309 119 9MB
English Pages [98] Year 2020
https://www.tarjomano.com/order
DESIGN AND PERFORMANCE OF
TALL BUILDINGS FOR WIND ASCE MANUALS AND REPORTS ON ENGINEERING PRACTICE NO. 143 EDITED BY PREETAM BISWAS JOHN PERONTO
TASK COMMITTEE FOR THE DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
ASCE Manuals and Reports on Engineering Practice No. 143
Design and Performance of Tall Buildings for Wind Edited by Preetam Biswas and John Peronto Prepared by Task Committee for the Design and Performance of Tall Buildings for Wind of the Structural Engineering Institute of the American Society of Civil Engineers
Published by the American Society of Civil Engineers
Library of Congress Cataloging-in-Publication Data Names: Biswas, Preetam, editor. | Peronto, John, editor. | Task Committee for the Design and Performance of Tall Buildings in Wind author. Title: Design and performance of tall buildings for wind / prepared by Task Committee for the Design and Performance of Tall Buildings in Wind; edited by Preetam Biswas, P.E., and John (Sp.) - Peronto. Description: Reston, Virginia : American Society of Civil Engineers, [2020] | Series: ASCE manuals and reports on engineering practice ; no. 143 | Includes bibliographical references and index. | Summary: “Design and Performance of Tall Buildings for Wind, MOP 143 provides a framework for the design of tall buildings for wind, based on the current stateof-practice in tall building structural design and wind tunnel testing”– Provided by publisher. Identifiers: LCCN 2020033269 | ISBN 9780784415658 (paperback) | ISBN 9780784483121 (adobe pdf) Subjects: LCSH: Tall buildings–Aerodynamics. | Wind resistant design. Classification: LCC TH891 .D37 2020 | DDC 690/.21–dc23 LC record available at https://lccn.loc.gov/2020033269 Published by American Society of Civil Engineers 1801 Alexander Bell Drive Reston, Virginia 20191-4382 www.asce.org/bookstore | ascelibrary.org Any statements expressed in these materials are those of the individual authors and do not necessarily represent the views of ASCE, which takes no responsibility for any statement made herein. No reference made in this publication to any specific method, product, process, or service constitutes or implies an endorsement, recommendation, or warranty thereof by ASCE. The materials are for general information only and do not represent a standard of ASCE, nor are they intended as a reference in purchase specifications, contracts, regulations, statutes, or any other legal document. ASCE makes no representation or warranty of any kind, whether express or implied, concerning the accuracy, completeness, suitability, or utility of any information, apparatus, product, or process discussed in this publication, and assumes no liability therefor. The information contained in these materials should not be used without first securing competent advice with respect to its suitability for any general or specific application. Anyone utilizing such information assumes all liability arising from such use, including but not limited to infringement of any patent or patents. ASCE and American Society of Civil Engineers—Registered in US Patent and Trademark Office. Photocopies and permissions. Permission to photocopy or reproduce material from ASCE publications can be requested by sending an email to [email protected] or by locating a title in the ASCE Library (https://ascelibrary.org) and using the “Permissions” link. Errata: Errata, if any, can be found at http://dx.doi.org/10.1061/9780784415658. Copyright © 2020 by the American Society of Civil Engineers. All Rights Reserved. ISBN 978-0-7844-1565-8 (soft cover) ISBN 978-0-7844-8312-1 (PDF) Manufactured in the United States of America. 25 24 23 22 21 20 1 2 3 4 5
MANUALS AND REPORTS ON ENGINEERING PRACTICE (As developed by the ASCE Technical Procedures Committee, July 1930, and revised March 1935, February 1962, and April 1982)
A manual or report in this series consists of an orderly presentation of facts on a particular subject, supplemented by an analysis of limitations and applications of these facts. It contains information useful to the average engineer in his or her everyday work, rather than findings that may be useful only occasionally or rarely. It is not in any sense a “standard,” however, nor is it so elementary or so conclusive as to provide a “rule of thumb” for nonengineers. Furthermore, material in this series, in distinction from a paper (which expresses only one person’s observations or opinions), is the work of a committee or group selected to assemble and express information on a specific topic. As often as practicable the committee is under the direction of one or more of the Technical Divisions and Councils, and the product evolved has been subjected to review by the Executive Committee of the Division or Council. As a step in the process of this review, proposed manuscripts are often brought before the members of the Technical Divisions and Councils for comment, which may serve as the basis for improvement. When published, each manual shows the names of the committees by which it was compiled and indicates clearly the several processes through which it has passed in review, so that its merit may be definitely understood. In February 1962 (and revised in April 1982), the Board of Direction voted to establish a series titled “Manuals and Reports on Engineering Practice,” to include the manuals published and authorized to date, future Manuals of Professional Practice, and Reports on Engineering Practice. All such manual or report material of the Society would have been refereed in a manner approved by the Board Committee on Publications and would be bound, with applicable discussion, in books similar to past manuals. Numbering would be consecutive and would be a continuation of present manual numbers. In some cases of joint committee reports, bypassing of journal publications may be authorized. A list of available Manuals of Practice can be found at https://www.asce.org/ bookstore.
AUTHORS
Preetam Biswas, Skidmore, Owings & Merrill Preetam Biswas is a director of structural engineering at Skidmore, Owings & Merrill (SOM). He has led the design of multiple tall buildings, airports, stadiums, and convention centers around the globe. He has authored many technical papers focusing on tall buildings and other structural system innovations and served as the chair for the task committee that authored this Manual of Practice. John Peronto, Thornton Tomasetti, Inc. John Peronto is a senior principal at Thornton Tomasetti and has led the design of many tall-to-megatall buildings, as well as unique and iconic structures around the world. He is highly published and a leader in professional organizations such as ASCE, ACI, IStructE, ICE, CTBUH, CCHRB, and has served as the chair for the Tall Buildings Committee of SEI since 2016. Kevin Aswegan, Magnusson Klemencic Associates Kevin Aswegan is an associate with Magnusson Klemencic Associates (MKA) in Seattle, where his design experience focuses on tall buildings in areas of high wind and seismicity in the United States and around the world. He is actively involved in wind-related research, has authored many articles on performance-based wind design, and is an active member of the Standard ASCE 7 update process. Roy Denoon, CPP Wind Engineering Dr. Roy Denoon is vice president of CPP Wind Engineering and has been involved in wind tunnel testing for 30 years, during which time he has worked on many globally iconic buildings and structures. He is a member of the ASCE 7 Wind Load Subcommittee, a contributing author to the ASCE/SEI Prestandard for Performance-Based Wind Design, a co-author of v
vi
AUTHORS
the CTBUH Technical Guide on Wind Tunnel Testing of High-Rise Buildings and is a CTBUH Fellow. John Kilpatrick, RWDI John Kilpatrick is a principal at RWDI. John is currently Wind Engineering Practice Leader at RWDI, a former chair of the UK Wind Engineering Society, and is a contributing author to the ASCE/SEI Prestandard for Performance-Based Wind Design and ASCE 49 Standard for Wind Tunnel Testing for Buildings and Other Structures. Sami Matar, LERA Consulting Structural Engineers Sami S. Matar is an associate partner with LERA Consulting Structural Engineers (LERA). Since joining LERA in 1995, he has built extensive experience in a variety of projects and has participated in or led the design of several of the firm’s high-rise building projects. Brian McElhatten, Arup Brian McElhatten is an associate principal in Arup’s Chicago office and leads the structural engineering group there. He has extensive knowledge of tall building systems, analysis, and design. During his time at both Arup and SOM, he has worked on and led the structural design of numerous tall and supertall buildings around the world. In addition to the Tall Buildings Committee of SEI, he is actively involved in CTBUH. Patrick Ragan, WSP Patrick Ragan is an associate with WSP in Chicago. His design experience includes towers in the United States, China, and the Middle East, with a primary focus on the lateral system design and optimization of high-rise structures. He has authored papers on various subjects including statistical methods for estimating wind turbine design loads, form-finding analysis of shell-shaped roof structures, and fluid viscous damper alternatives to conventional steel brace outrigger systems. CONTRIBUTOR Alexander W. Jordan, Skidmore, Owings & Merrill Alex Jordan is an Associate at Skidmore, Owings & Merrill (SOM). He has experience in the design of tall buildings and airports, and he specializes in digital design practices.
CONTENTS
PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi ACKNOWLEDGMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.3 Use of This Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Historic General Design Requirements . . . . . . . . . . . . . . . . . . . . . 3 1.5 Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.6 Nature of Wind . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.7 Limitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2
DESIGN PROCESS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Establish Performance Objectives . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.3 Preliminary Structural Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.4 Wind Climate Assessment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.5 Wind-Induced Loads and Responses . . . . . . . . . . . . . . . . . . . . . . 9 2.6 Structural Modeling and Analysis . . . . . . . . . . . . . . . . . . . . . . . 10 2.7 Comparison of Results to Acceptance Criteria . . . . . . . . . . . . . . 10 2.8 Wind Optimization Program . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.9 Final Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.2 Mean Recurrence Intervals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.2.1 Strength: Foundation and Lateral System (Main Wind Force Resisting System) . . . . . . . . . . . . . . . . . 14 3.2.2 Serviceability: Drift and Displacement . . . . . . . . . . . . . . . . 14 3.2.3 Serviceability: Accelerations and Motion Perception . . . . . 15 vii
viii
CONTENTS
3.3 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3.3.1 P-Delta (Second Order) Effects . . . . . . . . . . . . . . . . . . . . . . 16 3.3.2 Story Stability Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . 17 3.3.3 Stability Evaluation with P-Delta Analysis . . . . . . . . . . . . 17 3.3.4 Global Stability and Story Stability . . . . . . . . . . . . . . . . . . 18 3.3.5 Stability Acceptance Criteria . . . . . . . . . . . . . . . . . . . . . . . 19 3.4 Strength Evaluation of the Lateral Force–Resisting System . . . . 19 3.5 Building Displacements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.5.1 Overall Building Deflection . . . . . . . . . . . . . . . . . . . . . . . . 20 3.5.2 Story Drift . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.5.3 Drift Measurement Index . . . . . . . . . . . . . . . . . . . . . . . . . . 22 3.5.4 Recommended Drift Criteria . . . . . . . . . . . . . . . . . . . . . . . 23 3.6 Nonstructural Elements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.7 Occupant Comfort . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.7.1 Acceleration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 3.7.2 Visual and Auditory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.8 Project-Specific Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 4
PRELIMINARY STRUCTURAL DESIGN . . . . . . . . . . . . . . . . . . . 29 4.1 Purpose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.2 Preliminary Wind Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.2.1 Along-Wind Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 4.2.2 Crosswind Response. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.3 Estimation of Building Performance . . . . . . . . . . . . . . . . . . . . . . 30 4.3.1 Preliminary Structural Analysis . . . . . . . . . . . . . . . . . . . . . 31 4.3.2 Strength Checks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 4.3.3 Building Periods and Mode Shapes . . . . . . . . . . . . . . . . . . 32
5
WIND CLIMATE ASSESSMENT . . . . . . . . . . . . . . . . . . . . . . . . . 35 5.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 5.2 Davenport Wind Loading Chain. . . . . . . . . . . . . . . . . . . . . . . . . 35 5.3 Wind Climate: Storm Types and Data Sources. . . . . . . . . . . . . . 36 5.3.1 Windstorm Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 5.3.2 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 5.4 Influence of Terrain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 5.5 Extreme Value Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 5.6 Design Criteria: Mean Recurrence Intervals . . . . . . . . . . . . . . . . 40
6
WIND TUNNEL TESTING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6.2 Triggers for Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 6.3 Types of Wind Tunnel Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 6.3.1 High-Frequency Balance. . . . . . . . . . . . . . . . . . . . . . . . . . . 43 6.3.2 High-Frequency Pressure Integration. . . . . . . . . . . . . . . . . 44 6.3.3 Aeroelastic Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
CONTENTS
ix
6.4 Physical Testing versus Computational Estimates . . . . . . . . . . . 47 6.5 Testing Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 6.5.1 Timeline and Type for Testing . . . . . . . . . . . . . . . . . . . . . . 47 6.5.2 Inclusions and Exclusions. . . . . . . . . . . . . . . . . . . . . . . . . . 48 6.5.3 Required Input Information . . . . . . . . . . . . . . . . . . . . . . . . 48 6.6 Combining Climate and Wind Tunnel Data . . . . . . . . . . . . . . . . 48 6.7 Typical Outputs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 6.8 Additional Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.8.1 Shielding and Influence from Surrounding Buildings . . . . 51 6.8.2 Design Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 6.8.3 Minimum Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 7
DAMPING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 7.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 7.2 Inherent Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 7.3 Aerodynamic Damping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 7.4 Supplemental Damping. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 7.4.1 Direct Damping Systems . . . . . . . . . . . . . . . . . . . . . . . . . . 55 7.4.2 Indirect Damping Systems . . . . . . . . . . . . . . . . . . . . . . . . . 56 7.5 Supplemental Damping: Strength and Serviceability . . . . . . . . . 58
8
STRUCTURAL MODELING AND ANALYSIS . . . . . . . . . . . . . . 59 8.1 Structural Modeling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 8.1.1 Strength-Level and Serviceability-Level Analysis . . . . . . . 60 8.1.2 Primary Lateral Load–Resisting System and Nonparticipating Elements . . . . . . . . . . . . . . . . . . . . . . . . . 60 8.1.3 Building Mass . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 8.1.4 P-Delta (Second Order) Effects . . . . . . . . . . . . . . . . . . . . . . 62 8.1.5 Diaphragms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 8.1.6 Foundation Flexibility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 8.1.7 Panel Zone Deformations . . . . . . . . . . . . . . . . . . . . . . . . . . 64 8.2 Special Considerations for Reinforced Concrete Structures . . . . 65 8.2.1 Expected Strength and Modulus of Elasticity of Concrete Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 8.2.2 Stiffness Modifiers and Behavior of Cracked Reinforced Concrete Structures . . . . . . . . . . . . . . . . . . . . . 66 8.2.3 Simplified Method for Selecting Stiffness Modifiers . . . . . 67 8.2.4 Detailed Method for Selecting Stiffness Modifiers . . . . . . . 67
9
WIND OPTIMIZATION PROGRAM . . . . . . . . . . . . . . . . . . . . . . 71 9.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 9.2 Building Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 9.3 Building Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 9.4 Holistic Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
x
CONTENTS
10 CONCLUDING REMARKS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 10.1 Design Validation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 10.2 Peer Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 10.3 Concurrent Research and Future Directions . . . . . . . . . . . . . . . 78 10.3.1 Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 10.3.2 Performance-Based Design . . . . . . . . . . . . . . . . . . . . . . . 79 10.3.3 Computational Wind Engineering . . . . . . . . . . . . . . . . . 79 10.3.4 High-Performance and New Materials. . . . . . . . . . . . . . 79 10.4 Closing Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
PREFACE
The Tall Buildings Committee of the Structural Engineering Institute of the American Society of Civil Engineers comprises senior leaders active in the field of tall buildings from both industry-leading professional consulting firms and academic institutions. The goal of this committee is to publish an industry consensus manual of practice that provides guidelines on the design and performance of tall buildings for wind effects. This publication provides recommendations and guidance on tall building design industry standard practice and approaches to complement current literature, model codes, and other standards. This publication should be used in conjunction with local building codes and considered a current consensus document developed by industry leaders for the design and performance of tall buildings for wind effects. However, it is the opinion of the committee that this consensus will evolve over time and as the industry advances. Preetam Biswas, P.E., LEED Director of Structural Engineering Skidmore, Owings & Merrill Task Committee Chair—Design and Performance of Tall Buildings for Wind John Peronto, P.E., S.E., C.Eng, EUR ING, FIStructE, FICE, SECB, LEED AP Senior Principal Thornton Tomasetti, Inc. Technical Activities Division Chair—Tall Buildings
xi
ACKNOWLEDGMENTS
To all the individuals Maryam Asghari Mooneghi Andrew Bartolini Audrey Bentz Manotapa Bhaumik Melissa Burton Finley Charney Jon Galsworthy Mike Gibbons
who contributed to the Ramon Gilsanz Jennifer Goupil Jeremy Hasselbauer Johnn Judd Hessam Kazemzadeh Tracy Kijewski-Correa Jordan Komp Michael Montgomery Deepak Pant
discussion on content: Viral Patel Karl Rubenacker Rob Smith Seymour Spence John Tessem May Thu Nwe Nwe Un Yeong Jeong
To members of institutes and firms that contributed to the discussion on content: Arup Kinetica University of Michigan ASCE/SEI LERA University of Notre Dame BCE McNamara Salvia University of Wyoming CPP MKA Virginia Tech DCI Rad Urban Walter P Moore GMS RWDI WSP Gradient Wind SOM JHU TT A special acknowledgement to the members of the Blue Ribbon Panel for their thorough and detailed review of the contents of this manual: Abbas Aminmansour, Ph.D. William F. Baker, P.E., S.E., F.ASCE Daryl Boggs, Ph.D., P.E. David Farnsworth, P.E. William Faschan, P.E., S.E., F.ASCE xiii
xiv
ACKNOWLEDGMENTS
Robert A. Halvorson, P.E., S.E. Peter Irwin, Ph.D., P.Eng. Ahsan Kareem, Ph.D., Dist.M.ASCE Ron Klemencic, P.E., S.E. Robert Sinn, P.E., S.E.
