin-walled work material and is a clean, chipless process. The purpose of Y. Bhargavi (B) · V. Diwakar Reddy Mechanical Engineering Department, Sri Venkateswara University College of Engineering, Tirupati, Andhra Pradesh, India e-mail: [email protected] V. Diwakar Reddy e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_61

633

634

Y. Bhargavi and V. Diwakar Reddy

Fig. 1 Illustration of stages in form drilling

the bushing is to increase the thickness for threading and available clamp load. The length of the bushing is two to three times the original thickness of the work material. This process is a dry drilling process, and hence unlike traditional drilling, cutting fluids and coolants are not used. Figure 1 shows various stages of form drilling. In stage 1, the conical tool comes in contact with workpiece. In stage 2, the tool has nearly penetrated the workpiece, and in this stage, the thrust force is peak and slightly bushes are formed. Stage 3 shows the work material ductility encompassing the tool tip. Extrusion of materials sideward to form boss can be identified. In stage 4, the tool penetrates workpiece, and the bushing of required hole is formed. In stage 5, the tool is retracted from the workpiece.

2 Literature Survey Miller et al. [1] observed thrust force and torque decreased by preheating the workpiece temperature and bushing shape improvement with increased workpiece temperature. Miller et al. [2] studied the tool wear due to friction. Chow et al. [5] studied the machining characteristics of friction drilling on AISI 304 stainless steel. France et al. [6–8] investigated the strength characteristics of friction-drilled holes in metal tubes. Ku et al. [9] investigated the thermal friction drilling effects on surface roughness and bushing length, and the machining characteristics of the process were improved. Skovron et al. [10] observed the reduction of process time by preheating the material. Su et al. [11] investigated counter bore die to improve the petal formation without cracks. Kurt et al. [12] studied to utilize the Taguchi methods to optimize surface roughness and hole diameter accuracy in dry drilling of Al 2024. El-Bahloul et al. [13] investigated experimentally the optimal process parameters of thermal friction drilling process. Raju et al. [14] investigated the finite element simulation of a friction drilling process using Deform-3D. Miller et al. [15] studied tool wear in friction drilling. They quantified the tool characteristics by measuring its weight change and detected changes in its shape with a coordinate measuring machine. They also observed wear damage by using scanning electron microscopy. Gopichand et al. [16] performed numerical simulation and analysis of friction drilling process for alumina alloy using ANSYS.

Dimensional Analysis of Form Drilling Parameters by Buckingham …

635

Fig. 2 Center and cone region of tool

Fig. 3 Form drill tools

Tool Selection See Figs. 2 and 3.

3 Experimental Setup A LMW JV-55 CNC vertical machining center was used for the friction drilling of the work material Al 8011. The work material was held in a vice on the bed of the machine, and the tool was held by a standard collets/tool holder as shown in Fig. 4a. The program is feed to perform form drilling according to the design of experiments L18, by varying the process parameters as shown in Table 1. After running the form drilling, the tip of tool mates with workpiece and heat generated by frictional force between rotational tool and workpiece to soften, create a hole, and penetrate bushing. During this process, thrust force and torque were measured by using dynamometer. The temperature during form drilling was measured by infrared thermometer. The sectional view of the bushing is as shown in Fig. 4d.

636

Y. Bhargavi and V. Diwakar Reddy

Fig. 4 a Experimental setup in form drilling, b infrared thermometer, c dynamometer, and d crosssectional view of bush formation

Table 1 Process parameters

Parameter

Level 1

Level 2

Level 3

Mg powder

With

Without



Workpiece thickness (δ)

2

3

4

Tool diameter (D)

4

6

10

Speed (V )

2000

3000

4000

Feed (f )

0.1

0.2

0.3

4 Pi Terms Evaluated 4.1 Process Parameters The pi terms influencing the form drilling phenomenon are only four which are listed below. The list of evaluated pi terms is shown in Table 2. π2 =

f FD δ ; π4 = V t; π5 = ; π6 = D T D

Among the six pi terms, only four pi terms are influencing which are stated above. The evaluated pi terms are used to perform regression analysis and to establish equation for heat generated (Graphs 1, 2, 3, and 4).

Dimensional Analysis of Form Drilling Parameters by Buckingham …

637

Table 2 List of pi terms Exp. no.

π1

π2

π3

π4

π5

π6

Q T

f D

K t Dφ T

Vt

FD T

δ D

Torque (T ) (N-m)

Heat generated (Q) (J)

1

586.4286

0.035

418.608

93.333

2.885714

0.5

1.4

821

2

471.2368

0.0333

311.85

75

4.760526

0.333

1.52

716.28

3

502.6503

0.03

396.9443

80

6.721311

0.2

1.83

919.85

4

601.0867

0.05

397.8509

95.6667

3.49333

0.75

1.5

901.63

757.1176

0.05

120.5

4.771765

0.5

1.7

1287.02

233.33

5.2

0.3

1.65

2419.02

5

455.2603

6

1466.073

0.01

1265.918

7

1074.422

0.025

490.1349

171

2.468027

1.0

1.47

1579.4

0.0333

367.6616

124.667

4.159509

0.667

1.63

1308.11

8

802.5215

9

423.0643

0.03

666.6

10

787.4933

0.075

296.6189

11

607.3727

0.0166

596.9281

12

942.4757

0.02

987.1784

691.1513

0.075

339.57

110

3.573686

0.0166

538.65

166.667

3.195

13 14

1047.188

9.19883

0.4

1.71

723.44

125.33

67.333

3.213333

0.5

1.5

1181.24

96.67

3.91677

0.333

1.67

977.87

6.864865

0.2

1.85

1743.58

0.75

1.52

1050.55

0.5

1.6

1675.5

150

15

567.574

0.02

913.1096

90.333

8.08284

0.3

1.69

16

1135.172

0.05

442.3169

180.667

3.05655

1.0

1.45

75.667

5.368241

0.666

1.71

4.92228

0.4

1.93

17

475.4386

0.05

468.528

18

879.6425

0.01

900.5409

140

959.2 1646 813 1697.71

Graph 1 π2 versus heat generated at operation region

5 Results and Discussion Analysis of variance (ANOVA) has been studied, and the significant parameters affecting these responses are discussed below. The percentage (%) in ANOVA table indicates the significance rate of the process parameters on the quality characteristic.

638

Graph 2 π4 versus heat generated at operation region

Graph 3 π5 versus heat generated at operation region

Graph 4 π6 versus heat generated at operation region

Y. Bhargavi and V. Diwakar Reddy

Dimensional Analysis of Form Drilling Parameters by Buckingham …

639

Table 3 Response table for heat generation—ANOVA Parameters

DOF

Adj SS

Adj MS

F-value

P-value

Mg powder

1

0.08335

0.08335

1.68

0.231

Contribution (%) 3.93

Workpiece thickness

2

0.19808

0.09904

2.00

0.198

9.34

Tool diameter

2

0.09289

0.04644

0.94

0.431

4.38

Speed

2

0.90142

0.45071

9.08

0.009

42.49

Feed

2

0.44860

0.22430

4.52

0.049

21.15

Error

8

0.39694

0.04962







Total

17

2.12128









5.1 Analysis of Variance for Heat Generation The percent numbers depict that speed has significant effect on heat generated. From Table 3, it is observed that Mg powder, workpiece thickness, tool diameter, speed, and feed affect the heat generation by 3.93, 9.34, 4.38, 42.49, and 21.15%, respectively. Among all the process variables, speed influence is more followed by feed. The effect of magnesium powder is very low for heat generation in form drilling process.

5.2 Analysis of Variance for Petal Height The percent numbers depict that tool diameter has significant effect on petal height formation. From Table 4, it is observed that Mg powder, workpiece thickness, tool diameter, speed, and feed affect the petal height formation by 2.98, 0.77, 85.85, 1.83, and 1.68%, respectively. Among all the process variables, tool diameter influence is more followed by Mg powder. Table 4 Response table for petal height—ANOVA Parameters

DOF

Adj SS

Adj MS

F-value

P-value

Contribution (%)

Mg powder

1

0.00383

0.00383

3.45

0.100

2.98

Workpiece thickness

2

0.00098

0.00049

0.45

0.656

0.77

Tool diameter

2

0.11055

0.05527

49.77

0.000

85.85

Speed

2

0.00235

0.00117

1.06

0.391

1.83

Feed

2

0.00216

0.00108

0.97

0.418

1.68

Error

8

0.00888

0.00111







Total

17

0.12877









640

Y. Bhargavi and V. Diwakar Reddy

Table 5 S/N ratios for responses for heat generation—Taguchi analysis Exp. no.

Mg powder

Workpiece thickness in mm

1

With

2

2

With

2

3

With

4

With

5

With

6

With

7

Tool diameter

Speed

Feed

Heat generation

S/N ratio

4

2000

0.1

821.00

58.2869

6

3000

0.2

716.28

57.1017

2

10

4000

0.3

919.85

59.2743

3

4

2000

0.2

901.63

59.1006

3

6

3000

0.3

1287.10

62.1922

3

10

4000

0.1

2419.02

67.6728

With

4

4

3000

0.1

1579.40

63.9698

8

With

4

6

4000

0.2

1308.11

62.3329

9

With

4

10

2000

0.3

723.44

57.1881

10

Without

2

4

4000

0.3

1181.24

61.4468

11

Without

2

6

2000

0.1

977.87

59.8056

12

Without

2

10

3000

0.2

1743.58

64.8288

13

Without

3

4

3000

0.3

1050.55

60.4283

14

Without

3

6

4000

0.1

1675.50

64.4829

15

Without

3

10

2000

0.2

959.20

59.6382

16

Without

4

4

4000

0.2

1646.00

64.3286

17

Without

4

6

2000

0.3

813.00

58.2018

18

Without

4

10

3000

0.1

1697.71

64.5973

The bold values identifies the maximum heat generated value to the maximum S/N ratio

5.3 Taguchi Analysis for Heat Generation From Table 5, it is observed that the maximum heat generation is obtained at speed = 4000 rpm, feed = 0.1 rev/min, tool diameter = 10 mm, and workpiece thickness = 3 mm and with magnesium powder. Table 7 shows the Taguchi analysis for responses; in this, speed is the most influencing, and the next influencing parameter is feed. The ranks obtained for Mg powder, workpiece thickness, tool diameter, speed, and feed are 5, 3, 4, 1, and 2 (Graph 5; Tables 5 and 6).