CHAPTER 1 INTRODUCTION
1.1 PURPOSE This Manual of Practice is intended to provide best-practice guidelines for the design of tall buildings for wind. Current building codes and design standards focus primarily on strength design for ultimate wind loads, offering little guidance related to the evaluation and establishment of acceptance criteria for tall building performance under varying levels of wind effects. As such, current design practice varies widely across the industry. The goal of this document is to promote consistency and best practices within the industry for the design of tall buildings for wind. It is envisioned to apply to buildings with height greater than 120 m (400 ft) and/or a height to width aspect ratio greater than 5:1. 1.2 SCOPE The design recommendations provided in this manual are primarily focused on the wind design of building structures that are especially tall, slender, and/or prone to wind-induced movement. A discussion of both static and dynamic approaches to wind design are presented in subsequent chapters. Wind design performance objectives are achieved by • Ensuring structural integrity under ultimate loads; • Limiting lateral deflections under service loads to prevent permanent deformations, damage to nonstructural elements, or adverse effects on the serviceability of the building’s services and vertical transportation systems; 1
2
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
• Preventing excessive wind-induced motions, as well as visual and acoustical disturbances that may impact occupant comfort; • Limiting surface wind intensity to improve pedestrian comfort; and • Achieving these objectives while working to minimize structural materials and associated embodied carbon and cost. The manual provides recommendations on achieving the aforementioned performance objectives, including the selection of the wind loading mean recurrence interval (MRI) for serviceability, the establishment of acceptance criteria, guidance on addressing uncertainty in building structural properties such as stiffness and damping, as well as modeling of uncertainties in wind climate, wind loading, and other effects. A concurrent effort to this Manual of Practice is the ASCE/SEI Prestandard for Performance-Based Wind Design (ASCE 2019). The Prestandard provides a performance-based approach for the design of buildings for wind to help resolve the conflicts in performance objectives that exist in using prescriptive procedures for the wind design and performance-based approach for the seismic design of individual buildings. Among other things, the Prestandard will introduce the use of nonlinear dynamic analysis for wind design and limited nonlinearity in the main wind force resisting system elements. Both these topics are outside the scope of this manual.
1.3 USE OF THIS MANUAL This manual is compatible with the requirements of ASCE 7-16 (ASCE 2017) for the determination of wind loads and provides additional guidelines beyond the building code for the establishment and evaluation of performance objectives when designing tall buildings. The chapters are ordered in a manner that closely resembles the typical design approach for a tall building. This usually involves the establishment of performance objectives and completion of a preliminary structural design using code-based wind loads, followed by a more detailed assessment of the wind climate and wind-induced response through wind tunnel testing. The results of the wind tunnel testing program are then used to refine the preliminary design and complete the final design and construction documents. To the extent possible, the chapters are written in a self-contained format to allow ease of reference regarding specific topics. Significant topics, such as damping, have dedicated chapters and are cross-referenced from other chapters to avoid duplicating information.
INTRODUCTION
3
1.4 HISTORIC GENERAL DESIGN REQUIREMENTS Until the publication of ANSI A58.1-1972 (1972), design for wind loads was largely governed by local and regional building codes. These codes typically prescribed wind pressures to be applied to building surfaces. ANSI A58.1-1972 provided wind load criteria using probabilistically determined wind speeds and tabulated design load parameters. This probabilistic wind load approach has evolved through revisions made over the years in ANSI A58.1-1982, and then the various editions of ASCE 7 starting with ASCE 7-88 (1988), and most recently ASCE 7-16 (2017). The most significant changes have been the change to the reference wind speed from fastest-mile to a 3-s gust in ASCE 7-95, and more recently, the mapped wind speeds in ASCE 7-10. Until ASCE 7-05, the mapped wind speeds were based on a 50-year MRI. Importance factors, based on the ratio of 25-year, 50-year, or 100-year velocity pressure to the 50-year velocity pressure, were used to calculate the wind pressure for the desired MRI from the 50-year MRI wind pressure based on the building risk category. A load factor of 1.6, obtained through reliability analysis, was used to convert the service loads to ultimate loads. In ASCE 7-10, wind loads were specified directly as “ultimate loads,” with a load factor of 1.0. Back-calculations determined that the equivalent “ultimate” MRIs corresponding to the product of old importance factors and load factor aligned fairly well with MRIs 300, 700, and 1,700 years. Since ASCE 7-10, ultimate wind speed maps are published for different risk categories directly representing the 300-year, 700-year, and 1,700-year MRIs (SEAOSC 2010). Although ANSI A58.1 and ASCE 7 have code mandated the strength design requirements for wind loads, they have largely remained silent on serviceability or occupant comfort considerations. Various efforts to assess the state of the art of wind drift design practices by ASCE and other organizations such as the American Concrete Institute (ACI) revealed that a wide range of drift criteria and modeling assumptions are used by design professionals, and that an industry consensus has not been established. One of the goals of this manual is to document the industry best practices to achieve greater consensus and bring increased consistency to the practice.
1.5 STAKEHOLDERS Whereas the primary target audience for this Manual of Practice are structural engineers, it is important to recognize the different stakeholders with a vested interest in the design criteria and to coordinate the building performance objectives with these stakeholders (Riad 2016). Following is a listing of significant stakeholders and their potential interests.
4
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
• Owners: Owners are the individual(s) or organization that conceives, develops, and financially supports the project. Owners need to be involved in major decisions affecting the serviceability of the building as it may impact the functionality, cost, and value of their property. In addition, owners may request specific design criteria or design enhancements beyond what the building code provides. • Architect: The architect is conventionally the lead design consultant hired by the owner. The architect works to fulfill the owner’s brief and coordinates the design among the design consultants. The architect typically designs or participates in the design of the building enclosure, vertical transportation system, interior partitions, and other nonstructural components sensitive to the building’s response under wind. • Structural engineer: The structural engineer is engaged by the architect or owner to design the structure for the building. The structural engineer’s responsibilities include the design of the building structure under anticipated loading events including wind loads. The structural engineer is in the unique position to advise the owner, architect, and other stakeholders about the anticipated building movements in wind and resulting motions (such as accelerations) and how to control them within agreed-on acceptable limits. • Wind engineering consultant: The wind engineering consultant is typically engaged by the structural engineer, architect, or owner to provide the wind loading for use in the design of the building structure and building enclosure and to determine estimated building movements, which could impact occupant comfort. The wind engineering consultant will assess the wind climate and most often perform wind tunnel tests to obtain the wind loads for the design of the building structure and to calculate the dynamic response of the structure. • Vertical transportation consultant: The vertical transportation consultant designs the building elevator system. Excessive building sway may impact the operation of elevators. As such, the vertical transportation consultant has a stake in understanding the building’s drifts and its dynamic properties to adequately specify the building’s elevator system. • Façade consultant: The façade consultant is responsible for the design and/or specification of the building enclosure. Therefore, accurate estimation of the wind pressures on the building’s exterior surface and the wind drifts are essential for the building enclosure design process. • Building occupants: The building occupants are not usually part of the decision-making process during the design of a building unless the owner occupies the building. However, building occupants are
INTRODUCTION
5
perhaps most impacted by decisions made about the building performance. • Neighbors: Neighbors and neighboring properties may be adversely impacted by a new tall building next to their property. The wind loads on the neighboring property may be adversely affected as a result of the new tall building being constructed in the vicinity. Also, the potential noise caused by wind on exterior elements may become a nuisance. • Pedestrians: Pedestrians may be adversely impacted by a new tall building because wind speeds may increase or be redirected at the street level in the building surroundings. In some cases, recognizing the potential impact of a new building on pedestrian comfort and implementing measures to mitigate high wind speeds will help facilitate the community’s acceptance of the building. • General public: The impact of a tall building goes beyond the neighbors or pedestrians, although typically the impact is not wind related. The impact of a tall building on the general public is often reflected in the municipal approvals process, which may require additional studies of the impact of the building on the neighborhood and city.
1.6 NATURE OF WIND Wind is a dynamic and random phenomenon. Wind speed can be viewed as a mean value on which random fluctuations or “gusts” (dynamic effects) are superimposed. The fluctuating component is described in statistical terms such as standard deviation, power spectral density, and probability density function. Wind speeds usually increase with height aboveground, which is a characteristic of a boundary-layer flow (SP240-2) (ACI 2006). In addition to this boundary-layer phenomenon, wind has directionality, that is, winds from a particular direction may be stronger or dominate over wind from other directions. Directionality varies from region to region, and even from site to site, and is traditionally represented using a “wind rose” (i.e., a plot showing the variation of wind speed with compass direction). Further, the concept of a return period (or MRI) is used to calculate the recurrence interval of winds of a given velocity over time. The return period is estimated probabilistically using available wind records (or, when availability is limited, numerical simulation of wind events) and is an essential concept in the current wind design standards. It is also an important consideration in the building’s serviceability and performance.
6
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
The issue of wind climate assessment is discussed in more detail in Chapter 5, and broader descriptions of the nature of wind and climate are available in the literature (Simiu and Yeo 2019, Irwin et al. 2013). 1.7 LIMITATIONS This manual is intended to provide a framework for the wind design of tall buildings that is based on the current state of practice in tall building structural design and wind tunnel testing. The guidelines in this manual should result in the design of tall buildings that perform adequately in wind. Because each project may have unique circumstances or requirements, engineers and building officials using these guidelines must exercise their own independent judgment as to the suitability of these recommendations for a specific project. In addition, this manual does not cover other aspects of tall building design such as, but not limited to, seismic design, differential shortening, building lean under self-weight because of differential shortening, and so on. Engineers and building officials must exercise their own independent, professional judgment as to other design requirements.
CHAPTER 2 DESIGN PROCESS
2.1 OVERVIEW This book describes a design process which is intended to be primarily sequential, beginning with Chapter 3 and ending with Chapter 10 as shown in Figure 2-1. However, it is understood that most designs do not proceed in a linear fashion and often require redesign or reanalysis because of changes by stakeholders or natural design evolution. Furthermore, it will also frequently be necessary to skip ahead several steps in the design process. For example, the preliminary structural design will often require the development of a structural analysis model prior to wind climate assessment or wind tunnel analysis. 2.2 ESTABLISH PERFORMANCE OBJECTIVES The first step in the design process is the establishment of performance objectives in accordance with Chapter 3. The performance objectives should be determined through consultation with the building’s stakeholders and will serve as targets against which measured performance will be evaluated. This book follows conventional tall building design practice in which the performance objectives fall into two general categories: (1) structural strength and stability (under rare wind events), and (2) serviceability (under frequent wind events). Structural strength and stability requirements are related to life safety and are typically set by the local building code. Serviceability performance objectives are related to continued building operation and are not code 7
8
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Figure 2-1. Design process overview. mandated. As such, it is critical to establish early in the design process the performance objectives for design considerations such as occupant comfort, vertical transportation functionality, and damage to nonstructural elements. Because the serviceability objectives are not code mandated, for certain buildings the stakeholders may determine that project-specific performance, beyond what is achieved through conventional tall building design practice, is desired. These guidelines should not be interpreted to prevent the establishment and use of such project-specific performance objectives. 2.3 PRELIMINARY STRUCTURAL DESIGN In the preliminary structural design, the engineer must select the systems, materials, and geometric layout of the structure. Using general rules of thumb as well as code-based prescriptive methods, the engineer should determine proportions of overall structural systems and the individual structural elements. This preliminary design acts as the starting point for further design refinement and provides a basis for wind tunnel testing as described in Chapter 6. More detailed information on the development of a preliminary structural design can be found in Chapter 4. 2.4 WIND CLIMATE ASSESSMENT Wind climate analysis is key to any reliable tall building design. The wind climate analysis is the component of the design process that determines the probability of a particular wind speed occurring from any given wind direction. It is also a step that introduces a great degree of uncertainty owing to the imperfect, and sometimes incomplete, nature of meteorological data sets. To increase reliability, different probabilistic approaches may be required to best fit to the meteorological data for different return periods. For instance, for short return period events related to serviceability design,
DESIGN PROCESS
9
such as occupant comfort accelerations, a statistical fit that matches relatively commonly occurring wind events must be used. For strength design at extreme wind speeds the fit must be of an extreme value type suitable for extrapolation well beyond the data record length. For very tall buildings whose upper portions reach or exceed the height of the atmospheric boundary layer (ABL), it is advised to undertake an upper-level wind climate assessment to determine wind speeds and directions above the ABL, because these may have substantial influence on the global response. In some jurisdictions, it may be necessary to scale the strength design wind speeds to local statutory authority requirements. However, in other jurisdictions, alternative minimum design wind speeds can often be established using a suitably rigorous statistical analysis. It is recommended that serviceability assessments be based on a best estimate site-specific wind climate analysis. The second part of the wind climate analysis is the transference of wind data models to the project site, accounting for the upwind roughness for each direction. Although codes and standards typically provide simplified methods for assessing upwind terrain, more comprehensive methods may be used if appropriate. Refer to Chapter 5 for additional information related to wind climate assessment.
2.5 WIND-INDUCED LOADS AND RESPONSES The two methods of assessing wind-induced loads and responses are the use of published data or wind tunnel testing. Wind tunnel testing is recommended for the design of tall buildings because it is the only current method that allows the designer to determine the aerodynamic characteristics for specific shapes and over a full range of wind speeds. Published data can take the form of information in codes, standards, and published databases, all of which have been derived from wind tunnel testing. Some key limitations of published data are that on limited occasions they reflect the exact geometry of a building under design, the fluctuating wind-induced response of the structure may not be considered, and they do not account for the precise surroundings. Wind tunnel testing for a project overcomes these limitations by providing building and site-specific information. Although published data can provide reasonable estimates for the along-wind response, cross-wind response, which can dominate loads and accelerations for tall or slender buildings, can only be accurately determined through wind tunnel tests. This is because wind tunnel tests include the influence of surrounding buildings, whereas published data may not. Even relatively small architectural changes can significantly impact a tall
10
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
building’s response. Refer to Chapter 6 for additional information on wind tunnel testing and types of dynamic response. 2.6 STRUCTURAL MODELING AND ANALYSIS The structural analysis model should represent the properties of the full-scale building as accurately as is necessary. Appropriate structural modeling and analysis are important to ensure that tall building behavior will meet its assumed acceptance criteria once constructed. The model must include the mass and stiffness of the elements of the lateral system, inclusive of geometric nonlinearity inherent to the lateral system (P-Delta effects) and the effects of diaphragms on the distribution of lateral loads. The engineer must also make assumptions about a number of different building behavior parameters during the modeling and analysis process. These assumptions may vary depending on the particular performance objective being studied and the structural system of the building. Chapter 7 provides more information about inherent and supplemental damping. Chapter 8 provides recommendations on best practices for structural modeling and analysis and the appropriate assumptions that must be made during the process. 2.7 COMPARISON OF RESULTS TO ACCEPTANCE CRITERIA The engineer should establish acceptance criteria corresponding to the performance objectives in accordance with Chapter 3 and compare the results determined through structural analysis and wind tunnel testing to the established acceptance criteria to verify acceptable performance. Where the results do not show acceptable performance, additional design iterations may be required. 2.8 WIND OPTIMIZATION PROGRAM Although not required for all designs, there may be a desire to optimize building response with respect to wind. Chapter 9 provides guidance on the wind optimization program, which can take the form of either aerodynamic modification or aerodynamic design. Common optimization strategies include modification to the building orientation, geometry, and porosity. 2.9 FINAL DESIGN The final design phase documents the results of the design process. Once the acceptance criteria are met for the desired performance objectives, the
DESIGN PROCESS
11
final construction documents should be produced. The calculations and analyses that supported the design process should also be documented in detail where appropriate. If a peer review is deemed necessary, it should be completed at this stage of design. An appropriate and experienced peer reviewer may independently verify the assumptions made and the results of the design process. Refer to Chapter 10 for more information on the final design phase and the peer review process.
CHAPTER 3 PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
3.1 INTRODUCTION The focus of this chapter is the establishment of performance objectives for buildings subjected to wind loading. Where applicable, guidance is provided on acceptance criteria for use by the structural engineer and other stakeholders. The impacts of wind on tall buildings can be divided into two primary categories: strength and stability, and serviceability. In general, strength and stability requirements are addressed in the framework of building codes, ensuring that life-safety requirements are maintained. Issues related to the strength and stability of tall buildings are addressed briefly here, including recommendations for global stability acceptance criteria. The serviceability performance objectives for tall buildings are primarily related to controlling lateral movements (both displacements and accelerations) owing to wind. Where the acceptance criteria are not code mandated, they should be determined by the project’s stakeholders. 3.2 MEAN RECURRENCE INTERVALS Probabilistic modeling of the wind is critical for determining wind loading on a tall building structure. Sections 3.3 through 3.7 outline various performance objectives for stability, strength design, drift and deformation limits, and occupant comfort. Each performance objective should be associated with a return period or MRI appropriate for design. The following sections suggest MRIs for various performance objectives. The MRIs related 13
14
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
to serviceability should be considered as suggestions, and as such, they may vary in actual practice from those listed in this chapter. In all designs using the ASCE 7 standard or building codes based on the provisions of ASCE 7, the risk category of a tall building is determined, which helps set expectations for performance levels for service and strength design. 3.2.1 Strength: Foundation and Lateral System (Main Wind Force Resisting System) Tall buildings must satisfy strength limit states, where structural members and components are designed to support the ultimate wind loads and retain structural integrity. The MRI appropriate for strength considerations varies based on Building Risk Category. The risk categories and associated MRIs for strength design per ASCE 7-16 (ASCE 2017) are provided in Table 3-1. 3.2.2 Serviceability: Drift and Displacement ASCE 7-16 (ASCE 2017) standard advises that “serviceability limits (e.g., maximum static deformations or accelerations) shall be chosen with due regard to the intended function of the structure.” It goes on to note that “serviceability shall be checked using appropriate loads for the limit state being considered.” The code therefore allows significant freedom of choice with regard to drift and displacement limits and the mean recurrence interval for the assessment of the criteria. Griffis (1993) recommends a wind load with an MRI of 10 years for review of wind serviceability limit states of typical buildings. The basis for this MRI is similar to the average tenancy of office buildings in the United States, and exceedance of the drift and displacement limits is not a safety-related issue. Although 10 years may be suitable for interior partitions in office buildings, other building components and building types Table 3-1. Risk Categories with Associated MRIs Risk Category
MRI
I
300 years
II
700 years
III
1,700 years
IV
3,000 years
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
15
are expected to last well beyond this time frame, and it is important they can tolerate movements from larger MRIs. The commentary to Appendix C of ASCE 7-16 (ASCE 2017), Appendix CC, provides wind speed maps for MRIs of 10, 25, 50, and 100 years and notes that a 10-year MRI may be appropriate for a typical building, whereas 50- or 100-year MRIs may be more suitable for drift-sensitive buildings. For the design of tall buildings, the selection of the appropriate MRI for drift and displacement is left to engineering judgment and the project stakeholders. Suggested drift limits for 10-year and 50-year MRIs are presented in Section 3.5.4. In non-hurricane/non-tropical cyclone regions, buildings designed to 10-year MRI criteria will in general be able to meet the corresponding, less stringent limits at longer MRIs. The wind engineering consultant should provide wind loads and motion predictions appropriate for the agreed MRI. 3.2.3 Serviceability: Accelerations and Motion Perception In North America, a 10-year MRI has traditionally been used for the determination of occupant comfort. The guidelines were originally based on early comparisons of wind tunnel predictions and whether complaints had been received in the occupied buildings. Since then, wind tunnel testing and wind climate analyses have improved, and more recent comparisons have been used to refine the acceptable ranges (Irwin and Myslimaj 2008). More recent guidelines have moved to shorter return periods because these give more of an indication of how often motion is likely to be perceived in buildings—an important factor in occupant tolerance of motion. ISO 10137 (2007) is based on a 1-year MRI and has become the most commonly used guideline internationally. The office guidelines in ISO 10137 are largely consistent with the previous 5-year MRI guidelines published in ISO 6897 (1984), which were based on field experience of a wide range of building types, including both residential and commercial. The basis for the more onerous residential guidelines in ISO 10137 is not clear, and these can be difficult to meet for many buildings without the addition of supplementary damping. This highlights the need for stakeholder engagement in determining acceleration performance targets, based on the proposed occupancy and use of the building. For some supertall and super-slender buildings that may exhibit strong crosswind responses under moderate wind events, it may be appropriate to examine even shorter MRIs for comparison with perception thresholds: in many cases, it must be recognized that to do this would require extrapolation of wind tunnel data that limits the validity of the comparison. In locations with mixed wind climates, it is important to compare the influence of different windstorm types on the likely frequency of motion perception and the magnitude of accelerations during isolated extreme wind events.
16
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
3.3 STABILITY A fundamental performance objective for a tall building under wind loading is that the lateral system should maintain a minimum margin of safety against global instability, under all applicable ultimate load combinations. Structural instability is associated with excessive P-Delta (second order) amplification of deflections and forces. Global and local P-Delta effects are discussed, and limits for allowable P-Delta amplification are recommended in the following section. 3.3.1 P-Delta (Second Order) Effects The lateral stability of a structure is dependent on its lateral stiffness, lateral movement, and the vertical loads imposed on it. If adequate lateral stiffness is provided, the structure will maintain equilibrium under horizontal wind-induced movements. Regardless of the amount of lateral stiffness, any vertical compression load acting on a column or wall that has been laterally displaced will create a geometrically nonlinear action (secondorder effect) as shown in Figure 3-1. These are referred to as P-Delta (P-Δ) effects and are additional axial, bending, and shear effects that result from the gravity loads being resisted by a wind-displaced structure. Section 8.1.4 provides recommendations about the destabilizing load P, which should be considered to capture P-Delta effects under service-level and strength-level analysis. P-delta (P-δ) effects refer to the second-order deformation and forces that occur between a member’s endpoints owing to a destabilizing axial force P.
Figure 3-1. Second-order effects.
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
17
For most structures, P-delta amplification is primarily a consideration in design at the element level. For example, when a perimeter column resists a horizontal wind load, P-delta effects will amplify the moment from the horizontal load. However, there are some cases, such as a structural system that relies on slender moment frame columns to provide stability, in which the P-delta effect can appreciably reduce the story stiffness. In these structures, both P-Delta and P-delta effects should be considered in the analysis. The commentary for Chapter C of AISC 360-16 (AISC 2016) contains two benchmark problems that may be used to confirm that a given software program adequately considers these effects. In some cases, it may be necessary to subdivide column elements to capture P-delta effects. 3.3.2 Story Stability Coefficient In Figure 3-1, the lateral story stiffness, k, is represented by a spring that has first-order deflection Δ1 under lateral load Vx, with second-order deflection Δ2 resulting from vertical load P. The destabilizing effect is resisted by an additional load k*Δ2 in the spring. By statics, PΔ = kΔ2h. Substituting Δ2 = Δ−Δ1 and k = Vx/Δ1, it follows that the ratio of total deflection to first-order deflection (the P-Delta amplification) is Δ=Δ1 ¼ 1=ð1 − PΔ1 =V x hÞ The quantity PΔ1/Vxh is often referred to as the story stability coefficient, Q, and appears in various codes; namely, it appears as Q in ACI 318-14. For any given story, if the story height h, lateral stiffness Vx/Δ1, and gravity load P are known, the story stability coefficient Q and second-order amplification factor Δ/Δ1 may be estimated as follows: Q ¼ PΔ1 =V x h Δ=Δ1 ¼ 1=ð1 − QÞ
3.3.3 Stability Evaluation with P-Delta Analysis Historically, second-order amplification has been evaluated using the story stability coefficient approach, which lends itself well to hand calculations. In current practice, such calculations are less common because modern computer programs can easily account for these effects directly in a geometrically nonlinear analysis (often referred to as P-Δ analysis when only small displacements are considered). Still, the concept of the stability coefficient and the quantification of P-Δ amplification is useful for evaluating the stability of a structure. When second-order effects are directly considered in the analysis, a comparison between first-order and second-order
18
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
building deflections can be used to evaluate the margin of safety against instability, and a stability coefficient may be back-calculated as Q ¼ 1 − Δ1 =Δ 3.3.4 Global Stability and Story Stability Most tall building structures have substantial cantilever behavior (movements caused by axial compression and elongation of vertical elements), and thus the story drift may not be the most effective measurement of stability. For these structures, linear buckling analysis usually shows buckled mode shapes with displacement distributed throughout the structure rather than concentrated at a single story, so it is reasonable to evaluate stability globally. In this case, Δ/Δ1 may be calculated once in each direction of interest, based on the roof displacement, rather than at each level. For structures with lateral systems composed primarily of moment frames, it may be necessary to evaluate story stability, because the stability of an individual story could control the stability of the overall structure. A linear buckling analysis is useful for identifying whether buckling modes have deformations concentrated in an individual story, or if the deformations are distributed throughout the height of the structure (Figure 3-2). If linear buckling analysis is performed, the eigenvalue λ calculated for each mode represents the factor of safety against instability for that mode
Figure 3-2. Global versus local buckling determined for two structures using linear buckling analysis: (a) global buckling mode; and (b) story buckling mode.