5.4 Taguchi Analysis for Petal Height From Table 7, it is observed that the petal height formation is maximum at speed = 2000 rpm, feed = 0.1 rev/min, tool diameter = 10 mm, and workpiece thickness = 4 mm and with magnesium powder. Table 8 shows the Taguchi analysis for responses; tool diameter is the most influencing, and the next influencing parameter is Mg

Dimensional Analysis of Form Drilling Parameters by Buckingham …

641

Main Effects Plot for SN ratios Data Means

thickness

mg

speed

diameter

feed

Mean of S/N ratios

63

62

61

60

59

w

wo

2

3

4

4

6

10

2000

3000

4000

0.1

0.2

0.3

Signal-to-noise: Larger is better

Graph 5 S/N ratios for heat generation

Table 6 Response table for heat generation Level

Mg powder

Workpiece thickness in mm

Tool diameter

Speed

Feed

1

60.79

60.12

61.26

58.70

63.14

2

61.97

62.25

60.69

62.19

61.22

3



61.77

62.20

63.26

59.79

Delta

1.18

2.13

1.51

4.55

3.35

Rank

5

3

4

1

2

powder. The ranks obtained for Mg powder, workpiece thickness, tool diameter, speed, and feed are 2, 5, 1, 3, and 4, respectively, according to their influence on the response petal height. It is confirmed that the results obtained from the ANOVA are coincided with the results of Taguchi analysis (Graph 6).

6 Conclusion In form drilling, the heat is generated due to frictional force between rotational tool and workpiece. From the result, maximum heat generation is 2419 joules at speed 4000 rpm, feed 0.1 mm/rev, tool diameter 10 mm, and workpiece thickness 3 mm when magnesium powder is used. From Taguchi analysis, the most influencing parameter for heat generation is speed followed by feed. From ANOVA results, contribution of speed for heat generation is 42.49%. From ANOVA result, contribution of tool diameter for petal height formation is 85.85%. From the output responses,

642

Y. Bhargavi and V. Diwakar Reddy

Table 7 S/N ratios for responses for petal height—Taguchi analysis Exp. no.

Mg powder

Work piece thickness

1

With

2

2

With

2

3

With

4

With

5

With

6

With

7

Tool diameter

Speed

Feed

Petal height

S/N ratio

4

2000

0.1

3.4

10.6296

6

3000

0.2

5.22

14.3534

2

10

4000

0.3

5.68

15.0870

3

4

2000

0.2

2.48

7.8890

3

6

3000

0.3

4.51

13.0835

3

10

4000

0.1

6.92

16.8021

With

4

4

3000

0.1

2.92

9.3077

8

With

4

6

4000

0.2

4.99

13.9620

9

With

4

10

2000

0.3

7.25

17.2068

10

Without

2

4

4000

0.3

2.50

7.9588

11

Without

2

6

2000

0.1

4.04

12.1276

12

Without

2

10

3000

0.2

6.27

15.9454

13

Without

3

4

3000

0.3

2.96

9.4258

14

Without

3

6

4000

0.1

4.80

13.6248

15

Without

3

10

2000

0.2

4.25

12.5678

16

Without

4

4

4000

0.2

2.76

8.8182

17

Without

4

6

2000

0.3

3.93

11.8879

18

Without

4

10

3000

0.1

6.44

16.1777

The bold values identifies the maximum petal height value to the maximum S/N ratio

Table 8 Response table for petal height Level

Mg powder

Work piece thickness in mm

1

13.147

12.684

2

12.059

12.232 12.893

3

Tool diameter

Speed

Feed

9.005

12.051

13.112

13.173

13.049

12.256

15.631

12.709

12.442

Delta

1.087

0.661

6.626

0.997

0.856

Rank

2

5

1

3

4

it is observed that if the diameter increases, the petal height also increases for same workpiece thickness (i.e., mass of the material constant). The maximum temperature obtained in the form drilling is 363 K which is tolerable by system. Because of this, extended tool life is obtained when friction drilling process is adopted in comparison with conventional twist drilling process. The fourth pi term (i.e., Vt) is the most influencing parameter, and to control the process, it is significant. The accuracy in producing the holes is higher than conventional drilling process; the circularity of the produced holes is good. In form drilling, heat generation is more; because of this,

Dimensional Analysis of Form Drilling Parameters by Buckingham …

643

Main Effects Plot for SN ratios Data Means

mg

16

thickness

diameter

feed

speed

Mean of S/N ratios

15 14 13 12 11 10 9 w

wo

2

Signal-to-noise: Larger is better

3

4

4

6

10

2000

3000

4000

0.1

0.2

0.3

Graph 6 S/N ratios for petal height

it makes the flow of material higher which increases the petal formation capability and decreases the machining time.

References 1. Miller, S.F., Li, R., Wang, H., Shih, A.J.: Experimental and numerical analysis of the friction drilling process. ASME J. Manuf. Eng. (2004). Submitted for publications 2. Miller, S.F., Blau, P., Shih, A.J.: Micro structural alterations associated with friction drilling of steel, aluminum, and titanium. J. Mater. Eng. Perform. (2005). Accepted for publication 3. van Geffen, J.A.: Piercing tool. U.S. Patent No. 3,939,683 (1976) 4. van Geffen J.A.: Method and apparatuses for forming by frictional heat and pressure holes surrounded each by a boss in a metal plate or the wall of a metal tube. U.S. Patent No. 4,175,413 (1979) 5. Chow, H.-M., Lee, S.-M., Yang, L.-D.: Machining characteristic study of friction drilling on AISI 304 stainless steel. J. Mater. Process. Technol. 207, 180–186 (2008) 6. France, J.E., Davidson, J.B., Kirby, P.A.: Strength and rotational stiffness of simple connections to tubular columns using flowdrill connectors. J. Constr. Steel Res. 50(1), 15–34 (1999) 7. France, J.E., Davidson, J.B., Kirby, P.A.: Moment-capacity and rotational stiffness of endplate connections to concrete-filled tubular columns using flowdrill connectors. J. Constr. Steel Res. 50(1), 35–38 (1999) 8. France, J.E., Davidson, J.B., Kirby, P.A.: Strength and rotational response of moment connections to tubular columns using flowdrill connectors. J. Constr. Steel Res. 50(1), 1–14 (1999) 9. Ku, W.L., Hung, C.L., Lee, S.M., Chow, H.M.: Optimization in thermal friction drilling for SUS 304 stainless steel. Int. J. Adv. Manuf. Technol. 53(9–12), 935–944 (2011)

644

Y. Bhargavi and V. Diwakar Reddy

10. Skovron, J.D., Rohan Prasad, R., Ulutan, D., Mears, L., Detwiler, D., Paolini, D., Baeumler, B., Claus, L.: Effect of thermal assistance on the joint quality of Al6063-T5A during flow drill screwdriving. J. Manuf. Sci. Eng. 137(051019), 1–8 (2015) 11. Su, K.-Y., Welo, T., Wang, J.: Improving friction drilling and joining through controlled material flow. In: 46th SME North American Manufacturing Research Conference, NAMRC 46, Texas, USA, pp. 663–670 (2018) 12. Kurt, M., Bagci, E., Kaynak, Y.: Application of Taguchi methods in the optimization of cutting parameters for surface finish and hole diameter accuracy in dry drilling processes. Int. J. Adv. Manuf. Technol. 40, 458–469 (2009) 13. El-Bahloul, S.A., El-Shourbagy, H.E., El-Bahloul, A.M., El-Midany, T.T.: Experimental and thermo-mechanical modeling optimization of thermal friction drilling for AISI 304 stainless steel. CIRP J. Manuf. Sci. Technol. CIRPJ (2017) 14. Raju, B.P., Swamy, M.K.: Finite element simulation of a friction drilling process using deform3D. Int. J. Eng. Res. Appl. (IJERA) 2(6), 716–721 (2012). ISSN: 2248-9622 15. Miller, S.F., Blau, P.J., Shih, A.J.: Tool wear in friction drilling. Int. J. Mach. Tools Manuf 47, 1636–1645 (2007) 16. Gopichand, A., Brahmam, M.V., Bhanuprakash, D.: Numerical simulation and analysis of friction drilling process for alumina alloy using Ansys. Int. J. Eng. Res. Technol. (IJERT) 3(12) (2014). ISSN: 2278-0181 17. Prabhu, T., Arulmurugu, A.: Experimental analysis of friction drilling on aluminium and copper. Int. J. Mech. Eng. Technol. (IJMET) 5(5), 130–139 (2014). ISSN 0976–6340 (Print), ISSN 0976–6359 (Online)

Comparison of Ductile, Flexural, Impact and Hardness Attributes of Sisal Fiber-Reinforced Polyester Composites S. Sathees Kumar, V. Mugesh Raja, B. Sridhar Babu and K. Tirupathi

Abstract In this effort, mechanical attributes of sisal fiber-strengthened polyester material have been carried out with differing the fiber volume division of 10:4 and 10:5% separately. These outcomes show that the including of sisal fiber composites expanding the malleable, flexural, impact and hardness. The mechanical characteristics of sisal fiber composite material outcomes are contrasted and unsaturated polyester composites. The flexural and effect qualities accomplished from the sisal composites are up to 13.5% and 13.6 separately. The morphological examination of the sisal fiber composite example was breaking down through scanning electron microscopy (SEM). This sisal fiber mechanical outcome is additionally corresponded to the past work results.

1 Introduction For the most part, numerous sorts of filaments are bounteously accessible in the nature, for example, jute [1], sisal [2], coir, kenaf, oil palm [3], bamboo, ramie and flax straw and so forth. Common fiber demonstrated a successful and effectively accessible support material in the thermoset and thermoplastic frameworks. The investigation of mechanical properties, particularly interfacial exhibitions of the composites, depends on normal filaments because of the inadequate interaction holding among the hydrophilic common strands and the hydrophobic polymer frameworks. Having regard to this study, an improved strategy was intended that could further assess the interfacial attributes between normal fiber and polymeric lattices [4]. The S. S. Kumar (B) · B. Sridhar Babu · K. Tirupathi Department of Mechanical Engineering, CMR Institute of Technology, Hyderabad, Telangana, India e-mail: [email protected] V. Mugesh Raja Department of Mechanical Engineering, University College of Engineering, Ramanathapuram, Tamil Nadu, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_62

645

646

S. S. Kumar et al.

analysis of the life cycle ecological quality of traditional fiber mixtures with glass fiber reinforced composites and found that standard fiber combinations are naturally predominant in the specific purposes [5]. The sugar stick bagasse squanders as fortification to polymeric tars for manufacture of ease composites. They announced that composites with consistent microstate could be manufactured, and mechanical properties like wooden bundles can be accomplished [6]. Hassan et al. [7] have changed over the bagasse into a thermo-shapeable material through esterification of the fiber framework. The aspects’ steadiness and mechanical attributes of the composites arranged from the esterified filaments were accounted for in this effort. The practical use of existing filaments in polymeric materials as fortifications. The physicochemical, warm and mechanical attributes of the modern fiber fortifications were estimated and contrasted and those of other regular [8]. Safeguarding Sun et al. [9] recognized the characteristic bamboo fiber over Fourier transform infrared spectroscopy (FTIR), second subsidiary IR spectra and binary attributes infrared connection spectroscopy. All the artificially treated plant strands demonstrated comparable IR ghastly profiles. In their second subordinate IR spectrum, a few contrasts intermediate to bamboo fiber and additional plant fibers gave separation results. Tofanica et al. [10] inspected the concoction creation, fiber qualities and appropriateness of rapeseed stalks for the assembling of cellulosi, with its pertinence in mash and paper produce. Niranjan et al. [11] proposed a characteristic fiber composite, made of abaca fiber which goes about as a fortifying operator, wherewith epoxy pitch was the grid. Issues such as the need to improve the collection of useless manufacturing and the poor mechanical characteristics of polymers can be solved by providing polymer/natural blends with fortification efficiency tweaked from different beginnings using usual [12]. Some natural can be contrasted with manufactured ones, for example, fiberglass and carbon fiber, as far as support properties in composites, other than exhibiting diminished well-being perils, low thickness and high adaptability [13, 14], and the objectives were for the most part improve mechanical obstruction. The mechanical qualities of a polymer composite fortified by common are chiefly aftereffect of the amount and type, other than the associate quality among support and lattice. A choice to adjust the composite mechanical execution is creating half and half composites, a material delivered by mix of at least two kinds of reinforcement [15].