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
19
under the initial gravity load combination considered. The corresponding stability coefficient is Q = 1/λ. 3.3.5 Stability Acceptance Criteria Because second-order deflections increase asymptotically as Q approaches 1.0, building codes have historically set limits well below 1.0 to ensure a sufficient margin of safety against collapse. The upper limit for Q given in Seismic Design Requirements of ASCE 7-16 [Equation (12.8-17)] is Q ≤ 0.25, and this limit may also be used to demonstrate adequate stability under wind loads, with the gravity loads used in the calculation of Δ/Δ1 corresponding to the load combinations which include ultimate-level wind loads. The limit of Q ≤ 0.25 is equivalent to Δ/Δ1 ≤ 1.33 and implies a factor of safety of 4.0 against instability. For most high-rise structures, significant lateral stiffness is required to meet other performance objectives, and this limit will not control the design. For the rare structure where this is not the case, a higher stability coefficient—and lower corresponding factor of safety against instability—could potentially be justified. However, the designer should be aware of the asymptotic nature of instability, and that small changes or uncertainties in the structure will result in increasingly significant changes to structural forces and deflections if larger values of Q are allowed. Stability should be considered in both orthogonal directions corresponding to the primary translational dynamic mode shapes. Torsional stability should also be considered and may be evaluated by comparing first-order and second-order displacements under a pure torsional wind case. 3.4 STRENGTH EVALUATION OF THE LATERAL FORCE–RESISTING SYSTEM Each structural element of the lateral system should be proportioned to have adequate strength under all applicable ultimate wind load combinations, including second-order effects. Strength design of structural elements under combined wind and gravity actions is covered in detail by current codes including ACI 318-14 and ANSI/AISC 360-16, among others, and is not discussed further here. The global stability and associated life-safety performance objectives do not theoretically require that every element has a demand-to-capacity ratio less than unity. The building performance could be considered acceptable if overstressed elements are shown to have adequate ductility so that redistribution of loads can occur, and the secondary load path is designed for the redistributed loads. It is the current standard for performance-based seismic design of buildings to allow controlled inelastic response under ultimate load conditions (e.g., link/coupling beam yielding, wall tensile yielding, flexural
20
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
beam yielding, steel column panel zone deformation, and steel brace yielding). Such behavior under wind loads is beyond the scope of this document but has been addressed in the recently issued ASCE Prestandard for Performance-Based Wind Design (ASCE 2019). One type of material nonlinearity that is considered under current practice, albeit typically in an indirect way, is concrete cracking. The selection of these factors will be discussed in detail in Chapter 8. It is important to note that the second-order forces used for strength design are dependent on the chosen stiffness–reduction factors. 3.5 BUILDING DISPLACEMENTS Building displacements under lateral loading can be quantified and evaluated in several ways. Three commonly used measures, in order of increasing complexity, are 1. Overall building deflection, 2. Story drift, and 3. Drift measurement index (DMI). 3.5.1 Overall Building Deflection The simplest measure of describing the displaced shape of a building is to report the lateral displacement of a single point in plan for each level of the building. Historically, it has been common to describe the roof displacement under service wind loads as a fraction of the building height, H. Although the ASCE 7-16 commentary notes that common limits used for this ratio have ranged between H/600 and H/400, quantitative limits are not recommended in this Manual of Practice because there are no performance objectives that correlate to this simple measure for tall buildings. However, it is acknowledged that overall building deflection is commonly used by structural engineers as a gauge to quickly assess the general behavior of tall buildings prior to evaluating the more rigorous drift measurements described subsequently. Care should be taken when choosing the point in plan to consider, because results toward the center of the floor plate may vary substantially from those on the perimeter of the floor plate. 3.5.2 Story Drift Story drift is the relative horizontal displacement between two adjacent floors. The story drift ratio is this relative horizontal displacement divided by the vertical distance between these floors. Figure 3-3 illustrates this relationship. The points being considered are in the same location in the horizontal plane (i.e., the points align vertically). As noted in Section 3.5.1, care should be taken in choosing the point for consideration.
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
21
Figure 3-3. Definition of story drift ratio. Story drifts are important measurements to consider in tall building behavior, having implications on stability, serviceability, and strength. Drifts can be composed of deformations caused by flexure, shear, or a combination of both. Depending on the building height and lateral system used in a tall building, drifts will usually be dominated by either total building flexure or shear. Figure 3-4 shows a portion of a building undergoing deformation from wind loading, in which Bay 2 is exhibiting primarily flexural deformation, and Bays 1 and 3 are experiencing primarily shear deformations. The drift measurement index (DMI) discussed in Section 3.5.3 is a
Figure 3-4. Flexure and shear deformations.
22
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
measure that can more appropriately capture the component of drift most responsible for building damage. 3.5.3 Drift Measurement Index Although story drift considers only the relative horizontal movement between adjacent floors, it is helpful to isolate the shear deformation (shear racking) for a given story by considering the combination of relative horizontal and vertical movements at all four corners of a structural bay or other area of interest. The DMI was introduced by Charney (1990) and referenced by Griffis (1993), who noted that it is the racking deformation, rather than simple story drift ratio, that correlates with damage to cladding and partitions. The DMI may be calculated for each individual bay within a story using the equations shown in Figure 3-5. As noted in Section 3.5.1, care should be taken in choosing the location for consideration.
Figure 3-5. DMI for a structural bay. Source: Adapted from Griffis (1993). Note: (1) Where there are no differential vertical displacements, the calculation reverts to the average of the story drift ratios at points A and B. This corresponds to a lateral system in which the columns do not participate in the lateral system. (2) Where adjacent columns are linked in the lateral system (e.g., by braces or walls), the vertical deformation components (D3 and D4) will usually be of opposite sign as the horizontal deformation components (D1 and D2), and the DMI would be less than the story drift ratio. (3) Where a participating column and a nonparticipating column are adjacent to each other, the vertical and horizontal components for that bay may be additive and the DMI would be greater than the story drift ratio.
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
23
Compared to story drift, DMI is a more accurate estimate of the deformations to be experienced by cladding or partitions. Story drift values may significantly overestimate or underestimate these distortions, depending on the structural system. If story drift is used as an approximation for DMI, the accuracy of this approximation should be verified for the building under consideration (Figure 3-5, Note). 3.5.4 Recommended Drift Criteria Table 3-2 provides recommended DMI values under service-level winds. Acceptable drifts for nonstructural elements are a function of material types, system details, manufacturer requirements, and others. Although the values shown in Table 3-2 are commonly used in practice, it is important for the structural engineer to coordinate these with the stakeholders, preferably with input from potential manufacturers during the design phase, and to make it clear in the contract drawings or specifications what these assumptions are. Less stringent criteria may be possible but would need to be evaluated by the appropriate nonstructural component manufacturers or contractors to understand the cost implications and potential constructability considerations. Information on the cladding panel sizes, partition detailing, and/or elevator shaft sizes may be needed to appropriately perform DMI evaluations for a specific component. Appropriate MRIs for drift criteria are not code mandated and are at the discretion of the structural engineer and stakeholders. Larger return periods like those used for strength design of the main lateral system of the building may be considered to investigate extreme events and their effects. 3.6 NONSTRUCTURAL ELEMENTS Controlling tall building movements under wind load to limit potential damage to nonstructural elements is an important consideration in the Table 3-2. Recommended Damage Measurement Index Criteria Recommended DMI criteria MRI
10-year
50-year
DMI criteria
1/400
1/300
Note: Story drift ratio may be used instead of DMI in some cases, see discussion in Section 3.5.3. These criteria are applicable for buildings classified as Risk Category II (10-year MRI is most commonly used). Buildings classified as Risk Category III or IV may warrant enhanced performance. See ASCE (2019) for further guidance.
24
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
design process. Recommendations are given in Section 3.5.4 and further discussed in Griffis (1993). The amount of story drift drives the potential for damage or unacceptable behavior of nonstructural elements. These drifts are manifest in relative movements perpendicular to the plane of the elements and/or parallel to the plane of the elements, also referred to as out-of-plane and in-plane movements, respectively. The in-plane relative movements can result in racking of the elements and generate shear deformations. Some structural lateral systems exhibit movements that contain both horizontal drifts as well as not-insignificant vertical movements, which further exacerbate the racking noted previously. As discussed in Section 3.5.3, the DMI is usually a more appropriate measure to consider. Closely connected to DMI measurements is the drift damage index (DDI), which is a measure of how much damage an element has sustained owing to the structural deformations from wind or seismic actions. Griffis (1993) contains a thorough discussion on this subject. Because gravity effects can exacerbate the deflections of horizontal frame members subjected to wind load deformations, it is recommended that the structural engineer consider the total deflected shape of the frame including both effects when evaluating serviceability issues. Performance objectives and potential damage thresholds for nonstructural elements should be discussed with and agreed to by stakeholders during the design process. The three categories of nonstructural elements usually needing consideration related to tall building drifts are as follows: • Components and cladding: ASCE 7 defines these as elements of the building envelope that do not qualify as part of the main wind force resisting system. In the context of tall building design, this primarily consists of the façade and cladding system and any secondary structural supports that attach the cladding to the main structural system. It also includes roof cladding and any additional parapets or dunnage elements. Failure of components and cladding elements to perform properly can result in reduced thermal efficiency, air leaks, excessive movement or vibration, visual imperfections, falling elements or spalling causing danger below, or major cracks or other failures in the façade requiring replacement. Cladding materials will play an important role in drift limits that can be tolerated without special or unusual detailing, among others. In addition to the considerations at service-level wind loads, it is also imperative that building envelope design consider the strength and behavior at large MRIs to prevent failures and/or determine acceptable levels of damage (Table 3-2). • Interior partitions: In tall buildings these elements are influenced by similar concerns and require similar considerations to those of the components and cladding elements as noted. If building movements
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
25
are not properly considered in the design and detailing of the interior partitions, issues such as visual cracking or audible creaking may be encountered. Partition materials will play an important role in drift limits that can be tolerated without special or unusual detailing. FEMA P-58 (FEMA 2018) and USG Corporation (2014) provide guidance on cracking of interior partitions. • Vertical transportation: Although shaft alignment in the as-built condition is important for the initial installation of elevator systems, long-term concerns are usually governed by potential movements and accelerations under wind loads. Modern elevators have monitoring systems that will force a temporary shutdown in the event of excessive overall building deflections and associated story drifts or reduce the travel speed under excessive building accelerations. Building movements can potentially cause vibration in the elevator cables that can also lead to temporary shutdowns. Communication with the elevator consultant regarding anticipated building movement and dynamic properties is important to limit or avoid these operational issues. 3.7 OCCUPANT COMFORT Tall buildings, like all structures, sway in the wind. Resonant motion is a design consideration because of its potential impact on the comfort of the occupants of the building. However, as human response to motion can vary significantly from person to person, it is not reasonably possible to satisfy everyone. When considering the effects of motion on the occupants, the primary aim is to limit perception of motion under normal conditions (~0.1 year MRI), to keep occupants comfortable under frequently occurring events (1-year MRI), and limit discomfort to manageable levels under less frequent events (10-year MRI). 3.7.1 Acceleration Although people are sensitive to motions in several ways, one of the most important ways is the sensitivity to sudden changes in building motion, which is usually characterized by peak acceleration. (The rate of change of acceleration, or jerk, is another possible measure but it is not commonly evaluated). This peak acceleration occurs as a tall building moves through a cycle of motion from one extreme to the other, speeding up through the center of the dynamic swing before slowing down toward the end of the dynamic swing. At the end of the dynamic swing, as the motion changes direction, the maximum acceleration occurs. Building motions caused by wind consist of two components: a static or sustained action, which is not apparent to occupants but is included in the
26
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
estimation of the building drift, and oscillatory or resonant vibration, which is caused by the dynamic and varying action of wind. This dynamic response of a tall building to wind excitation is influenced by many factors as discussed in Chapters 5 and 6. In general, occupants are more tolerant of discomfort felt infrequently or for short periods of time, as long as they feel confident about structural robustness. Early practice focused on accelerations occurring at a 10-year MRI or a 5-year MRI. ISO 6897 (1984) provided guidelines for a 5-year MRI based on interpretation of field experience from a large number of buildings and structures, some that experienced complaints and some that did not. More recently, a 1-year MRI has been the more common basis for guidelines [e.g., ISO 10137 and Architectural Institute of Japan (AIJ) Recommendations (2004)], recognizing the importance of the frequency of occurrence of perceptible motion in occupant satisfaction. These guidelines take into account the building’s natural frequency and its impact on motion perception. Figure 3-6 shows a comparison of the ISO 10137 and AIJ criteria. It is observed that ISO residential criteria roughly align with AIJ 90 curves, which indicates that 90% of people will perceive the motion. As noted, perception of motion should be expected. ISO 10137 guidelines provide frequency-dependent peak acceleration limits for the 1-year return period for both residential and commercial occupancy. However, although
Figure 3-6. Comparison of AIJ (2004) and ISO 10137 acceleration criteria MRI = 1 year. Source: Courtesy of Arup.
PERFORMANCE OBJECTIVES AND ACCEPTANCE CRITERIA
27
the commercial occupancy guidelines are consistent with ISO 6897, the stricter residential guidelines are based on factoring the commercial values by 0.7. This results in a guideline that can be hard to achieve for many buildings and more demanding than for many existing successful residential buildings. The choice of acceleration targets should be made in conjunction with all stakeholders based on the likelihood, acceptability, and impact of tenant dissatisfaction or complaints. Although peak accelerations are normally considered, some people are prone to discomfort from extended exposure to low-amplitude accelerations (Burton et al. 2015). 3.7.2 Visual and Auditory Although kinesthetic perception of sway motion in the form of accelerations is the perception mechanism most commonly discussed, there are a number of other, often more important, cues that may trigger perception of motion by building occupants. A comprehensive discussion of these can be found in Burton et al. (2015). The most common of these are visual and auditory cues. Visual cues include swinging lights, moving blinds/curtains, swaying plants, sloshing liquids, movement of other hung or loosely suspended fittings and objects, and perception of sloping floors. Significant twisting will often be detected by observers looking outside through windows and perceiving movement relative to surrounding buildings or other objects. Auditory cues are most commonly nonstructural noise, such as creaking resulting from the building sway. 3.8 PROJECT-SPECIFIC PERFORMANCE Occasionally, it may be desirable to provide enhanced performance beyond the standard performance objectives discussed in the previous sections. The project-specific performance could take the form of more restrictive drift limits, lower acceleration limits, or other criteria. These adjusted performance objectives may come at a cost premium and should be discussed with the stakeholders early in the design process. More restrictive drift limits may be desirable to accommodate brittle finishes and minimize cracking observed by building occupants in the finished interior spaces. They may also be required to accommodate smaller joints in the cladding where desirable. On some occasions, the owner may request lower acceleration limits for improved occupant comfort.
CHAPTER 4 PRELIMINARY STRUCTURAL DESIGN
4.1 PURPOSE In all but the simplest cases, the structural design process is iterative rather than linear. This is particularly true for the design of tall buildings for wind, in which accurate wind loads are necessary to proportion the building’s lateral system. Wind loads, however, cannot be accurately estimated before the lateral system has been proportioned and the associated dynamic properties have been calculated. This chapter addresses the initial steps in this iterative process that are necessary before wind tunnel testing can take place. 4.2 PRELIMINARY WIND ESTIMATES 4.2.1 Along-Wind Response A first step in the preliminary structural design of a tall building is typically to estimate the design wind loads in the along-wind direction (parallel to the wind) using either code provisions such as those in ASCE 7-16 (ASCE 2017), or engineering judgment from past wind tunnel testing. For many tall buildings, code-estimated wind loads in the along-wind direction may be conservative because shielding from other buildings and wind climate directionality are not directly considered. A site-specific wind climate analysis may be used to improve estimates of wind-induced structural loads and responses. The ASCE 7-16 (ASCE 2017) commentary contains a simplified method for interpolating between upwind 29
30
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
roughnesses, an approach based on the work of Deaves and Harris (1978). This approach can be used along with, or without, a directional wind climate analysis (described in Chapter 5) to more accurately assess design wind speeds at a project site. 4.2.2 Crosswind Response For many tall buildings, response in the crosswind direction (perpendicular to the wind) will exceed response in the along-wind direction. Crosswind vibration in buildings is most commonly caused by vortex shedding. This is a phenomenon in which vortices are created and shed alternately from each side of the building. The rate at which these vortices are shed is a function of the wind speed, building width, and building shape. Where the frequency of the vortex shedding is close to the natural frequency of the building, the structure will experience crosswind resonance leading to, depending on the building and climate, large accelerations and/or loads. For very tall, slender buildings with low natural frequencies these large responses can occur at relatively low MRIs, with implications for occupant comfort. Even for stockier buildings, the peak crosswind response can occur at wind speeds below that associated with the strength design MRI wind speed and should be accounted for in the design. Methods of preliminary estimation of crosswind response are found in some codes and standards such as AS/NZS 1170.2:2011 (2011) or NRC (2015). These are, however, relatively crude estimates for simple building shapes without significant influence from surrounding buildings. Other publicly available sources of aerodynamic data for the preliminary design of tall buildings include the NatHaz Aerodynamic Load Database (NALD) maintained by Notre Dame University. This resource provides estimates for base reactions and accelerations for various building shapes under along-wind, crosswind, and torsional responses, based on a database of wind tunnel results for isolated buildings. The NALD and other database-enabled design (DED) tools related to wind loads are available at https://www.vortex-winds.org/. Other sources of wind-loading data can be used to refine these estimates, such as wind tunnel test data for buildings of similar shape and/or surroundings. Often this information is not available in the public domain, but wind engineering consultants may hold this data in-house for use in this type of early analytical prediction. 4.3 ESTIMATION OF BUILDING PERFORMANCE The next step for the designer is to lay out and proportion a structural system to meet key performance objectives based on the estimated wind
PRELIMINARY STRUCTURAL DESIGN
31
loads. A preliminary structural analysis should be performed to estimate a building’s lateral stiffness with reasonable accuracy and to ensure that the lateral stiffness is sufficient to meet the drift performance objective under the preliminary estimates for service-level wind loads. It is equally important to estimate a building’s mass realistically, because mass and stiffness (and their spatial distribution) are the fundamental properties affecting a building’s periods of vibration and the associated mode shapes. The periods, mode shapes, and assumed damping ratios are combined with wind tunnel test results by the wind consultant to determine wind loads and building response. These attributes, along with wind climate, the site’s surroundings, and the shape of the building, compose the key variables that determine the structural loads to be used in the final design. 4.3.1 Preliminary Structural Analysis For preliminary analysis, the primary goal is to generate sufficiently accurate structural properties for wind tunnel testing. Recommendations for this level of analysis are listed in this section. A more detailed discussion of structural modeling and analysis follows in Chapter 8. The structural analysis model may be relatively simple at the preliminary stage, but at a minimum should include the primary lateral system elements; these elements should be proportioned so that the maximum allowable movements from Chapter 3 are not exceeded under serviceability conditions. Gravity elements (those that are not part of the lateral system) may be omitted for the analysis model, but gravity loads, including the self-weight of any structural elements not included in the model, need to be included to allow • Estimation of building mass for calculation of building periods and mode shapes, • Evaluation of P-Delta effects, and • Evaluation of resistance to overturning. P-Delta effects should always be considered, and a comparison of the deflection including P-Delta effects, Δ, to the first-order deflection, Δ1, should be made to evaluate the global stability performance objective from Section 3.3.5. For concrete construction, appropriate stiffness modifiers should be used to account for cracking. The simplified approach of selecting stiffness modifiers from a table (see Section 8.2.3) is typical for preliminary analysis. The extent of concrete shear wall tensile cracking is particularly important and should be closely scrutinized. If extensive wall tensile cracking is observed under serviceability conditions, building drifts may increase significantly, and revisions to the structural system will likely
32
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
be required to meet the drift performance objective. For most tall concrete shear wall buildings, wall tensile cracking under service loads should be limited to small localized areas or avoided completely. Various other topics affecting lateral stiffness in an analysis model are discussed in Chapter 8, including panel zone deformation and foundation flexibility, and these issues may or may not warrant consideration at the preliminary stage. 4.3.2 Strength Checks It is not necessary to perform detailed member design (e.g., selecting concrete reinforcement to meet strength checks) at this stage. The focus should be to determine whether key elements of the lateral system, such as link/coupling beams, outriggers, shear walls, and so forth, are within code-mandated design limits. If they are not, these elements should be reproportioned or the structural system should be revised, or both. The feasibility of the foundation system should also be considered. If foundation uplifts are present, the ability of the foundation system to resist these uplifts should be confirmed. 4.3.3 Building Periods and Mode Shapes The preliminary structural analysis may be considered complete when the following are confirmed: • Structural mass has been accurately estimated; • The lateral system has been selected, and preliminary sizes have been assigned for lateral system components; • Applicable structural stiffness reductions, especially concrete cracking, have been considered; • P-Delta effects have been considered; • Structural drifts have been verified not to exceed the target drift limits defined for the performance objective under estimated serviceability wind loads; • Key lateral system elements can be designed; and • Foundation elements can be designed. At this point the structural mass and stiffness in the analysis model may be deemed sufficiently accurate to be used to determine the building periods and mode shapes to be reported to the wind tunnel. Ultimately, as the design progresses past the preliminary stage and wind estimates are replaced with wind tunnel loads, it is almost certain that the building periods and mode shapes will evolve. This is especially true if a wind optimization program (as described in Chapter 9) is pursued to reduce wind effects. In either case, the revised structural properties should be discussed with the wind engineering consultant to determine if it is necessary to
PRELIMINARY STRUCTURAL DESIGN
33
update the wind loads to be used for design, which could lead to further changes in the structural properties or the requirement of a new wind tunnel test. At the preliminary stage, however, the designer’s goal is simply to estimate the structural properties with a level of accuracy reflective of the information available, which should minimize the number of future design iterations.
CHAPTER 5 WIND CLIMATE ASSESSMENT
5.1 OVERVIEW An understanding of the strength and directionality of the wind climate local to the project site is crucial to ensure the safe and efficient design of a tall building. This chapter outlines the role that wind climate plays in determining wind loads on a tall building and discusses best practices for obtaining and processing data that describes the wind climate, the statistical modeling of the data, and how it is combined with the results of wind tunnel testing. 5.2 DAVENPORT WIND LOADING CHAIN A convenient approach for evaluating wind loads and the ensuing wind-induced responses of tall buildings can be described by way of the Davenport wind loading chain. This approach was developed by Alan G. Davenport over a lifetime of work in the field of wind engineering (Davenport 1977, 1982), and it ties together the various concepts used to evaluate wind actions on structures using a rational “chain of thought” process. The wind loading chain is shown in Figure 5-1. The basis of the analogy to a chain is that predictions of wind loads and effects are only as sound as the strength of the links in the chain. It is therefore rational to ensure that each link of the chain (and the understanding of each component’s impact on the predicted response) is equally strong. The first link in the chain, and the link contributing the largest amount to the variability of predictions of wind-induced response, is wind climate. 35
36
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Figure 5-1. Davenport wind loading chain. The following sections review windstorms, the influence of terrain on local wind speeds, statistical approaches to integrate wind climate with wind tunnel measurements of response, and the recurrence intervals for various limit states.
5.3 WIND CLIMATE: STORM TYPES AND DATA SOURCES The type of windstorm contributing to the wind hazard at a site will vary with the geographic location. This is important to the design of tall buildings for two primary reasons: (1) different windstorms possess different mean and gust wind profiles; and (2) it has been shown that statistical analysis of wind speed and direction data, especially through extreme value analysis, should be completed on data sets separated by storm type. 5.3.1 Windstorm Types Windstorms can be classified into the following types: • Synoptic low-pressure systems: These are large-scale storms, with their stronger winds usually thought of in terms of gale events. • Tropical cyclones: Depending on severity and location, these may be referred to as tropical depressions, tropical storms, hurricanes, typhoons, or cyclones. They are associated with extremely high wind speeds, heavy rainfall, and coastal storm surges and swells. • Thunderstorms: These are much more localized storms in which the highest wind speeds result from downdrafts impinging on the ground and spreading out. Consequently, their highest wind speeds are often close to the ground, and they have very different profiles and temporal characteristics to synoptic-scale storms. • Tornadoes: These are small-scale, fast moving windstorms that can produce extremely high wind speeds with very high rotational flows. • Topographic or thermally driven winds: These are winds created by high density air flowing from a high elevation down a slope under the force of gravity. Special regions for these types of winds are identified in the ASCE 7 wind maps.
WIND CLIMATE ASSESSMENT
37
A discussion of the local wind climate should be included as part of the submission from a wind tunnel laboratory. This discussion should include a description of the prevalent windstorm type(s) that impact the project site, how these storms have been reflected in the subsequent wind tunnel testing, and how they have been treated in the statistical analysis of the wind climate. The treatment of the wind climate may differ for strength and serviceability design requirements. The wind tunnel consultant should clearly document that the local wind climate characteristics and classification have been incorporated into the wind loads and effects predicted for the building. 5.3.2 Data Sources Data that describe the local wind climate can be either measured or simulated. Measured data are typically obtained from surface weather stations, where the data have often been archived for many decades. The wind velocities measured at surface height, typically 10 m (33 ft) above ground level, are extrapolated to building height or some other clearly defined reference height using models of boundary layer winds. Wind climate analyses conducted for tall buildings typically use measured data as their primary data set, although it may be augmented by other data sources. Surface data are commonly recorded at 1 h intervals, although some stations record data once every 3 h. Many international stations may have sparser records, depending on the location. In addition, many weather stations may have a long history of seemingly consistent data, yet the anemometer may have been relocated or replaced with one of a different type, or vegetation or architectural build-ups may have occurred, resulting in a change in its effective exposure, height above ground, and its gust response characteristics. Such effects must be identified, and the appropriate analysis must be executed to convert all readings to a consistent 3 s gust speed at a height of 10 m (33 ft) in open country. Care and attention must be given to stations that do not have continuous records, and corrections should be applied as appropriate. Simulated wind data is often used in regions having frequent hurricane or cyclonic wind events to overcome the limitations of measured data. Reliable measurements of hurricane winds are also hampered somewhat by the robustness of instrumentation in extreme events. Monte Carlo methods are used to simulate a large number of time histories of hurricane events (typically between 10,000 and 100,000 events) to increase the population of events used to predict extreme winds, which also enhances the reliability of the statistical analyses for ultimate limit state design.