1.1 Materials and Methods 1.1.1

Sisal Fiber

Sisal is gathered from a plant Veracruz and is accessible in southern pieces of India. The sisal fiber was acquired by GVR enterprises, Madurai. It was cut somewhere in the range of 0.5 and 1.0 cm of length. The physical and mechanical attributes of sisal are given in Table 1. The sisal plants are isolated from sisal leaves by physically and furthermore by precisely. The sisal leafs are cut from sisal plant and tied into bundles by using bags. Then bags contain the sisal leafs are retted in tanks or River or Well

Comparison of Ductile, Flexural, Impact and Hardness Attributes …

647

Table 1 Characteristics of sisal fiber Fiber

Thickness (mm)

Solidity (g/cc)

Cellulose/lignin content (%)

Elastic modulus (GN/m2 )

Tenacity (MN/m2 )

Elongation (%)

Ductile strength (MPa)

Flexural modulus (GPa)

Sisal

50–200

1.45

67/12

9–16

568–640

3–7

54

2.5

Fig. 1 Raw sisal fiber

for 3–4 days. The retted leafs are washed in running water and the top portion of the leafs are removed by manually (May by removed mechanically) to get the fiber separately and cleaned and dried in the sun. Figure 1 demonstrates the unrefined sisal fiber.

1.2 Fabrication of Specimen The sisal leaves are cut from sisal plant and integrated with groups by utilizing packs. The accompanying advances were utilized to remove the sisal fiber from crude sisal. (a) Sisal plant, (b) retting in water for 3–4 days, (c) evacuating the best bit of the leaves, (d) dried utilizing daylight and (e) last type of sisal. Figure 2 shows the sisal fiber removed from unrefined fiber.

1.2.1

Preparation of Fibers

Sisal fiber can be separated from its foliage by retting, bubbling and mechanical removal techniques. The sisal is cleansed and extremely handled. This procedure proceeds 15–21 days for a solitary sequence of abstraction and debases the nature of fiber.

648

S. S. Kumar et al.

Fig. 2 Sisal fiber

1.2.2

Preparation Method of Specimen

The mold used in this work was made of well-seasoned plywood of 250×250×3 mm dimensions with beadings. Fabrication of the composite material was done in this mold by the hand lay-up process. The top and bottom surfaces of the mold and walls were coated with wax, remover (PVA) and allowed to dry. The functions of the top and bottom plates are to be covered. The mold is placed at atmosphere condition for preparing to make a composite material. The sisal filaments are sliced into 20-mm pieces disseminated consistently at the base of the shape which is set up previously. The created example is appeared in Fig. 3.

Fig. 3 Fabricated specimen

Comparison of Ductile, Flexural, Impact and Hardness Attributes …

649

2 Mechanical Testing 2.1 Tensile Strength The ductile test is directed on a Tinius Olsen UTM 10 KN all-inclusive testing equipment with a measure length of 75 mm, and crosshead speed of the machine is set at 5 mm min−1 . The sample estimate for malleable test is 115 mm length, 20 mm width and 3 mm thickness as per ASTM D638-01. The energy rate was kept consistent at a rate of 2 mm/min for all samples.

2.2 Flexural Strength The flexural test was performed on UTM by three-point bending test. Every one of the tests was led according to the ASTM standard D790-02 with 120 mm length, 15 mm width and 3 mm thickness measurements. Various samples had been tried, and interpretations were taken.

2.3 Impact Strength and Hardness The impact test samples are set up as per the required measurement following the ASTM D256 standard. A rectangular bit of 65 ± 2 mm long and width 15 ± 0.2 mm and thickness of 3 mm was readied. The test was accomplished on general testing machine. The ASTM D2240 standard example was done in the Rockwell hardness analyzer. Hardness esteems are estimated by R scales.

3 Scanning Electron Microscope (SEM) The face attributes of the mixture material were done at the environment temperature to know the fiber—matrix boundary and component of disappointment of the syntheses utilizing SEM. The examples obtaining the most extreme and least estimations of rigidity were chosen for morphological examination. Before the morphological analysis was finished, experiments were selected to improve the conductivity.

650

S. S. Kumar et al.

4 Results and Discussions 4.1 Ductile Properties Rigidity and elastic modulus of sisal fiber-strengthened mixtures are organized in Table 2; what’s more, the chart is plotted with comparing information in Fig. 4. Ductile properties of the composite 10:5 wt. were considered most extreme to be contrasted with 10:4 wt. sisal composites. It was seen that there were 10:5 wt. proportion great upgrades in elasticity and malleable modulus of sisal fiber. It was observed slightly small in any case 10:4 wt. compared with 10:5. Further increment in the fiber stacking results in an expansion in properties. At lesser fiber stacking, scattering of fiber is weak, so stress exchange will not happen legitimately. At maximum fiber stacking, nearby is a solid inclination for fiber–fiber interface.

4.2 Flexural Properties Flexural quality of sisal-strengthened composites is organized in Table 3. Furthermore, relating information is plotted in Fig. 4. The aftereffect of flexural test demonstrates the beneficial outcome of sisal fiber reinforcement into the polyester matrix. Table 2 Tensile attributes of sisal fiber-reinforced composites Weight ratio

Ductile strength (MPa)

10:4

20

10:5

21.1

Fig. 4 Ductile, flexural, impact strength and robustness of sisal fiber-fortified polyester composites

Ductile modulus (MPa)

Break force (N)

Elongation (%)

658

2723

2.62

672

2760

2.70

Comparison of Ductile, Flexural, Impact and Hardness Attributes …

651

Table 3 Ductile attributes of sisal fiber-reinforced composites Weight ratio

Ductile strength (MPa)

Flexural strength (MPa)

Impact strength (kJ/m2 )

References

Unsaturated polyester

23.4

56.7

19.1

[17]

Polyester/sisal (10:4)

20

61.9

21

Present work

Polyester/sisal (10:5)

21.1

64.4

21.7

Present work

Fig. 5 Ductile and flexural modulus of sisal fiber-reinforced polyester composites

The estimation of flexural quality and flexural modulus discovered that the limit of the composite was 10:5 wt. The flexural modulus of sisal fiber composites is appeared in Fig. 5.

4.3 Impact Properties Figure 4 exhibits the variety of effect quality of sisal fiber-strengthened polyester compounds having fiber content by hand lay strategy. From the figure, plainly the effect quality of sisal fiber-fortified composites 10:5 is smaller than that of the 10:4 wt. sisal fiber-strengthened composites. The least esteem is seen for the sisal fiber-fortified syntheses; owed to the fiber/substance attachment, all things considered decides the quality of composites. Improved collaboration can prompt immaculate holding, and consequently, the disappointment of the composites can happen at generally low effect. The effect quality of sisal fiber composites is appeared in Table 3.

652

S. S. Kumar et al.

Table 4 Mechanical attributes of the previous work results compared with the present work Fiber content

Ductile strength (MPa)

Flexural strength (MPa)

Sisal/unsaturated polyester composites

19.6

54.9

Admicellar-treated sisal

78

Sisal fiber-reinforced urea formaldehyde resin composites Banana/sisal-reinforced hybrid composites Glass fiber/sisal fiber composites



19.93 176.2

Mustard cake and pine needle-reinforced sisal fiber composites

41.45

Sisal fiber-reinforced polyester composites

21.1

Impact strength (J/m)

Hardness

References





[14]

12.5

72.5

[15]

58.58



72 HR

[16]

60.86





[17]

18.67



[18]

8



[19]

21.7

92

Present work

102

– –

64.4

In this work results are high compared with sisal unreinforced polyester composites and sisal reinforced natuaral fibers. The comparision values are represeted in bold

4.4 Hardness Properties Hardness of a certain example denotes its stiffness or obstruction of being cracked to have its profile varied everlastingly when load is connected to it. Solidity of a compound relies upon the scattering of the stuffing into the matrix. For the most part, the nearness of an increasingly adaptable matrix makes the resultant composites show lower hardness. The estimation of hardness discovered that limit of the composite was 10:5 wt. This could be because of the milder appropriation of sisal into the matrix with diminish of hollowness and more grounded amalgamate bond to the grid. The stability characteristics of sisal fiber syntheses are appeared in Fig. 4. Mechanical attributes of the past work results contrasted and the present work are appeared in Table 4.

5 Analysis of SEM The exterior attributes of the composite material utilized for the examination are analyzed as a result of SEM. The picture of the malleable break compound material of sisal fiber is exhibited in Fig. 6a, b. Figure 6a demonstrates the interrelationship

Comparison of Ductile, Flexural, Impact and Hardness Attributes …

653

Fig. 6 SEM image of a 10:4 wt. ratio, b 10:5 wt. ratio

between the sisal fiber and polyester resins. Ductile break picture of 10:4 wt. proportion demonstrates a few splits and voids development. Because the inappropriate scattering of the sisal and polyester tar might be made the miniaturized scale splits in the internal area of the composites. This might made negative impacts in the 10:4 wt. proportion of sisal and polyester composites. In Figure 6b, 10:5 sisal fiber shows the strong interactions between the fibers/matrix and arranged in the two-layer longitudinal directional alignment of the fiber which indicates high ductile load-carrying capacity and less delamination. The SEM picture in Fig. 6b unmistakably uncovers the correct dispersion of polyester resin and sisal fiber composites. Because of proper scattering of above resin and fiber, it might give great positive outcomes.

6 Conclusions The sisal fiber composite examples are manufactured by hand layup method. The sisal fiber composites are exposed to mechanical testing, for example, elastic, flexural, impact test and hardness test. The outcomes showed that the sisal composite material of 10:5 wt. indicates greatest malleable and flexural quality that can hold the quality up to 21.1 and 64.4 MPa separately. The effect quality gotten from the sisal fiber composite material of 10:5 wt. has the most extreme estimation of 21.7 J/m. The inner structure and inside fractures and fiber gaps are watched for the broken surfaces of the tested samples utilizing SEM. This kind of composite materials can be helpful for automobile applications.