38
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
5.4 INFLUENCE OF TERRAIN In tall building design practice, the structural engineer works with the client to identify specific performance objectives for serviceability and strength limit states. As noted in Chapter 2, a key design parameter is wind speed—pressure varies as a function of wind speed raised to the power of 2, but building response may vary as a function of speed to higher powers, typically in the range of 2.3 to 4. Site wind speeds for design are affected by both near-field and far-field effects, related to the proximity of surrounding buildings and the aerodynamic roughness of the terrain, respectively. Near-field effects are understood but difficult to quantify accurately using an analytical approach. It is understood that neighboring buildings having massing similar to the target building can provide shelter if located in close proximity, but interference effects such as turbulent wakes or channeling by multiple buildings can also be more onerous for design. Interference is not typically addressed in codified approaches to calculating wind-induced loads because of the dependency of the wake effect on spacing and upwind building massing and geometry. Because near-field effects are best understood through wind tunnel modeling, they form the focus of Chapter 6. Far-field effects related to the aerodynamic roughness of the ground influence both the mean and gust wind velocity profiles at a site. In nature, the progression from one terrain type to another is usually gradual, and the range of aerodynamic roughness length covers several orders of magnitude. For tall buildings it is recommended that the influence of terrain on velocity and turbulence be evaluated using rational approaches complementary to code, such as the model of the planetary boundary layer provided by the Engineering Sciences Data Units (ESDU) based on work by Harris and Deaves (1980). Changes in local topography can affect wind speeds in the near and far field. Slopes with grades greater than about 5 degrees can locally increase wind speeds near the crest of hills and escarpments, particularly close to the earth’s surface. Topographic effects are calculable (e.g., the Kzt factor in ASCE 7), and analytical methods are provided in all codes of practice. In complex terrain, topographic effects are better quantified through the use of reduced scale wind tunnel models (1:2000 to 1:5000 scale) or computational wind engineering. The wind consultant’s report should include an explanation of the techniques used to assess terrain influence and provide a description of the wind tunnel profiles used in the simulations of wind effects.
WIND CLIMATE ASSESSMENT
39
5.5 EXTREME VALUE ANALYSIS The determination of an applicable design wind speed or pressure is conventionally based on extreme value analysis techniques. This analysis involves fitting a statistical distribution to independent maxima. Most maxima extraction techniques fall into the following categories: • Annual maxima: The highest magnitude wind speed in a calendar year (nominally January 1 to December 31) is extracted for each year in the available period of record. This limits the number of maxima for fitting to the number of years in the period of record. • Monthly maxima: The highest magnitude wind speed in each month of each year is extracted. For example, if there are 30 years of data in the historical record, then 12 × 30 = 360 maxima are available for fitting. • Method of independent storms: In this method, independent maxima are extracted using a two-pass algorithm that first identifies independent events and then identifies the maxima within the events. This significantly increases the number of independent maxima to upward of 80 or 90 per year. The method of independent storms is considered more robust than the annual or monthly maxima methods because statistical independence is implicit in this approach. The wind consultant’s report should include a description of the extreme value analysis techniques used as well as a discussion regarding the maxima extraction techniques. Before fitting, extracted maxima should be classified according to storm type (e.g., synoptic, thunderstorm, and so forth). It is well established that in mixed climates, where multiple storm types contribute to the extreme wind climate, individual extreme value fits should be conducted for each storm type. For example, a site may be subject to thunderstorm, synoptic, and tropical cyclone winds. Extreme value fits should be conducted on the extracted maxima individually and then recombined to determine the overall risk. For serviceability limit states, it is common practice that the wind demand determined for any given MRI will be based on the wind consultant’s statistical evaluation of the extreme wind climate, which is typically at or below the code-prescribed wind speeds or pressures. This analysis also allows for directional wind effects to be included. For strength design, wind loads are typically provided at the MRI of interest based on the codeprescribed wind speed or pressure. In areas where no code-prescribed wind speed or pressure is available, or where the wind consultant and design team feel that the code is inappropriate for the site or building under study, a detailed wind climate report can be prepared by the wind consultant and
40
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
presented to the local building code official for approval. Sections 26.5.3 and C26.5.3 of ASCE 7-16 (2017) describe the procedures that should be followed for estimating basic wind speeds from regional climatic data. A peer review of the analysis may be recommended. 5.6 DESIGN CRITERIA: MEAN RECURRENCE INTERVALS Probabilistic modeling of the wind hazard is critical for determining the demand on a tall building structure. In most designs, the occupancy category of a tall building is determined, which sets expectation on performance levels for service and strength design. Various performance objectives for stability, occupant comfort, drift limits, and strength design are outlined in Chapter 3. Each performance objective is associated with an MRI appropriate for design considerations.
CHAPTER 6 WIND TUNNEL TESTING
6.1 OVERVIEW This chapter outlines the role that wind tunnel testing plays in determining wind loads on a tall building and accounting for the uncertainties that are contained in code-based wind analysis. The following sections discuss triggers for testing in the wind tunnel, the types and benefits of each test, how to identify the strengths and weaknesses of physical and computational testing, and test inputs and outputs. The benefits of wind tunnel testing for the design of a tall building are highlighted. 6.2 TRIGGERS FOR TESTING Satisfying the ultimate strength demand for a tall building is not a guarantee that serviceability limit states (commonly, building motion and drift) will be met. For example, even in high seismic zones where strength design is governed by seismic requirements, this is no guarantee that occupant comfort under wind excitation will be achieved. This results from the random nature of the applied wind force having energy spread across a broad range of natural frequencies and the sensitivity of a building’s response at varying frequencies. The sensitivity of a building’s response is a function of its dynamic structural properties, its geometric form, the velocity and turbulence of the approaching wind (which includes the effects of local topography), and interference effects from nearby structures. Any combination of the aforementioned variables may serve as the triggers for wind tunnel testing. However, if more than one of the following 41
42
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
conditions are fulfilled, the design team should seek specialist advice from a wind consultant and consider the use of a wind tunnel from the early stages of design. Items that might affect the wind effects on a tall building include the following: 1. Height and slenderness ratio: Crosswind response, which is not covered by many codes, will often govern the wind response of buildings with height greater than 120 m (400 ft) and/or a height-to-width aspect ratio greater than 4:1. 2. Irregular plan forms not covered by code provisions: Examples include buildings with unusual architectural features, such as tapered, twisted, or offset stories; rooftop crown features; and building appurtenances. 3. Buildings with plan forms known to be susceptible to crosswind loading or vortex shedding should be evaluated using the wind tunnel approach. 4. The site location: When a tall building is located on a site where buffeting in the wake of upwind obstructions or channeling effects caused by proximity to neighboring buildings is anticipated, special consideration is warranted. 5. Linked tower structures: Where multiple towers have linking structures, such as skybridges, it is important to understand the relative movements between the towers and/or the load transfer through the linking elements. 6. Unusual structural dynamic properties or asymmetric structural systems that may not be well covered by code provisions.
6.3 TYPES OF WIND TUNNEL TESTS For the design of tall buildings, stakeholders typically seek distributed wind loads for the design of the main wind force resisting systems (MWFRS) and foundations, and predictions of building drifts and motions for serviceability and occupant comfort. To evaluate the wind-induced responses of a tall building in the wind tunnel, there are two basic types of wind tunnel tests: (1) static (rigid) model tests, and (2) elastic model tests. In general, static model tests are undertaken using one of two methods: the high-frequency balance (HFB) approach, in which wind forces (moments and shears) are measured directly at the model’s base, or by using integration of simultaneously measured surface pressures known as high-frequency pressure integration (HFPI). The model is rigid for both approaches. The elastic model technique, known as the aeroelastic model method, incorporates a scaled elastic model of the tall building. Brief explanations of the wind tunnel test types are provided subsequently.
WIND TUNNEL TESTING
43
6.3.1 High-Frequency Balance The HFB is a device capable of measuring up to six components of force and/or moment. The balance is typically integrated with a lightweight wind tunnel model to develop high stiffness, yielding natural frequencies of vibration of the combined model/balance system that are beyond the range of interest for post-test analyses, hence the term, high-frequency balance (Figure 6-1). Mean and fluctuating loads integrated over the rigid model’s surface are measured directly by the balance. The measured, applied base moments are combined with the estimated structural modal properties to determine design wind loads. Care must be taken by the wind consultant to ensure that corrections are applied to the measured data where the building mode shapes deviate from a linear variation with height. This is an aerodynamic type model test, in which the effects of the dynamic properties are incorporated analytically. The high-frequency balance approach is well suited to wind tunnel investigations early in the design process because the models are relatively simple to build and instrument. This method is the most suitable for formfinding workshops in which geometric form is being evaluated and early design feedback is required for evaluating different structural systems.
Figure 6-1. HFB model: One Bangkok Tower. Source: Courtesy of RWDI.
44
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
6.3.2 High-Frequency Pressure Integration The HFPI technique is another aerodynamic model test that uses a rigid pressure-tapped model, typically the same model used for the determination of local cladding pressures (Figure 6-2). For this reason this approach is best suited for a building whose geometry and external features that may impact building aerodynamics are relatively finalized. Rather than measure forces directly, the simultaneous pressure integration approach instantaneously combines pressures over the envelope to derive mean and fluctuating components of the applied wind pressures. The model must contain a density of pressure taps on the surface of the model sufficient to capture key aerodynamic flow characteristics. The resonant component of the wind-induced response is calculated by directly integrating local surface pressures with the dynamic modal properties. This also permits higher modes of vibration to be included in the dynamic analysis, something that is not possible with a typical HFB test. This may be important for tall buildings with extremely long fundamental periods of vibration (i.e., greater than 10–12 s), in which secondary sway modes may be sensitive to wind-induced excitation.
Figure 6-2. High-frequency pressure integration model: Torre Shyris 18, Quito, Ecuador. Source: Courtesy of CPP.
WIND TUNNEL TESTING
45
The pressure integration technique allows the wind force distribution with height to be resolved with improved precision compared to the high-frequency balance approach. This is beneficial in predicting the distribution of quasi-static (background) loads among floors and resonant response when torsion is important. The pressure integration technique is also useful for tall buildings with different structural systems over its height, for example, a concrete tower with a lightweight steel rooftop crown feature. The integration of pressures can be undertaken to inform the design of smaller architectural/structural features such as rooftop crowns and winter gardens, among others, and does not require a separate test. Where higher modes of vibration may be excited by the wind, it is recommended to verify wind-induced response predictions from early HFB studies using the simultaneous pressure integration approach where possible. For particularly tall and slender buildings, or those with aerodynamically significant architectural elements that cannot be pressure tapped, the high-frequency pressure integration approach may not be possible because of the large number of pressure taps required and the inability to extract the tubes associated with these within the small plan form of the building. Similarly, for buildings with very complex architecture, it may not be possible to include a sufficient number of pressure taps to adequately define the instantaneous pressure fields over the building envelope. 6.3.3 Aeroelastic Method An aeroelastic model is designed to have scaled geometry, mass, stiffness, modal displacements, and damping characteristics that represent the full-scale prototype of the building (Figure 6-3). In other words, the model will respond to applied wind forces elastically by vibrating in the same fashion as the actual building. The benefit of the elastic response is that aerodynamic forces that may occur owing to large-amplitude motion of the building will be captured in the measurements of the model’s response. Aeroelastic models may be simplified to evaluate only the fundamental sway modes, or they may be more complex to permit measurement of the response of the building in higher modes or in torsion. An important aeroelastic response occurs when the motion of a structure results in a change in the wind loading. If the resulting change in wind loading acts in phase with the structure’s velocity, it may reduce the aerodynamic stability of the structure. It is convenient to include these velocity-induced forces in the response equation of the structure, and the commonly used term for these forces is aerodynamic damping. For most tall buildings, aerodynamic damping will be positive. This can be beneficial in cases in which the predicted wind-induced response(s) marginally exceeds the performance criteria established by the design team, and the positive
46
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Figure 6-3. Full aeroelastic model: 432 Park Avenue. Source: Courtesy of RWDI. aerodynamic damping will lead to small reductions in the response. For particularly slender buildings, aerodynamic damping in the crosswind direction may be negative, particularly around critical wind speeds for vortex shedding. This effectively reduces the total damping of the building. This is not an effect that may be measured using rigid model methods approaches, although the results from HFB and HFPI testing can give an indication that the building is operating in a wind speed range where this is possible. Where negative aerodynamic damping may occur, an aeroelastic model test is recommended. When designing for ultimate limit states, the controlling response may occur at shorter MRI wind speeds. This would be as a result of vortex shedding, and the impact of negative aerodynamic damping needs to be considered. The aeroelastic study is the most cost- and time-intensive wind tunnel test for a tall building. If deemed necessary, it is often performed in the latter stages of design. The aeroelastic model can provide a final high-fidelity confirmation of the wind-induced response when required, although care needs to be taken with buildings with complex coupled modes to ensure that the necessary parameters are adequately modeled (ASCE 49-12).
WIND TUNNEL TESTING
47
6.4 PHYSICAL TESTING VERSUS COMPUTATIONAL ESTIMATES Wind tunnel testing is the established tool used to model wind flow around buildings and structures to develop design data for cladding and structural wind loads. It is also used to conduct pedestrian-level wind studies and assess pollutant dispersion. Wind tunnel modeling provides a good representation of both mean and turbulent wind effects, and there are a number of full-scale validation studies of model-scale wind tunnel predictions (Li et al. 1998, Kijewski-Correa et al. 2006) which substantiate the approach. Computational wind engineering (CWE), typically using computational fluid dynamics (CFD) techniques, can be a useful tool for simulating flow behavior and some flow physics that cannot be achieved in a wind tunnel. However, CWE modeling requires specially trained analysis, and techniques have not yet reached the stage of development at which they can routinely and accurately predict separation and turbulent eddies and gusts within the urban environment, as well as flow around buildings. Although significant advances are being made in the field of CWE, at present there are limited CWE studies that provide accurate validation of global wind loads for structural design for simple isolated buildings or clusters of buildings. Given the massive computing resources that would be required to adequately model atmospheric flows through a typical urban environment to predict loads on an architecturally complex building, it will likely be a significant number of years before CWE is capable of predicting overall wind loads on buildings with the necessary accuracy. In the meantime, wind tunnel testing is the recognized approach to determine the various wind effects on buildings. 6.5 TESTING PROCEDURE A number of standards and guidelines for wind tunnel testing procedures are available, such as ASCE 49-12 (ASCE 2012), which is referenced by ASCE 7-16 (ASCE 2017), the ASCE Manual of Practice 67 (ASCE 1999), and the Council on Tall Buildings and Urban Habitat (CTBUH) guideline, Wind Tunnel Testing of High-Rise Buildings (Irwin et al. 2013). The following section reviews wind tunnel model types and what stages of design they are appropriate for. 6.5.1 Timeline and Type for Testing The design of a tall building typically progresses through three distinct design phases: concept and schematic design, design development, and the technical design or construction drawings phase. During the concept and schematic design phase, high-frequency balance tests are most often used
48
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
to supplement the building code for the initial assessment of overall wind loads and responses. The HFB approach permits the most economical and quickest investigation of wind effects and provides rapid feedback to the design team. Once the architectural design has been finalized, the overall wind loads and responses can be determined using either the HFB or HFPI method, depending on the complexity or slenderness of the architecture. Aeroelastic tests are normally specified after HFB or HFPI testing if the results from these tests indicate that aerodynamic damping effects may be significant for design. 6.5.2 Inclusions and Exclusions Existing buildings and buildings currently under construction should be included in the wind tunnel models. Where future development is known, it should be considered for inclusion as an additional test configuration if the wind tunnel consultant believes it may significantly affect the results. Some local planning authorities maintain a list of developments that have received planning consent. This information can be used to guide the design team regarding future developments to include in the additional tests. If this information is not available, the design team in conjunction with the wind consultant should reach a consensus regarding future development to include in the additional tests. As noted in Section 6.8.3, specific testing of alternative surroundings may be required to address minimum thresholds required by code. 6.5.3 Required Input Information To undertake wind tunnel studies for structural loads, several inputs are required by the wind tunnel consultancy from the design team. Wind tunnel models may be constructed using two-dimensional (2D) plans, sections, and elevation drawings, and are commonly built of balsa wood, high-density foam, and thermoplastic resins. However, it is now more common for the wind engineering consultant to be provided information in three-dimensional (3D) format, and 3D rapid prototyping technologies are increasingly used for model manufacture. Analysis of wind tunnel data requires the following structural information as basic inputs: mass distributions, stiffness centers and mass moments of inertia, modal displacements provided about a reference coordinate, frequencies of vibration, structural damping ratios, drawings of the structural model coordinate origin, and orientation of the structural axis system. 6.6 COMBINING CLIMATE AND WIND TUNNEL DATA Several approaches may be used to incorporate wind climate statistics into the prediction of wind-induced response of tall buildings, each with
WIND TUNNEL TESTING
49
its own strengths and limitations. These approaches, in order of their complexity, are as follows: • Nondirectional: This is a simple approach that is compatible with code-mandated design wind speeds, but is the most conservative approach, not taking into account any directionality. • Sector method: This method introduces directionality, with the highest directional wind speed(s) scaled to match code-mandated values. As such, it is simple to understand and allows easy checking of results but relies strongly on the judgment and expertise of practitioners developing the directional model. • Storm passage: This is a direct method of calculation using time histories of storms, but it is computationally intensive and relies heavily on the quality of the input wind data. • Up-crossing: This was the first method used to provide a statistical fit to the probability of occurrence of a load effect, but it includes some approximations and is difficult to explain and check, owing to its mathematical complexity. • Multisector method: This is a more computationally intensive joint probability approach that allows determination of the statistical probability of a load effect using statistical fits to wind climate data without any additional approximations. Further details and comparisons of these techniques can be found in a number of publications, including Davenport (1977), Lepage and Irwin (1985), Gamble et al. (2001), Isyumov et al. (2003), Irwin et al. (2005), Holmes (1990), and Holmes and Bekele (2015). There is not a complete consensus in current wind engineering practice regarding the method used to integrate measured wind-induced response with directional climate effects, and this can vary geographically. For the design of a tall building it is sensible to account for directional wind effects in a more robust way than through the directional factor Kd in ASCE 7-16 (ASCE 2017). Regardless of the approach taken, the wind consultant’s report should include a discussion of the statistical method used to integrate wind climate statistics with the wind effects measured in the wind tunnel and address the limitations of the approach. 6.7 TYPICAL OUTPUTS On completion of the wind tunnel tests, aerodynamic data are combined with wind climate data to predict the overall wind-induced responses for varying return periods, using approaches presented in Chapter 5. Typical outputs from the predictions include maximum base moments and shears for foundation design, effective static floor-by-floor loads at
50
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
appropriate MRIs for strength and serviceability considerations, and a reasonable number of recommended load combinations to permit evaluation of critical load effects in the building’s structure. An example is provided in Figure 6-4, which illustrates the variation of base moments with wind direction and the sensitivity of loading to building frequency. The total number of recommended load combinations may vary, but for a single tower it is common to have between 10 and 24 possible loading scenarios to consider for design.
Figure 6-4. Base moment response versus wind direction: sensitivity to frequency. Source: Courtesy of CPP.
WIND TUNNEL TESTING
51
Component and resultant building accelerations are also predicted at appropriate MRIs to permit evaluation of occupant comfort and tolerance to building motions. The building accelerations should be compared to applicable guidelines [e.g., ISO 10137 (2007)] and reviewed with the building owner and other stakeholders to evaluate the serviceability performance of the building for occupant comfort. 6.8 ADDITIONAL CONSIDERATIONS In addition to the triggers for wind tunnel testing outlined earlier in this chapter, there may be additional considerations that require a wind consultant’s input to ensure a successful wind design for a tall building. 6.8.1 Shielding and Influence from Surrounding Buildings The wind loads on a tall building in an urban setting may be influenced by the interaction of wind with adjacent structures upstream of the building. Large buildings in close proximity to a study building may offer sheltering effects, which are beneficial. In other cases, the effects may be adverse. The magnitude of wake effects created by an upstream building is sensitive to the massing (shape, width, and height) of the adjacent structure, as well as the spacing between the adjacent building and the building under study. Because of the relative complexity of the issue, codes of practice do not currently offer guidance on wake interference effects. Wind tunnel tests are recommended to evaluate interference effects where they are suspected to be adverse. If future development is known, this should be included as an additional test configuration. 6.8.2 Design Evolution Development of design may continue beyond the date of submission of the wind consultant’s reports. Where revisions to the geometry of the building, effective sail area (changes in height or width), and/or dynamic structural properties has occurred, it is recommended that the revisions be reviewed with the wind consultant to assess the impact of the changes. If changes are deemed significant, additional testing may be required. 6.8.3 Minimum Thresholds Lower limits on loading for MWFRS and pressures for components and cladding should be considered. The following reflects the current state of the art regarding minimum thresholds for wind loads. When code-mandated minimum design wind speeds are reasonably close (within approximately 5%) to actual predicted strength-level design wind speeds, base overturning moments determined from wind tunnel
52
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
testing should be limited to not less than 80% of the design base overturning moments determined in accordance with Part 1 of Chapter 27 of ASCE 7-16 (2017). If it can be shown that lower values are not a result of shielding from specific influential nearby buildings, the lower limit can be reduced to 50%. Additional testing with influential nearby buildings removed is usually required to justify the reduced lower limit.
CHAPTER 7 DAMPING
7.1 OVERVIEW Damping plays a key role in the wind-induced response of tall buildings. The assumed level of damping in the design for strength and serviceability is an important consideration for the structural engineer and wind tunnel consultant. In general, three types of damping are considered for tall buildings: inherent damping, aerodynamic damping, and supplemental damping. 7.2 INHERENT DAMPING Two primary sources contribute to inherent damping in tall buildings. The first and most commonly known is the damping provided by the structural system itself. The second source is damping provided by secondary nonstructural components of a building such as the exterior wall, interior partitions, and finishes. Damping provided by a structural system for wind-induced responses varies depending on the type of structural system, the material composing the system, and the amplitude of displacements under different recurrence periods of considered wind events. Table 7-1 provides a summary of common structural system critical damping ratios typically considered for the design of tall buildings. Assumed inherent damping levels for wind-induced response have been traditionally lower than for seismic-induced responses because of assumed lesser structural and nonstructural damage owing to wind loads. Building height, aspect ratio, and structural system should be considered when selecting 53
54
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Table 7-1. Recommended Critical Damping Ratios Recommended critical damping ratios Structural steel systems (%)
Reinforced concrete systems (%)
1–10
0.75–1.0
1.0–2.0
50–1,700
1.0–1.5
1.5–2.5
MRI (years)
damping values to use for analysis. Damping ratios for hybrid or composite systems may fall between the steel and concrete values noted previously. Given the variability in reported damping levels in tall buildings, it would be prudent for the structural designer and wind consultant to consider the impact of damping assumptions parametrically as part of the design process. 7.3 AERODYNAMIC DAMPING Aerodynamic damping of a tall building structure can be a significant contributor to a slender tall building’s response to wind-induced forces for both strength and serviceability considerations. This source of damping is a result of the aerodynamic response of a tall building structure and is influenced primarily by the building’s structural stiffness, generalized mass, and building shape. Although this damping is traditionally considered to be a positive addition to a tall building response (i.e., increase in total damping), in certain cases it can also negatively contribute to the effective damping of a building, especially in the case of crosswind response close to the critical reduced velocity. Estimates of aerodynamic damping can be made for simple shapes from published data (e.g., Steckley 1989). This aerodynamic damping of tall buildings is most appropriately evaluated using aeroelastic wind tunnel model testing as discussed in Chapter 6. 7.4 SUPPLEMENTAL DAMPING With the relatively recent increase in the number of supertall towers, tall building designs are now more commonly utilizing supplemental damping systems to further control and optimize the structural system’s responses to wind. With the goal of implementing additional effective damping to the building’s response, supplemental damping can be provided by either direct damping systems or, more commonly, indirect damping systems.
DAMPING
55
7.4.1 Direct Damping Systems Where damping devices are installed and integrated directly into the structural system of a tall building to provide supplemental damping, this is considered a direct damping system. Although there are many proprietary products available, the vast majority of the devices available use viscous or viscoelastic materials that provide additional damping when load (or displacement) is applied to the device. Some examples of viscous or viscoelastic damping devices are piston-style fluid viscous damping devices (Figure 7-1) and wall coupling/link beam polymer-joint damping devices (Figure 7-2). If direct damping devices are used, the design and analysis of the tall building structural system must directly incorporate the dynamic behavior of the devices being implemented because the devices are in the direct load path of the lateral system.
Figure 7-1. Piston-style fluid damping device. Source: Courtesy of Arup.
Figure 7-2. Wall coupling/link beam polymer-joint damping device. Source: Courtesy of Kinetica.
56
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
7.4.2 Indirect Damping Systems Indirect damping systems are the most common devices used to provide supplemental damping for tall buildings and improve performance. These devices are independent of the lateral load-resisting system. They function by providing inertial mass that is tuned to oppose the resonant wind-induced response of the structure. For a tall building, the peak response of interest is typically at the top of the structure, which corresponds to the most effective location to provide an indirect damping system. The following different devices are used and are selected depending on the level of supplemental damping desired, as well as the required effective mass: • Tuned liquid column damper (Figure 7-3), • Tuned sloshing tank damper (Figure 7-4),
Figure 7-3. Tuned liquid column damper. Source: Courtesy of RWDI.
Figure 7-4. Tuned sloshing tank damper. Source: Courtesy of RWDI.
DAMPING
57
• Tuned mass damper examples include a simple pendulum damper (Figure 7-5) and compound pendulum damper (Figure 7-6), and • Active mass damper (Figure 7-7).
Figure 7-5. Tuned mass damper: Simple pendulum damper. Source: Courtesy of RWDI.
Figure 7-6. Tuned mass damper: Compound pendulum damper. Source: Courtesy of RWDI.
58
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Figure 7-7. Active mass damper. Source: Courtesy of RWDI. 7.5 SUPPLEMENTAL DAMPING: STRENGTH AND SERVICEABILITY Traditionally, supplemental damping for wind resistance has been implemented to improve the serviceability performance of the building. However, in recent years, the tall building industry is taking steps to use appropriate supplemental damping technology for strength design of the primary structural system. Many factors are leading the industry to push in this direction, including the high reliability of such damping devices, and the consideration that supplemental damping systems are currently used and prevalent in the strength design for seismic loading events in the tall building industry. If supplemental damping is used to reduce strength-level wind loads, special consideration should be given to the appropriate strength-level return periods to be considered with and without the additional damping, such that temporary conditions before damper installation are sufficiently covered in a tall building’s strength design. It is critical that the chosen damping system is robust and redundant enough to achieve appropriate performance for strength-level design events. A robust long-term inspection and maintenance program must also be implemented, in compliance with manufacturers’ recommendations and design professionals’ requirements. It must be guaranteed that this system will perform and that the required maintenance is conducted over the life of the building for a supplemental damping system to be used for strength-level design considerations.
CHAPTER 8 STRUCTURAL MODELING AND ANALYSIS
8.1 STRUCTURAL MODELING The two primary objectives in the structural modeling and analysis of a tall building for wind effects are to (1) estimate the structure’s stiffness and mass as accurately as reasonably possible to allow reliable predictions of the structure’s dynamic properties and movements, and to proportion the structure with adequate stiffness for serviceability considerations; and (2) evaluate the loads that are resisted by the elements that participate in the lateral load–resisting system so that these elements may be proportioned with adequate strength for ultimate load combinations. For design purposes, it is neither realistic nor necessary for a structural model to exactly replicate the building it is intended to represent. A relatively simple model is often sufficient to achieve the design objectives if the significant contributors to lateral stiffness and mass are considered and appropriately included. Numerous decisions and assumptions are necessary to construct and analyze a structural model of a tall building, and these choices affect the results of the analysis model. In turn, these results affect the structural properties provided to the wind tunnel consultant, which will affect the structural wind loads and accelerations, and how those loads are distributed to the individual elements composing the structural system. This chapter discusses a variety of decisions that must be made in the modeling and analysis phase of the design process, including • Designation of lateral load–resisting elements, • Mass assumptions, 59
60
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
• • • • • •
Consideration of P-Delta effects, Modeling assumptions for floor diaphragms, Foundation flexibility, Panel zone deformations at beam-column joints, Expected strength and modulus of elasticity of concrete materials, and Selection of stiffness reduction factors for reinforced concrete elements.