654

S. S. Kumar et al.

References 1. Varma, I.K., Anantha Krishnan, S.R., Krishnamoorthy, S.: Composites of glass/modified jute fabric and unsaturated polyester resin. Composites 20, 383–388 (1989) 2. Joseph, K., Thomas, S., Pavithran, C.: Effect of ageing on the physical and mechanical properties of sisal-fiber-reinforced polyethylene composites. Compos. Sci. Technol. 53, 99–110 (1995) 3. Jacob, M., Thomas, S., Varughese, K.T.: Mechanical properties of sisal/oil palm hybrid fiber reinforced natural rubber composites. Compos. Sci. Technol. 64, 955–965 (2004) 4. Li, Y., Mai, Y.W., Lin, Y.: Sisal fibre and its composites: a review of recent developments. Compos. Sci. Technol. 60, 2037–2055 (2000) 5. Joshi, S.V., Drzal, L.T., Mohanty, A.K., Arora, S.: Are natural fiber composites environmentally superior to glass fiber reinforced composites? Compos. Part A Appl. Sci. Manuf. 35, 371–376 (2004) 6. Monteiro, S.N., Rodriquez, R.J.S., De Souza, M.V., d’Almeida, J.R.M.: Sugar cane bagasse waste as reinforcement in low cost composites. Adv. Perform. Mater. 5, 183–191 (1998) 7. Hassan, M.L., Rowell, R.M., Fadl, N.A., Yacoub, S.F., Christainsen, A.W.: Thermoplasticization of bagasse. I. Preparation and characterization of esterified bagasse fibers. J. Appl. Polym. Sci. 76, 561–574 (2000) 8. Mayandi, K., Rajini, N., Pitchipoo, P., Sreenivasan, V.S., Jappes, J.W., Alavudeen, A.: A comparative study on characterisations of Cissus quadrangularis and Phoenix reclinata natural fibres. J. Reinf. Plast. Compos. 34, 269–280 (2015) 9. Sun, B., Huang, A., Wang, Y., Liu, J.: Natural bamboo (Neosinocalamus affinis Keng) fiber identification using FT-IR and 2D-IR correlation spectroscopy. J. Nat. Fibers 12, 1–11 (2015) 10. Tofanica, B.M., Puitel, A.C., Gavrilescu, D.: Environmental friendly pulping and bleaching of rapeseed stalk fibers. Environ. Eng. Manag. J. 11 (2012) 11. Niranjan, R.R., Junaid Kokan, S., Sathya Narayanan, R., Rajesh, S., Manickavasagam, V.M., Vijaya Ramnath, B.: Fabrication and testing of abaca fibre reinforced epoxy composites for automotive applications. Adv. Mater. Res. 718, 63–68 (2013) 12. Kumar, S., Kumar, Y., Gangil, B., Patel, V.K.: Effects of agro-waste and bio-particulate fillers on mechanical and wear properties of sisal fibre based polymer composites. Mater. Today Proc. 4, 10144–10147 (2017) 13. Rana, R.S., Purohit, R.: A review on mechanical property of sisal glass fiber reinforced polymer composites. Mater. Today Proc. 4, 3466–3476 (2017) 14. Yusoff, R.B., Takagi, H., Nakagaito, A.N.: Tensile and flexural properties of polylactic acidbased hybrid green composites reinforced by kenaf, bamboo and coir fibers. Ind. Crops Prod. 94, 562–573 (2016) 15. Sangthong, S., Pongprayoon, T., Yanumet, N.: Mechanical property improvement of unsaturated polyester composite reinforced with admicellar-treated sisal fibers. Compos. Part A Appl. Sci. Manuf. 40, 687–694 (2009) 16. Zhong, J.B., Lv, J., Wei, C.: Mechanical properties of sisal fibre reinforced urea formaldehyde resin composites. Express Polym. Lett. 10, 681–687 (2007) 17. Venkateshwaran, N., ElayaPerumal, A., Alavudeen, A., Thiruchitrambalam, M.: Mechanical and water absorption behaviour of banana/sisal reinforced hybrid composites. Mater. Des. 32, 4017–4021 (2011) 18. Ramesh, M., Palanikumar, K., Hemachandra Reddy, K.: Mechanical property evaluation of sisal–jute–glass fiber reinforced polyester composites. Compos. Part B Eng. 48, 1–9 (2013) 19. Yoganandam, K., Ramshankar, P., Ganeshan, P., Raja, K.: Mechanical properties of alkalitreated Madar and Gongura fibre-reinforced polymer composites. Int. J. Ambient Energy. https://doi.org/10.1080/01430750.2018.1477066

Optimization of EDM Process Parameters Using Standard Deviation and Multi-objective Optimization on the Basis of Simple Ratio Analysis (MOOSRA) J. Anitha and Raja Das Abstract The process of decision making involves finding out all the attributes which are quite conflicting in nature and selecting the best alternative based on the choice of the decision maker. Multi-objective techniques can be used in the selection process. In this paper, a new multi-objective optimization method called multi-objective optimization on the basis of simple ratio analysis (MOOSRA) is used to find the best alternative. MOOSRA in combination with standard deviation is used as an improvement procedure. Standard deviation is applied to decide the weights that are used for normalizing the performance measures which are obtained from the experimental outcomes. Electric discharge machining (EDM) process has widely emerged as an outstanding method for cutting electrically conductive materials which are difficult to machine by any traditional machining process. Four EDM parameters, namely peak current (Ip), pulse on time (T on), duty cycle (T ) and voltage (V ) were considered as input parameters, and material removal rate (MRR) and surface roughness (Ra) are the output parameters. MRR and surface finish are quite contradicting in nature. Higher values of MRR are required to acquire high productivity and lower values of surface roughness are required to achieve better surface quality. The objective is to maximize MRR and minimize surface roughness. The aim of the current study is to recommend optimized input parametric combination to enhance the productivity and the quality.

1 Introduction Electric discharge machining (EDM) is a non-conventional machining process used on electrically conductive material irrespective of its hardness using thermal energy. In the dielectric environment, the metal is removed from the workpiece using a J. Anitha (B) · R. Das School of Advanced Studies, VIT University, Vellore, India e-mail: [email protected] R. Das e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_63

655

656

J. Anitha and R. Das

series of electrical sparks. EDM technology becomes indispensable because it can cut even the hard materials. In the recent years, EDM technology has brought many improvements in the machining process. It can machine intricate parts and even hard materials. In the present industrial scenario, the conventional machining process is replaced using modern machining process because of significant advantages. To optimize the performance characteristics, various optimization techniques have been discussed by Manish et al. [1]. Many problems in manufacturing industry involve simultaneous optimization methods which have conflicting objectives. Maximizing the profits, minimizing the production cost or transportation cost are some typical examples for problems involving multi-objective optimization techniques according to Chakraborty [2]. To solve multi-objective optimization problem, one of the methods is multi-criteria decisionmaking MCDM according to Jaimes et al. [3]. Multi-objective optimization on the basis of simple ratio analysis (MOOSRA) is MCDM technique. The steps in MOOSRA are quite similar to MOORA method. In MOOSRA method, the negative values do not appear as in MOORA method since it is less sensitive to large variations according to Jagdish and Ray [4]. This method was used by Bhowmik [5] for obtaining the optimum parameters for surface roughness. Later, this method was used for selecting the optimum fluid for gear hobbing process by Jagdish and Ray [4], non-traditional machining by Sarkar et al. [6] and selecting material by Kumar and Ray [7]. MOORA and standard deviation are used to optimize the gas metal welding parameters by Achebo and Odinikuku [8] and Muniappan et al. [9] for optimizing WEDM parameters. Adalı and I¸sık [10] used MOOSRA and MULTIMOORA for laptop selection problem. Based on the above literature, very less research is done using MOOSRA method. The novelty of this paper is standard deviation is used to find the relative importance of the output parameters material removal rate and surface roughness based on the experimental data, and further, MOOSRA method is used to recommend optimized input parametric combination to enhance the productivity and the quality.

2 Experimental Environment Electronica Electra plus PS 50ZNC die-sinking EDM machine is used for conducting the experiment, and AISI D2 tool steel is used as workpiece material having a rectangular shape with a thickness of 4 mm (with negative polarity) and density 7.7 g/cc. Electrolytic copper with a positive polarity and 30 mm in diameter is used as the electrode material, side flushing technique with pressure of 0.3 kg f/cm2 , and for the dielectric fluid, commercial grade EDM oil with freezing point of 94 °C and specific gravity of 0.76 are used. The four input process parameters: current, pulse on time, duty cycle and voltage along with the range of values are shown in Table 1. The experimental values of MRR and Ra are shown in Table 2.

Optimization of EDM Process Parameters Using Standard Deviation …

657

Table 1 Machining parameters along with units and levels Input parameters

Unit

Ip

A

Level 1 5

10

15

T on

µs

50

75

100

50

66.5

83

40

45

50

T V

V

Level 2

Level 3

3 Methodology 3.1 Optimization Problem If MRR = f 1 (Ip, T on, T, V ) and Ra = f 2 (Ip, T on, T, V ), then the multi-objective optimization problem is Maximize f 1 (Ip, T on, T, V ) and Minimize f 2 (Ip, T on, T, V ) Subject to 5 ≤ Ip ≤ 15 50 ≤ T on ≤ 100 50 ≤ T ≤ 83 40 ≤ V ≤ 50 And Ip, T on, T, V e R

3.2 Standard Deviation Concept for Weight Calculation Standard deviation is related to calculation of unbiased assignment of weights. Step 1: Calculate

X i1j =

X i j − min1< j= 0. –ve sample is categorized as +ve c) 3.

V=V

{x}. The –ve sample is accounted to the –ve training set.

return V

Fig. 4 Bootstrap pseudo code

known after a couple of phase of learning. The same equivalent techniques, classifiers β1 , β2 . . . β Z are adapted as well. In this way, in perspective on the testing unraveling, honest booster chain learning figure could be further enhanced with small changes on weightage diagram and readiness plan. Initially, positive example loads are straightaway brought into the generous learning system. But the negative examples, gathered by bootstrap strategy and the loads are balanced by the order mistakes of each past feeble classifier. Like the condition utilized in boosting preparing technique, the modifying should be possible by: wg ij+1



wg 0j

exp

 i w k 

 α − y jkt , h kt (x)

(11)

k=1 t=1

where yj is the sample label x j , wg0j is the starting weight for sample x j , and i is the index of the current node. We no longer require the initial weak learner hi,0 (x) and based on the earlier boosting classifier the successive training will be done. In Fig. 5, the algorithm description is given: In light of the current methodology, the booster chain can be viewed as a variation of AdaBoost calculation with comparative and speculative execution with an error limit.

3 Optimization of Boosting Chain The performance at the present step requires an exchange in the middle of precision and speed at every step of boosting chain. The more the attributes used, the higher the recognition precision attained. At that time, the evaluation will be more for the classifiers with more features. The stability between the selected recall and FP rates

818

A. S. Venkata Praneel et al.

1. The values for all positive sample xj is Initialized as: i=0, FO0=1,

= {},

wgj=1/p and for all -ve sample xj, wgj=1/ni 2. It has to be observed that ,While FOi >FO a) Then i=i+1 b) Training

i

to reach the requirements di and fi requirements on

validation data set. − by using primary weights wgj, training data set Ni and P − by training a classifier node

i

c) Optimization of Node classifier d) FOi=FOi-1* fi,

=

{

i}

e) Evaluation of Booster Chain

on non-face picture , as well input

false recognitions in the data set at Ni+1 f) For each example of xj in data set Ni+1, a modified weight wgj for

i+1

as per to Equation (11). g) Loads of missed +ve examples are valued to zero, as well as the loads

of stable +ve examples are unaltered. Fig. 5 Booster chain learning pseudocode

will be achieved by simply adjusting the threshold for each classifier by using the naive optimization method. Already mentioned sharp increases in FP rates in the results are frequent with this method. A new algorithm for boosting optimization is studied to address this issue, established on a straight model.

3.1 Booster Optimization Based on a Linear Model For reducing the complexity, the following norms are used: Q = wi , h j (x) = h i, j (x), α j = αi, j , b = bi , and α = {α1 , α2 . . . αT }. By then, an official end limit of AdaBoost defined Eq. (5) can be seen as the straight mix of weak learners {h 1 (x), h 2 (x) . . . h t (x)}.

A Survey on Accelerating the Classifier Training Using Various …

819

Every weak learner hi (x) will be marked after the booster training. Whenever it is fixed, the slight understudy corresponds to the model x i from the primary attribute space F to a different point in a new space F * with a newer dimensionality Q. xi∗ = hx(i) = {hx, hx1 (i), x2 (i), . . . , h x T (i)}

(12)

With this, the streamlining of α parameter is viewed as a finding in an ideal isolating hyperbola-plane in the latest space F * .

3.2 Adjusting the Classifier According to Eq. (13), the answer for getting enhanced hyperbola-plane could be achieved by settling the accompanying quadratic programming issue: L(γ ) =

n 

γ−

i=1

Subject to the requirements

n     1  γi γ j = yi − yi j h xi j i .h x j 2 i, j=1

n 

(13)

γi yi = 0 and Ui ≥ γi ≥ 0, i = 1 . . . n. Coefficient

i

U i is initialized by the constant U and classification hazard “r” over the data set: Ui = rU. If ‘ xi’ is a face pattern

(14)

where “C” is a characterization hazard, “r” is dictated by the location rate necessity; the rate of recognition is enhanced by expanding the estimation of value of “r” with an expense of false positive value. The arrangement of the augmentation issue is indicated by: γ 0 = γ10 , γ20 . . . , γn0

(15)

Then, the optimized α will be given by: α=

n  i=1

γi yi h i (xi )

(16)

820

A. S. Venkata Praneel et al.

1.

Train the direct SVM classifier with the set of {hi(x)}, i=1… Z and weight wgj.

2.

Then sort the vector α by worth. Assume the latest file will be like i1, i2, …, iZ etc.

3.

k=1… N, where, N is the constant value for highlight end. a) Delete the component hik, b) Calculate recent learning exactness pk c)

Include hik

4. Finally, delete the element hik, with the biggest pk

Fig. 6 n-level booster feature reduction algorithm

3.3 Removal of Redundancy in Boosting An AdaBoost is a successive forward pursuit system utilizing the eager choice methodology, repetition during the learning strategy cannot be maintained a strategic distance from. FloatBoost embraces the backtrack technique. It erases horrible frail classifiers from the group when another powerless classifier is included. Despite the fact that FloatBoost gives a trustful method to lessen the repetition throughout the booster preparation, and such a technique is a struggle with the booster weight pattern, the exhibit of learning methodology is precarious, which appears in Fig. 6. As indicated by the direct model of booster classifier, the outcome classifier can be communicated as: f (x) =

Q 

αi h(xi ) + b

(17)

i=1

The path of gradient f (x) on hi (x) is: ∇h(xi ) · f (x) = α

(18)

A Survey on Accelerating the Classifier Training Using Various …

821

300

False Positives

250

FP for Boostin g Chain

200 150

FP for FloatBo ost

100 50 0 0.91 0.89 0.87 0.86 0.85 0.84 0.83 0.82 0.81 0.76 Detection Rates

FP for AdaBoo st

Fig. 7 Different numbers of FPs for detection rates on the MIT + CMU test set

4 Performance Comparisons of the Three Detectors There are three identifiers depending on booster chain; FloatBoost course and Adaboost course are executed over a similar preparing set. The detection rate of FP bend and over the test of MIT-CMU is appeared in Fig. 7. What is more, the normal quantities of highlights utilized in every finder are recorded in Table 3. The False Positive Rate (FPR) is FP/N where FP is the number of false +ves. N is the total number of −ves. FPR = FP/N = FP/NFP + NTN

(19)

where N = NFP + NTN, NFP—Number of false positives and NTN—Number of true negatives. So as to evade any distinctions coming out of the fundamental foundation frameworks of identifier, with a preparation set of data of 16,000 pictures (i.e., 8000 countenances and another 8000 of non-faces) as well as a test data set of 16,000 pictures (i.e., 6000 appearances and another 10,000 non-faces) have been utilized to assess the calculations. In Fig. 8, we have shown FP rates under different features and fixing the detection rate to 96%.

5 Observations of Empirical Outcomes From Figs. 7, 8 and Tables 2, 3 and 4 some experiment results shown, the empirical results are seen in the show of the studied process in the following things: 1. In Table 3, the best performance is boosting chain cascade algorithm.

822

A. S. Venkata Praneel et al. 1 0.9 FP rates

0.8 0.7

FPR for BoostingChain

0.6 0.5

FPR for Adaboost

0.4 0.3

FPR for Floatboost

0.2 0.1 0

Features Used

Fig. 8 Number of Features for the FP rate on the MIT + CMU testing set

Table 2 False positives and the detection rates for the three detectors on MIT-CMU test set FP for boosting chain

Detection rate

FP for FloatBoost

Detection rate

FP for AdaBoost

Detection rate

252

0.96

200

0.91

225

0.91

214

0.95

180

0.90

190

0.89

186

0.91

160

0.89

160

0.87

172

0.90

120

0.88

96

0.86

81

0.89

105

0.87

86

0.85

69

0.86

74

0.86

75

0.84

51

0.85

61

0.85

64

0.83

42

0.82

52

0.84

56

0.82

30

0.80

38

0.82

48

0.81

21

0.78

22

0.79

25

0.76

Table 3 Average no. of features utilized in face detection on MIT-CMU test set

Boosting chain

FloatBoost

AdaBoost

21

22

25

2. AdaBoost and FloatBoost cascade approaches are outperformed by boosting chain approach in Fig. 7, from the Detection-FP rate curve. When there is a higher recall rate, then it works great. 3. The higher recall rate works in enhancing the effectiveness of the approach after post-filtering process.

214

186

172

164

158

147

135

102

81

69

51

42

30

21

47

58

150

200

250

300

350

400

450

500

550

600

650

700

750

800

900

1000

328

341

394

374

345

322

301

272

240

222

201

195

188

180

172

146

0.177

0.121

0.05

0.08

0.13

0.18

0.23

0.29

0.37

0.422

0.44

0.44

0.46

0.48

0.51

0.59

0.675

FPR

36

48

56

64

75

86

96

101

116

135

160

190

225

300

275

250

300

NFP

121

NTN

NFP

252

AdaBoost

Boosting chain

100

Features

36

59

99

100

110

114

116

121

130

145

150

150

160

185

101

76

75

NTN

0.31

0.34

0.36

0.32

0.40

0.44

0.45

0.45

0.47

0.48

0.51

0.55

0.58

0.61

0.73

0.766

0.80

FPR

16

20

32

33

34

35

38

45

52

61

74

96

105

120

160

180

200

NFP

FloatBoost

142

170

270

294

295

295

295

290

241

240

226

203

180

172

150

145

140

NTN

0.101

0.10

0.10

0.10

0.10

0.10

0.11

0.15

0.17

0.20

0.24

0.32

0.36

0.41

0.51

0.55

0.59

FPR

Table 4 Number of features used and the NFP—number of false +ve’s, NTN—No. of true −ve’s and FPR—false +ve rate on MIT-CMU test set are given in the table for the three detectors boosting chain, FloatBoost and AdaBoost

A Survey on Accelerating the Classifier Training Using Various … 823

824

A. S. Venkata Praneel et al.

4. Even though it looks like the curve of FloatBoost, cascade is nearer to the curve of boosting chain cascade. From the results shown in Fig. 8, the AdaBoost cascade and FloatBoost cascades are outperformed by boosting chain classifier. 5. From the primary attribute space when the feature gets removed, at 800 features the flawed alarm rate of boosting chain classifier hits its minimum at the point. 6. The boosting chain will gradually degrade if we process more features and this fact will be studied that the algorithm reaches the needed aspect of boosting linear model at 800. The empirical results from the two test data set convey that the quality and effectiveness of recommended framework in totality.

6 Conclusion In the current paper, another system for quick object recognition has been studied. In the present system, booster cascade and bootstrap preparation are joined into a solitary learning process, which is not just given a hypothetical worth establishment to course preparing, yet additionally improves classifier execution by fusing heritage information of course learning. In any case, in view of a straight examination model for existing booster classifier, a classifier changing excess—decrease calculation is additionally displayed. The empirical outcomes state most of the testing data sets have illustrated the strength, versatility, precision and prevalence of the presented system. Likewise, we accept the propelled system that has been studied in the paper can be connected to other characterization conflicts in success PC vision.

References 1. Poggio, T., Sung, K.K.: Example-based learning for view based human face detection. IEEE Trans. PAMI 20(1), 39–51 2. Rowley, H.A., Baluja, S., Kanade, T.: Neural network-based face detection. IEEE Trans. Pattern Anal. Mach. Intell. 20, 22–38 (1998) 3. Roth, D., Yang, M., Ahuja, N.: A snowbased face detection. In: Neural Information Processing, vol. 12 (2000) 4. Schneiderman, H., Kanade, T.: A statistical method for 3D object detection applied to faces and cars. In: Proceedings IEEE Computer Soc. Conference on Computer Vision and Pattern Recognition (2000) 5. Serre, T., et al.: Feature selection for face detection”. AI generalization of on-line learning and an application to Memo 1697, Massachusetts Institute of Technology, 2000 boosting. J. Comput. Syst. Sci. 55(1), 119–139 (1997) 6. Vapnik, V.N. (ed.): Statistical learning theory. Wiley, New York (1998) 7. Papageorgiou, C.P., Oren, M., Poggio, T.: A general framework for object detection. In: Proceedings of International Conference on Computer Vision (1998)

A Survey on Accelerating the Classifier Training Using Various …

825

8. Alvira, M., Rifkin, R.: An empirical comparison of SNoW and SVMs for face detection. CBCL Paper#193/AI Memo #2001–004, Massachusetts Institute of Technology, Cambridge, MA, Jan 2001 9. Viola, P., Jones, M.: Robust real time object detection. In: IEEE ICCV Workshop on Statistical and Computational Theories of Vision, Vancouver, Canada, July 13 2001 10. Freund, Y., Schapire, R.E.: A decision-theoretic generalization of on-line learning and an application to boosting. Computational Learning Theory: Eurocolt ‘95, pp. 23–37. Springer, New York (1995) 11. Freund, Y., Schapire, R.E.: Experiments with a new boosting algorithm. In: Proceedings of the 13th Conference on Machine Learning, Morgan Kaufmann, pp. 148–156 (1996) 12. Li, S.Z., Zhang, Z., Shum, H.-Y., Zhang, H.: FloatBoost learning for classification. In: Advances in Neural Information Processing Systems, pp. 1017–1024 (2003) 13. Li, S.Z. et al.: Statistical learning of multi-view face detection. In: Proceedings of the 7th European Conf. on Computer Vision. Copenhagen, Denmark. May 2002 14. Brubaker, S.C., Wu, J., Sun, J., Mullin, M.D., … Rehg, J.M.: On the design of cascades of boosted ensembles for face detection. Int. J. Comput. Vis. 77, 65–86 (2008) 15. Verschae, R., Ruiz-del-Solar, J., Correa, M.: A unified learning framework for object detection and classification using nested cascades of boosted classifiers. Mach. Vis. Appl. 19, 85–103 (2008) 16. Guyon, I., Weston, J., Barnhill, S., Vapnik, V.: Gene selection for cancer classification using support vector machines. BIOWulf Technical Report (2000) 17. Xiao, R., Zhu, L., Zhang, H.: Boosting chain learning for object detection. In ICCV 3, 709 (2003)

Service Layer Security Architecture for IOT Using Biometric Authentication and Cryptography Technique Santosh Kumar Sharma and Bonomali Khuntia

Abstract Data security and authentication mechanism is a very challenging job for smart devices. And more ever, IOT is suffering with login and verification process. Here, in our paper, we have focused on human characteristics-based security system which cannot be pinched easily such as iris, thumb, palm, DNA and voice-based authentication system. Using biometric authentication theory, we have presented that how biometric systems are the boundless computational resources and prospective of flexibility, reliability and cost reduction along with high-security performance resources. To maintain the security of biometric traits over the Internet channel, end user can apply the cryptography algorithm such as ElGamal, MAC Omura, Cramer– Shoup, RSA. As a final point, this paper is contributed for evidencing the strength of integrating the biometric authentication system with cryptography techniques and its application on Internet-based applications. In order to develop strong security, we have proposed an integrated approach of three mechanisms using biometrics, OTP and cryptography. The work is validated for biometrics through AVISPA (SPAN) security tool which is worldwide acceptable for approving the security architecture.

1 Introduction This paper has presented a dynamic verification process using finger scan-based authentication scheme, which provides runtime authentication among end users, and both side verification will be done through dynamic verification with additional key security exchange between data server and user as a final security verification process. Biometric system is having very high efficiency during the identity of any human S. K. Sharma Department of MCA, Vignan’s Institute of Information Technology, Visakhapatnam, Andhra Pradesh, India e-mail: [email protected] B. Khuntia (B) Department of Computer Science, Berhampur University, Berhampur, Odisha, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0_80

827

828

S. K. Sharma and B. Khuntia

being, and due to this reason, many government and private sector organizations are using the thumb biometric system to maintain proper attendance management system without any bias. The working principle of biometrics is to take the input of small portion of finger surface showing to the sensor for feature mining and comparison, thus leading to relatively high matching speed and accuracy with moderate cost. Even though biometric is a reliable authentication and identity system, it cannot be assured as secure and concerns to worry about. The major problem of biometrics is pirate, the biometrics key, which may happen only through stolen biometrics, replacing compromised biometrics, frauds done by administrators, denial of service and intrusion, etc. Biometrics is suffering with data leakage due to vulnerable insecure shipper. In this regard, we have integrated the biometric technique with OTP and powered by secret key exchange as a final phase for security verification. Here, the security verification work is divided into three steps, in which the first step is the user authentication using finger-based biometric authentication, and in the second step, we are using one-time password for registered MAC address verification, and in the third phase, verification using cryptography. In service layer security, service layer receives the data from element layer and secures data from insider attack which restricts unauthorized access along with protecting the data from malicious, unauthorized access attacks and denial of service attacks.

2 Related Works Author [1, 2] has investigated the different types of attacks and used BAN logic to solve the synchronization problem where security analysis has been done formally and informally to check the efficiency of the proposed protocol. Here, the author has used pseudo-random number generator with hash function for securing biometric data for which they have validated the entire work using AVISPA tool to verify the authentication architecture. The author discussed smart card and its application [3, 4] with smart security technology against the Ann’s scheme which is the scheme during mutual authentication; further, the new scheme is validated through SPAN tool to verify this scheme for evaluating the passive attacks followed by active outbreaks. [5, 6] In this contributed paper, the author has monitored many devices which are very sensitive and prone to risk, and in this regard, the author proposed the scenario for how to handle the future IOT application and connected devices. Here, the author has proposed a framework on contract basis which is dividing the entire framework into access control contract—ACC, judge contract—JC, register contract—RC to monitor IOT devices for different purposes. [7, 8] Here, the author has kept focus on the VoIP on application layer for managing and controlling the participants by session initiation protocol. SIP can be implemented in TCP/UDP networks. [9, 10] At presently, IOT technology is promising to human society for making smart world to convert each physical object in smart gadget that can make control of daily needs activity on fingertip by using Internet from remote location. Since the introduction of internet base technology work is become very easy to access, control and manage the

Service Layer Security Architecture for IOT Using Biometric …

829

smart appliances using smart applications. After all smart technology is very flexible for all types of smart applications but it is invited the security threats to the system resources which causes serious damage to them. Due to available multiple types of threats and security challenges it is motivating the researchers to build strong security architecture for IOT which helps to provide integrated security services. [11–13] projected the future of IOT by estimating the subjugated heavy content-oriented traffic and cohesive conversations. [14–16] has gone through the on-off attacks and observed the impact of Kalman filter technique for analyzing the different behaviors of attacks. Lionel, Abderrahim and Hamdi extended the work with new identical homogeneous encryption proposal and adoptable introduction analysis for IOT security services. [17–21] has discussed the threat assessment with the formal security framework design and RFID role in security management. Also discuss how security certification services is issuing the authenticated certificate to provide end to end security in smart network. Sameera and Yutaka [21–26] touch the security architecture for security services and discuss the issue for providing secure framework of cloud and how to maintain defense system with UES algorithm. Heydam [27–31] contributed their work verification and identity technique using reverse engineering. Jadhav has handled variety of OS threats related to application layer for things and kept focus on other standards which are creating threat to the smart devices, professional and social projects such as authentication identity, data storage and recovery management for runtime data along with other vulnerable flaws. [31–35] has put focus on physical layer security to stop unauthorized access of devices using jamming technique and with the help of IOT agent it reduces the illegal machine login. [36–38] proposed the channel based mechanism to provide the corporal security system against eavesdropping and replay attack. [39, 40] has discussed the physical layer security by analyzing noise ratio.

2.1 System Architecture 2.1.1

Biometric Authentication Feature

Managing identity and accessibility control is a very important issue in several applications which should be handled very carefully; regarding this, physical and virtual access control for every end user and participated device must be maintained regularly. This type of protocol checks two things—first is for the object “who is the user and what is his/her role,” and second is that particular user is eligible to proceed. In order to construct resilient security framework, we have proposed an integrated approach of three mechanisms using biometrics, OTP and cryptography Fig. 1.

830

S. K. Sharma and B. Khuntia

Fig. 1 Three-phase architecture for security management

2.1.2

One-Time Password

OTP is a hardware authentication password that is applied for single session transaction on different devices, and OTP is generally accompanied by the additional layer of authentication. End user makes request to authorize him using the user name and password before proceeding verification with the OTP. The main purpose of the OTP is the physical verification process of the device which is registered to the OTP server with the unique ID, and objective of the OTP token is the user’s need to physically carry it, so it is more secure for remote user and other stakeholder’s authentication (Fig. 2).

2.1.3

Cryptography

2.2 Algorithm (Honey Encryption) This encryption technique is used for shielding the password of customer who participates in communication against security breaches. The objective of the honey encryption algorithm is to generate a set of duplicate passwords to confuse illegal user and prevent the system from password attack. The set of passwords will give

Service Layer Security Architecture for IOT Using Biometric …

831

Fig. 2 Interaction diagram for the objects

small session login to user. Honey encryption is a strong algorithm to prevent the data from passwords attacks and stop breaches (Table 1). Table 1 Honey encryption and decryption algorithm description

Algorithm keywords description

Step-by-step process for encryption

Decryption process

2. K—Key

HF ← Enc(K,M)

HF ← Dec(K, (RN,CT))

3. M—Message

Se ← $encode(M)

Se” ← HF(RN,K)

4. RN—Random number

RN ← ${0,1}n

Se ← CT ⊕ Se”

5. S—Seed value

Se ← HF(RN,K)

M ← decode(Se)

6. CT—Cipher text

CT ← Se’ ⊕ Se

Return (M)

1. HF—Hash function

832

S. K. Sharma and B. Khuntia

3 AVISPA Code H:hash_func, SEND,RECV:channel(dy)) Played_by Ui def= local State=nat, F:hash_func const ui_as_xu,B,Au,PIDu,Pwu: text, USK,PSK,P,SIDs.Xu,RU, Ruu:text, AIDu,M1,M4,Xs,N1:text, s1,s2,s3:protocol_id init State:=0 transition %user registration phase 1.State=0/\RECV(start)=|> Sate’:=1/\B’:=new()/\Au’:=H(PWu.B’) /\PIDu’:=H(IDu,B’) /\secret({IDu,B’PWu},s1,{Ui}) /\SEND({PIDu’.Au’}_SKuirs} 2.State=1/\RECV({H(H(IDu.B’).H(PWu,B’).USK).SIDs.xor(H(PWu.B’),H(H(IDu.B’).H(SIDs.P

SK))).H(SIDs,PSK))).H((SIDs,PSK).P.H)_SKuirs) =|> State’:=2/\secret({USK,PSK},s2,RS) /\Xu’:=new()/\Ru’;=F(Xu’.P) /\Ruu’:F(Xu’,F(Xs’.P)) /\AIDu’:=xor(H(IDu.B’),Ruu’) /\M1’:=H(H(IDu.B’).H(H(IDu.B’).H(SIDs.PSK)).Ru’.Ruu’) /\secret(Xs’,s3,{RS,AS}) /\SEND(AIDu’,M1’.Ru’) 3.State=2/\RECV(xor(H(IDu.B’),N1’).H(H(F((Xu’’,P)).H(H(IDu,B’).H(SIDs.PSK)).SID.N1’).H(IDu.B’) .N1’.H(H(IDu.B’).H(SIDs,PSK)).F(Xu’.P)))=|> State’:=3/\M4:=H(H(F(Xu’.F(Xs’.P)).H(IDu.B’).H(SIDs,N1’),N1)/\SEND(M4’)/\request(AS,Ui,as_ui_n1 ,N1’)

end role role applicationserver(Ui,RS,AS:agent, H:hash func SEND,RECV:channel(dy)) played_by AS def= local State:nat, B,Bu,SIDs,IDu,PWu,USK,PSK,P,P,Cus:text, Dus,Xu,Xs,Rs,SK,M2,N1,M3:text, F:hash_fun const as_rs_xs,ui_as_xu,as_ui_n1, s1,s2,s3:protocol_id init Sate:=0 transition %application registration phase 1.Sate=0/\RECV(start)+|> State’:=3

Service Layer Security Architecture for IOT Using Biometric … /\Xs’:new() /\Rs’:F(Xs.P) /\SEND(SIDs,Rs’)\ /\witness(AS,RS,as_rs_xs,Xs’) 2.State=3/\RECV(xor(H(IDu,B’);F(Xu’.F(Xs’.P))).H(H(IDu.B’).H(H(IDu.B’).H(SIDs.PSK)).F(Xu’F(Xs’ .P))).F(Xu’.P))=|>

State’:=6/\secret({IDu,B’PWu},s1.{Ui}) secret({USK,PSK},s2,RS) secret(Xs’s3,{RS,AS}) /\N1:=new() /\Sk’:H(F(Xu’.F(Xs’P)).H(H(IDu.B’).H(SIDs.PSK)),SIDs.N1’) /\M2’:xor(H(IDu.B’).N1’) /\M3’:H(SK’.H(IDu.B’).N1’.H(H(IDu.B’).H(SIDs.PSK)).F(Xu’,P)) /\witness(AS,Ui,as_ui_n1,N1) 3.State=6/\RECV(H(H(F(Xu’.F(Xu’.F(Xs’.P)).H(H(IDu.B’).H(SIDs,PSK)).SIDs.N!’))=|> State’:=8/\request(Ui,AS,ui_as_xu,Xu’) end

role role registrationserver(Ui,RS,AS:agent, SKuirs:symmetric_key, H:hash_func, SEND,RECV:channel(dy)) played_by RS def= local State=:nat, B,Bu,SIDs,IDu;PWu,USK,PSK,P,Cus,Dus,Xs:text, F:hash_func

const as_rs_xs,s1,s2,s3:protocol_id init State:=0 transition %user registration phase %Receive request message from Ui securely 1. State=0/\RECV({H(IDu.B’).H(PWu.B’)}_SKuirs=|> State’:1/\secret({USK,PSK},s2,RS) /\Bu’=H(H(IDu.B’).H(PWu.B’).USK)\ /\Cus’:=H(H(IDu.B’).H(SIDs.PSK)) /\Dus’:=xor(H(PWu.B’),Cus’) /\SEND({Bu’.SIDs.Dus’.H(SIDs.PSK).P.H}_SKuirs) 1.State=0/\RECV(SIDs.F(Xs’.P))=|> Stat’:4/\secret(Xs’,s3,{Rs,AS}) /\request(AS,RS,as_rs_xs,Xs’) end role role session(Ui,RS,AS:agent, SKuirs:symmetric_key, H:hash_func) def= local TX1,TX2,TX3,RX1,RX2,RX3:channel(dy)

833

834

S. K. Sharma and B. Khuntia composition user(Ui,RS,AS,SKuirs,H,TX1,RX1) /\registrationserver(Ui,RS,AS,SKuirs.H,TX2,RX2) /\applicationserver(Ui,RS,AS,H,TX3,RX3) end role role environment() def= const ui,rs,as:agent, skuirs:symmetric_key, h,f:hash_func, sids,p,ru,aidu,m1,m2,m3,m4:text, as_rs_xs,ui_as_xu,as_ui_n1, s1,s2,s3:protocol_id intruder_knowledge={ui,rs,as,h,f,sids,p,ru,aidu,m1,m2,m3,m4} composition session(ui,rs,as,skuirs,h) /\session(I,rs,as,skuirs,h) /\session(ui,i,as,skuirs,h) /\session(ui,,rs,i,skuirs,h) end role goal secrecy_of s1 secrecy_of s2 secrecy_of s3 authentication_on as_rs_xs authentication_on ui_as_xu authentication_on as_ui_n1 end goal environment()

4 Result Analysis See Results 1 and 2.

Result 1 Protocol Simulation for identifying the different types of attack

Service Layer Security Architecture for IOT Using Biometric …

835

Result 2 Intruder attacks analysis attempts at multiple sources

5 Conclusion and Future Work In this paper, we have discussed strong authentication and cryptography mechanism for handling futuristic security challenges in IOT domain using integrated framework of different techniques. We have developed resilient security using an integrated approach of three mechanisms using biometrics, OTP and cryptography. From the result, it is concluded that biometrics is also under the threat of different types of attacks, which has motivated the researcher to enhance the exiting mechanism of biometrics for better security. So our research work enhanced the exiting biometrics mechanism with the one-time password for authenticating the individual physical devices associated with cryptography to provide the security for our confidential data. At last, our work is providing the security at three different phases to provide strong security system. The biometric strength is validated through AVISPA (SPAN) security tool which is worldwide acceptable for approving the security architecture. Our future work is to develop multilayer security framework protocol to handle diverse security challenges for IOT environment.

References 1. Chikouche, N., Cherif, F., Benmohammed, M.: An authentication protocol based on combined RFID-biometric system. IJACS, 62–67 2. Das, A.K., Goswami, A.: A robust anonymous biometric-based remote user authentication scheme using smart cards, pp. 3–19. Elsevier, Amsterdam (2014) 3. Reddy, A.G., Das, A.K., Odelu, V. and Yoo, K.Y.: An enhanced biometric based authentication with key-agreement protocol for multi-server architecture based on elliptical curve cryptography. PLOS ONE, 3–10 (2016)

836

S. K. Sharma and B. Khuntia

4. Chaudhry, S.A., Mahmood, K., Naqvi, H.: An improved and secure biometric authentication scheme for telecare medicine information systems based on elliptic curve cryptography (2015) 5. Tyagi, N., Wang, J., Wen, K., Zuo, D.: Honey encryption applications implementation of an encryption scheme resilient to brute-force attacks. In: SPRING, pp. 13–05 (2015) 6. Murthy, S., Mulchandani, M.: Improving security of honey encryption in database: implementation. In: ICSESD (2017) 7. Yin, W., Indulska, J., Zhou, H.: Protecting private data by honey encryption. In: HINDAWI, pp. 21–11 (2017) 8. Jaeger, J., Ristenpart, T.: Honey encryption beyond message recovery security. In: EUROCRYPT, pp. 23–02 (2016) 9. Gracy, P.L., Venkatesan, D.: An honey encryption based efficient security mechanism for wireless sensor networks. IJPAM (2018) 10. Sahu, R., Ansari, M.S.: A secure framework for messaging on android devices with honey encryption. IJECS, 09 (2017) 11. Challa, S., Wazid, M., Das, A.K., Kumar, N.: Secure signature-based authenticated key establishment scheme for future IoT applications. In: IEEE, pp. 1–16 (2016) 12. Sain, M., Kang, Y.J., Lee, H.J.: Survey on security in internet of things: state of the art and challenges. In: ICACT, pp. 1–6 (2017) 13. Huh, S., Cho, S., Kim, S.: Managing IoT devices using blockchain platform. In: ICACT, pp. 1–4 (2017) 14. Zahra, B.F., Abdelhamid, B.: Risk analysis in Internet of things using EBIOS. In: IEEE, pp. 1–7 (2017) 15. Majeed, A.: Internet of things (IoT): a verification framework. In: IEEE, pp. 1–3 (2017) 16. Abels, T., Khanna, R., Midkiff, K.: Future proof IoT: composable semantics, security, QoS and reliability. In: IEEE, pp. 1–4 (2017) 17. El-Maliki, T., Seigne, J.M.: Efficient security adaptation framework for internet of things. In: IEEE, pp. 1–6 (2016) 18. Mohsin, M., Anwar, Z.: IoTSAT: a formal framework for security analysis of the internet of things (IoT). In: IEEE, pp. 1–9 (2016) 19. Khan, W.Z., Zangoti, H.M., Aalsalem, M.Y.: Mobile RFID in internet of things: security attacks, privacy risks, and countermeasures. In: IEEE, pp. 1–6 (2016) 20. Baldini, G., Le Gall, F.: Security certification and labelling in internet of things. In: ICT, pp. 1–6 (2015); Nawir, M., Amir, A., Yaakob, N.: Internet of things (IoT): taxonomy of security attacks. In: ICED, pp. 1–6 (2016). (Thailand) 21. Sklavos, N., Zaharakis, I.D.: Cryptography and security in internet of things (IoTs): models, schemes, and implementations. In: IEEE, pp. 1–2 (2016) 22. Abderrahim, O.B., Elhdhili, M.H.: TMCoI-SIOT: a trust management system based on communities of interest for the social internet of things. In: IEEE, pp. 1–6 (2017) 23. Metongnon, L., Ezin, E.C., Sadre, R.: Efficient probing of heterogeneous IoT networks. In: IFIP, pp. 1–7 (2017) 24. Abderrahim, O.B., Elhedhili, M.H.: CTMS-SIOT: A context-based trust management system for the social internet of things. In: IEEE, pp. 1–6 (2017) 25. Zouari, J., Hamdi, M., Kim, T.H.: A privacy-preserving homomorphic encryption scheme for the internet of things. In: IEEE, pp. 1–6 (2017) 26. Midi, D., Rullo, A., Mudgerikar, A., Bertino, E.: Kalis—A system for Knowledge-driven adaptable intrusion detection for the internet of things. In: ICDCS, pp. 1–11 (2017) 27. Dorsemaine, B., Gaulier, J.P., Wary, J.P., Kheir, N.: A new threat assessment method for integrating an IoT infrastructure in an information system. In: ICDCSW, pp. 1–8 (2017) 28. Sicari, S., Grieco, L.A., Coen-Porisini, A.: A secure ICN-IoT architecture. In: IEEE, pp. 1– 6 (2017); Massonet, P., Deru, L., Achour, A., Dupont, S., Levin, A.: End-to-end security architecture for federated cloud and IoT Networks. IEEE, pp.1–6 (2017) 29. Batool, S., Saqib, N.A., Khan, M.A.: Internet of things data analytics for user authentication and activity recognition. In: FMEC, pp. 1–5 (2017); Jerald, A.V., Rabara, S.A.: Algorithmic approach to security architecture for integrated IoT smart services environment. In: WCCCT, pp. 1–6 (2016)

Service Layer Security Architecture for IOT Using Biometric …

837

30. Stergiou, C., Psanni, K.E.: Architecture for security monitoring in IOT environments. In: IEEE, pp. 1–4 (2017) 31. Nakagawa, I., Shimojo, S.: IoT agent platform mechanism with transparent cloud computing framework for improving IoT security. In: COMPSAC, pp. 1–6 (2017) 32. Praveena, A.: Achaqieving data security in wireless sensor networks using ultra encryption standard version—IV algorithm. In: ICIGEHT, pp. 1–5 (2017) 33. Tellez, M., El-Tawab, S., Heydari, M.H.: IoT security attacks using reverse engineering methods on WSN applications. In: IEEE, pp. 1–6 (2017) 34. Ahmed, I., Beheshti, B., Khan, Z.A., Ahmad, I.: Security in the internet of things (IoT). In: ITT, pp. 1–7 (2017). (Dubai) 35. Sowmya, S.N, Kulkarni, N.: Security threats in the application layer in IOT applications. In: I-SMAC, pp. 1–4 (2017) 36. Hu, L., Wang, X.: Cooperative jamming for physical layer security enhancement in internet of things. In: IEEE, pp. 1–10 (2018) 37. Pechetti, S.V., Jindal, A., Bose, R.: Channel-based mapping diversity for enhancing the physical layer security in the internet of things. In: IEEE, pp. 1–6 (2017) 38. Burg, A., Chattopadhyay, A., Lam, K.Y.: Wireless communication and security issues for cyber–physical systems and the internet-of-things. In: IEEE, pp. 1–23 (2018) 39. Atat, R., Ashdown, J., Yi, Y.: A physical layer security scheme for mobile health cyber-physical systems. In: IEEE, pp. 1–15 (2017) 40. Pechetti, S.V., Jindal, A., Bose, R.: Channel based mapping diversity for enhancing the physical layer security in the internet of things. In: IEEE, pp. 1–15 (2017)

Author Index

A Abhinav, 11, 77 Adeshara, J. V., 199 Ahmad, Inzamam, 261 Alex Luke, K. A., 43, 473 Allaka, Gopichand, 157 Al Madanat, Iyad, 53 Amaresh Kumar, D., 371 Ananthu, J, 731 Angadi, Gangadhar, 533 Anitha, J., 655 Arun, J., 43 Ashok Babu, T. P., 361, 523 Ashok Kumar, M., 85

Chitra, K, 663 Choppavarapu, Vamsi Krishna, 759

B Baini, Rubiyah, 575 Baisakh, Barada P., 593 Balaji Ganesh, N., 509 Balaji, R., 309 Balaji Reddy, V., 473 Balakrishnan, P., 685 Balamurugan, P., 329 Balthazar, Pravinth, 123 Bharath, L., 489 Bhargavi, Y., 633 Bharj, Rabinder Singh, 239, 565 Biswas, Sanjoy, 93 Bucheli, José, 173 Bustamante, Carlos, 173

E Escobar, Ivón, 173, 181 Escudero, Miguel, 173

C Chandra, A. C. Prapul, 533 Chandra, Ishwar, 607

D Dang, Pranvat Singh, 749 Das, Raja, 655 Dave, Dhara H., 109 Deheri, G. M., 199 Delvadiya, Hardikkumar, 499 Devaraju, A., 483 Diwakar Reddy, V., 633 Dutta, Avisankha, 553 Dutta, Bibhash Kumar, 703

F Fong, Lim Soh, 575

G Gaddam, Bhanodaya Reddy, 453 Gangadhar, T. G., 533 Ghosh, Angkush Kumar, 33 Girisala, Vijay Kumar, 583 Girish, D. P., 533 Gnanadhas, Diju Samuel, 191 Gonjari, Shreyas, 781 Gopalakrishnan, E. A., 731 Gopalakrishnan, Ramanan, 191 Gopala Krishna, S. V., 289

© Springer Nature Singapore Pte Ltd. 2020 A. N. R. Reddy et al. (eds.), Intelligent Manufacturing and Energy Sustainability, Smart Innovation, Systems and Technologies 169, https://doi.org/10.1007/978-981-15-1616-0

839

840 Gugale, Akshay, 781 Gundala, Harika, 781 Gupta, Vikash Kumar, 261

H Haran, Rudraksh Raajesh, 749 Hari Haran, V., 271 Hari Sankar, P., 85 Harish, K. R., 319 Hatti, Prashant S., 421 Hazmi, Helmy, 739 Hemanth Prasanna, R., 43 Hussain, Mohd Hasnain Md, 575 Hyma, J., 759

Author Index Kumar, Sandeep, 21 Kumar, S. Sathees, 645 Kumar, Surender, 239, 565 Kumar, VaitlaSai, 21 Kumar, Vikram, 93 Kunapareddy, Aditya, 157 Kustagi, Harish Kumar, 77

L Lakshminarayana, G., 353 Limbai, Michelle Maya, 739

J Jadeja, Rajendrasinh, 147, 279 Jagannati, Venumurali, 453 Jaikumar, Mayakrishnan, 339 Jain, Rakesh, 711 Jamali, Annisa, 739 Jaya Kishore, Sunkara, 583 Jeshrun Shalem, M., 483 Jethva, Prapti, 429 Jimenez, Icler, 181 Joshi, Shivam, 401 Jyothu Naik, Ramavathu, 623

M Magikar, Atul, 221 Mahapatra, Debasish, 523 Mahesha, C. R., 379 Mallikarjuna, K., 85 Mamat, Hussin Bin, 123 Mangeelal, D., 583 Maslehuddin, Mohammed, 299 Mazumdar, Anuraag, 799 Megat Johari, Megat Azmi, 299 Menon, Vijay Krishna, 731 Mishra, Shubham, 703 Mistry, Pavak, 401 Mohamaddan, Shahrol, 575, 739 Mugesh Raja, V., 645 Muli, Madhu Kiran Reddy, 191 Mullo, Santiago, 173 Muthukumaraswamy, Senthil Arumugam, 693

K Kalaria, Tarang, 147 Kamal, Maswida Mustafa, 575 Kamatchi Kannan, V., 721 Kapoor, Rohan, 231 Karthik, K., 483 Khalid, Syed Sha, 461 Khan, Abid Hossain, 1, 33 Khanna, Pradeep, 231 Khuntia, Bonomali, 827 Kothari, Nishant, 543 Krishnakumari, A., 133 Krishna, Sandeep, 411 Kulkarni, Pallavi, 221 Kumar, Dinesh, 429, 499, 543 Kumar, Kintali Pavan, 791 Kumar, Pankaj, 93 Kumar, Punith, 11

N Nadgauda, Nikhita, 693 Naga Jyothi, P., 769 Nagananthini, R., 329 Nagarajan, Kailash, 731 Nagavinothini, R., 329 Nagmohan Rao, N., 309 Najamuddin, Syed Khaja, 299 Najeeb, Nithin Sha, 53 Nandakumar, P., 441 Narasimha Murthy, K., 421 Nasir, Mohammad Nazmi, 123 Naveen, Anantapalli, 791 Naveen, Arnipalli, 67 Neelima, N., 675 Negi, Prashant, 261 Nilesh, Mutyam, 791 Nilugal, Raghavendra P., 319

I Islam, Md. Shafiqul, 1 Ismail, Mohd Azmi, 123

Author Index P Panda, Sasank Shekhar, 703 Parvez, Mahmud, 249 Paswan, M. K., 93 Patel, Chandresh, 147 Patel, Krishna, 429, 499 Patel, Rajesh, 401 Patel, R. M., 199 Patel, Vinod, 147, 401 Patil, Neelkant, 371 Pendse, Tejas, 221 Ponnuri, Sai Manikanta, 133 Prabha, S. U., 693 Prajapati, M. B., 199 Prajwal, D., 11 Prasad, Anil K., 261, 593 Prasad, Naresh, 93 Pravin Kumar, M., 685 Preethi, G., 721 Pruna, Edwin, 173, 181

R Rahman, Aliff, 739 Rahman, A. N. M. Mizanur, 33 Rajagopal, M. S., 461 Rajeswara Rao, K. V. S., 411 Raj, K. V. Karthik, 533 Raju, J. V. V. S. N., 675 Rajya Lakshmi, D., 769 Rakesh, P., 675 RamaKrishna Murty, M., 809 Ramakrishnan, M., 133 Ramanathan, Velmurugan, 339 Rama Rao, K. V. S. N., 769 Ramdan, Muhammad Iftishah, 123 Ramesha, D. K., 489 Ramesh, Ajith, 731 Ramesh, N., 607 Rangajanardhana, G., 441 Rathinam, Kalidasan, 21 Rathore, Ajay Pal Singh, 711 Razzaque, Muhammed Mahbubur, 1 Regunathan, Rajeshkannan, 781, 799 Relangi, Sree Pradeep Kumar, 791

S Sachdeva, Aditya, 231 Sandeep Kumar, Y., 411 Sangeethkumar, E., 339 Sanjeev Kumar, P. V., 85 Sarada Devi, M., 109

841 Saravanan, M., 133 Sasikumar, N., 339 Sathishkumar, S., 721 Seela, Chiranjeeva Rao, 67 Selokar, Ashish, 43, 309, 473 Selvakumar, Raja, 339 Senapati, Ajit Kumar, 703 Sethia, Mayank, 781 Sethia, Mehul, 781 Shah, Vivek J., 209 Shaik, Sharmas Vali, 361 Shakil, Shawkat Imam, 249 Shankar, Arun R., 77 Sharma, Santosh Kumar, 827 Sharma, Vivek, 21 Shashikumar, C. B., 319 Shivarudraiah, 371, 379 Shridhar, V. A., 339 Simlandi, Sudip, 553 Singh, Awwal, 231 Sivaramakrishnaiah, M., 441 Solomon, Gnanadurai Ravikumar, 309 Somanakatti, Anupama B., 421 Soman, K. P., 731 Soori, Prashant Kumar, 53 Sridhar Babu, B., 645 Srihari, P. V., 509 Srikanth, N., 309 Srinadh, Reddy, 133, 339 Srinivasa Rao, D., 289 Srinivasa Rao, T., 809 Sudamalla, Moulica, 759 Suneel, K., 309 Suresh Babu, P., 289

T Thirupathi Reddy, Kota, 623 Tirupathi, K., 645 Trivedi, Janak D., 109 Trivedi, Tapankumar, 147, 279, 401

U Udayakumar, R., 209 Uday Chandra Rao, P., 289 Ullegaddi, Kalmeshwar, 371, 379 Upadhyay, Anima, 607

V Vaghela, Shivani, 543 Vamsi, Anupoju Sai, 67 Vanaparthi, Dharma Teja, 759

842 Vasantha Gowri, N., 391 Ved, Amit, 279 Velmurugan, R., 685 Venkata Praneel, A. S., 809 Venkataramaiah, P., 271 Venkatesh, Sai Vijay, 209 Venkateswara Rao, K., 289 Venugopal, S. M., 411 Vijaya, G., 461 Vinnakota, Koushik, 759 Viswanatha Rao, J., 353

Author Index Vyas, Chitrang, 279

W Wagire, Aniruddha Anil, 711 Wahab, Andyqa Abdul, 123

Y Yalamalle, Sunil R., 411 Yusuf, Moruf Olalekan, 299