This chapter is not intended to provide a prescriptive set of recommendations, but rather to highlight which important concepts should be considered in the analysis of a tall building. 8.1.1 Strength-Level and Serviceability-Level Analysis Because both strength-level and serviceability-level performance objectives are considered, it is appropriate for distinct structural analysis to be done for each of these levels. For reinforced concrete structures, different stiffness assumptions may apply to each case. A strength-level analysis alone may be sufficient in some cases in which building drift and accelerations do not control the building’s performance. Multiple levels of serviceability (e.g., 1-, 10-, 25-, and 50-year MRIs) could also be considered. 8.1.2 Primary Lateral Load–Resisting System and Nonparticipating Elements The engineer should identify the structural elements that are intended to contribute to the primary lateral load–resisting system and those intended to be gravity load–resisting elements only (nonparticipating). Appropriate measures should be taken to either limit or account for the wind forces that may develop in these nonparticipating elements owing to their relative stiffness and/or connectivity. Simplified structural models are usually preferred where possible; however, the development of more powerful and user-friendly structural analysis software has made it more practical for structural engineers to add greater levels of detail to their analysis models. Although more detailed structural models may be beneficial in some situations, unnecessary increased detail can create more difficulty to clearly understand and verify the appropriate load path for the distribution of the wind loads throughout the structure. Further, modeling of all structural elements without proper consideration may also result in the unintended consequence of gravity-only elements attracting wind loads without proper design or detailing. The designer may choose to apply end releases or stiffness modifiers to elements that are not intended to contribute to the resistance of lateral loads. This is of primary importance for the strength-level model, in which it is critical to ensure that no portion of the lateral load is neglected and that
STRUCTURAL MODELING AND ANALYSIS
61
the load is instead redistributed to the intended lateral load–resisting elements. For the service-level model, it may be preferable to omit these end releases or stiffness modifiers to allow a more realistic estimate of stiffness for serviceability considerations. Both aforementioned approaches—using member end releases or stiffness modifiers—implicitly rely on structural ductility, or at least concrete cracking, to justify the redistribution of loads away from nonparticipating elements. The engineer should identify conditions for which an unrealistic amount of ductility is required. If this is the case, the originally “nonparticipating” element should be considered part of the lateral load–resisting system and should be designed and detailed for the induced lateral loads. Nonstructural elements are always considered nonparticipating elements. Cladding and partitions should be detailed based on the anticipated building movements, with joints large enough to avoid the participation of these elements in the resistance of lateral loads. If nonstructural elements do attract lateral loads, the primary concern is that these elements would be damaged, as discussed in Section 3.6. A secondary effect may be that if the participation of nonstructural elements is significant (as has been observed in some buildings with precast concrete cladding), the building may be noticeably stiffer than predicted by the analytical model. Although such stiffness contributions from nonstructural elements may benefit the final dynamic behavior of the building, these effects cannot be reliably predicted in the design phase, and it would not be appropriate to include any stiffness contribution from an element that is not designed and detailed for the load it attracts. 8.1.3 Building Mass The wind loads and accelerations determined by the wind tunnel consultant are dependent on the modal information provided by the structural engineer. The self-weight of the structural elements of the building contributes significantly to the overall mass, and typical assumptions for the mass density of concrete may not be appropriate in cases in which high strength and/or heavily reinforced concrete is used. When the structural mass is determined automatically by the structural analysis software, care should be taken to include mass for structural elements that were intentionally omitted for simplification of the analysis model. Mass from the superimposed dead load (SDL) and live load (LL) should be realistic rather than based on the values used for gravity system design, which are typically conservatively set by the building code. The percentage of SDL considered for building mass can vary depending on the building type and the specifics of the finishes, but typical values range from 50% to 75% of the design loading. The LL measured in actual buildings has been found to be substantially lower than the code-prescribed design LL, with
62
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
realistic ranges between 5% and 25%. Special consideration should be given to ensure the mass at mechanical floors is appropriately estimated. ASCE 7-16 (2017) Table C4.3-2 provides additional guidelines on expected in-service live loads. 8.1.4 P-Delta (Second Order) Effects P-Delta effects should always be considered, both for static and dynamic/modal analysis. Most modern analysis software programs have built-in capabilities to consider P-Delta effects. To implement these capabilities, it is critical that the gravity loads are properly modeled such that the software can accurately determine the P-Delta demands. The loads applied should be those expected during the wind event being analyzed. For serviceability models, the expected realistic dead and live loads should be used. For strength models, the applied loads should be the factored gravity loads from the ultimate load combination being considered. For the case in which the analysis model for the lateral system omits portions of the gravity system for simplicity, it is still necessary to include the destabilizing effect of all gravity loads in the lateral analysis model. 8.1.5 Diaphragms Diaphragms are structural elements that transfer loads from their points of application to the lateral force–resisting elements or between different elements of the lateral system. Diaphragm forces in tall buildings are typically carried by the floor system. It is not always necessary to fully model diaphragms in a lateral analysis for wind, but accurate assumptions in model simplification are key to achieving an accurate distribution of lateral loads in the building. There are three main types of diaphragm modeling assumptions: flexible, in which wind loads are distributed to lateral force–resisting elements based solely on tributary wind sail area; rigid, in which all points in the diaphragm are rigidly connected, resulting in wind loads distributed based the lateral stiffness of the elements that are connected to the diaphragm; and semirigid, in which the stiffness of the diaphragm is explicitly modeled, resulting in wind loads distributed based on the lateral stiffness of the elements that are connected to the diaphragm and the stiffness of the diaphragm itself. Flexible diaphragm assumptions are typically not applicable in tall building analyses and are not addressed further in this document. Rigid diaphragm assumptions are reasonable for most above-grade diaphragms in conventional tall buildings. As such, the diaphragm can be modeled as a set of rigid body constraints applied to the vertical elements at that level. This can save considerable computing demands on large buildings and is common in modern practice.
STRUCTURAL MODELING AND ANALYSIS
63
Semirigid diaphragms are necessary when modeling more complex diaphragms or where transfer between lateral elements is expected. Some examples of common configurations necessitating the explicit modeling of semirigid diaphragms are as follows: • Large reentrant corners, large openings, or openings located such that local deformations can influence force distribution; • Setbacks, transfers, or other major discontinuities of the lateral force–resisting system; • Outrigger and belt truss levels and other levels in which floors are coupled by elements other than the continuous lateral force–resisting system; • Ground level floors where significant shear is transferred to basement walls; and • Any level where structural elements within the plane of the diaphragm carry significant axial forces, such as braced frame chords, truss chords, or in-plane bracing. Diaphragms should be designed to carry the loads determined by the analyses. When the diaphragm is not explicitly modeled (i.e., a rigid diaphragm assumption is used), a simplified analysis can be performed with loads applied based on tributary area or another rational distribution method. 8.1.6 Foundation Flexibility Before the advent of modern computing, most tall buildings were designed assuming pinned bases at the foundation level, and this is usually understood to have resulted in satisfactory designs. With the increase in computing power and advancements in analysis software, it is now more practical to model the flexibility of the foundations of a building. This can be accomplished by assigning spring stiffnesses at the base of the structure, or even by explicitly modeling foundation elements and soil layers within the structural analysis model. Modeling the flexibility of the foundations is not usually considered in the calculation of the building dynamic properties provided to the wind tunnel consultant. When foundation flexibility is included, the amount of inherent damping used in the calculation of wind response should be increased to account for the additional damping contributed by the soil– structure interaction. Ultimately, engineering judgment should be used to determine when the foundation flexibility should be modeled. In some cases, doing so may be necessary to adequately capture the load distribution in the structure for strength design, even if foundation flexibility is neglected for the computation of dynamic properties.
64
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
8.1.7 Panel Zone Deformations Panel zones are the regions of beam-column intersection in moment frames (Figure 8-1). At these locations, flexural deformations are limited by the depth of the connecting member, but the concentrated moments on each side of the column create a localized shear within the panel zone. Because the resulting shear deformations within the panel zones may be significant, panel zones should not be modeled as rigid. There are several options for considering panel zone deformations. The simplest method is to use a centerline model without explicit consideration of panel zones. The resulting racking story drifts in centerline models are typically larger than those found when incorporating elastic panel zone deformations and beam/column offsets, because the overestimation of flexural deformation will usually exceed the underestimation of shear deformation within the panel zone. However, there are some cases in which a centerline model will overestimate stiffness, particularly if there is shear yielding within the panel zone. The more accurate approach is to model the panel zones, which will usually result in improved stiffness estimates compared to a centerline model. This will also result in more accurate estimates of building dynamic properties, and thus a more accurate wind load determination. Published methods for modeling the panel zones include the Krawinkler joint model and the scissors joint model (Charney and Marshall 2006). Alternatively, the panel zones may be studied using a series of subassembly finite-element models, and stiffness modifiers may be calculated and applied to centerline models.
Figure 8-1. Panel zone shear at moment frame beam-column intersection.
STRUCTURAL MODELING AND ANALYSIS
65
8.2 SPECIAL CONSIDERATIONS FOR REINFORCED CONCRETE STRUCTURES In contrast to structural steel, which demonstrates linear-elastic behavior below the yield stress with a well-defined modulus of elasticity, concrete behavior is more complicated in several ways. Concrete demonstrates: • Nonlinear stress-strain relationship; • Differing behavior in compression and tension, with tensile cracking occurring at stresses far below the compressive strength; • Significant changes in stiffness based on degree of cracking; and • Inherent variability in concrete material properties and the expected strength and stiffness. These issues merit special consideration during the design of reinforced concrete structures and are addressed in the following sections. 8.2.1 Expected Strength and Modulus of Elasticity of Concrete Materials In consideration of the statistical variability of concrete strength, ACI 301-16 (ACI 2016) requires that the average strength of a concrete mix is greater than the specified strength by some margin, which may be determined based on the sample standard deviation. This ensures a suitably low probability that a given sample will have a strength less than the specified value, but it also typically results in in situ strength greater than the specified value. This also implies that most of the concrete placed in the field will have a higher than anticipated corresponding modulus of elasticity (MOE). Because the lateral stiffness of a tall building is most closely related to the average MOE of the concrete, rather than the minimum value, it is appropriate for analysis to use the average MOE for a given concrete mix. ACI 363R-10 (ACI 2010) summarizes empirical data, which suggest that the standard ACI 318-14 (ACI 2014) equation for the modulus of elasticity for normal density concrete is usually an appropriate predictor of the average MOE for high-strength concretes used in tall buildings. Refer to ACI 363R-10 for further discussion, including alternative equations for MOE that may be more accurate if information about the mix design, coarse aggregates, and curing conditions are known. When data on expected average MOE for a given mix design is available, this may be used as a more accurate alternative to calculating the modulus based on the compressive strength. When stiffness is a critical design consideration, the designer should specify the required MOE, and the contractor should evaluate the MOE through direct measurement and preconstruction testing if representative mix design test data is not available.
66
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
The testing should demonstrate that the average MOE of the samples meets or exceeds the specified value; it is not necessary to require that all samples exceed the specified value. It is reasonable to allow the desired MOE to be achieved later than the standard 28-day or 56-day ages used for strength testing, because this will be more representative of the stiffness available during the service life of the building. For practical reasons, it still may be preferable to test 28-day or 56-day samples rather than samples that are significantly older. In this case, the designer may perform serviceability checks using an extrapolated MOE for more mature concrete if realistic concrete strength-versus-time data is available for a representative mix. 8.2.2 Stiffness Modifiers and Behavior of Cracked Reinforced Concrete Structures For buildings with reinforced concrete lateral load–resisting systems, the structure’s lateral stiffness is usually sensitive to the extent of concrete cracking, so it is essential to account for cracking in the lateral analysis of any reinforced concrete structure. Cracking may either be modeled explicitly with material nonlinearity of concrete or may be modeled implicitly by applying stiffness reduction factors. An implicit approach using stiffness reduction factors is common in practice for wind design because it allows for an elastic analysis. The discussion in this section focuses on this approach. When stiffness reduction factors are used, the goal should be to calibrate the reduction factors such that a linear analysis at a given load level results in displacements similar to those obtained from a more rigorous nonlinear material analysis or from field measurements of the completed building. Realistically, however, there will inevitably be differences in predicted versus actual stiffness. In acknowledgment of this uncertainty, it may be prudent to consider a range of stiffness modifiers, especially if wind effects are highly sensitive to the building period (see Figure 6-4 and Section 6.7 for further discussion). In general, different stiffness reduction factors should be used for different magnitudes of load. Because the typical reinforced concrete forcedisplacement curve continues to flatten as the load increases and cracking becomes more extensive, the stiffness reduction factors are usually lower at ultimate load levels compared to service load levels. There are two different approaches for implementation of stiffness reduction factors to account for cracking: (1) the simplified method in which stiffness modifiers are selected prior to the analysis, with little or no iteration required; and (2) the detailed method in which stiffness modifiers are selected in consideration of reinforcement and load in the element. In the detailed method, iteration is required.
STRUCTURAL MODELING AND ANALYSIS
67
8.2.3 Simplified Method for Selecting Stiffness Modifiers The first approach is to select modifiers based on the type of structural element (wall, column, beam, and slab, among others) and level of loading (serviceability or strength) and use the same modifier for elements of the same type throughout the height of the building. If this method is used, Table 8-1 shows an example set of reduction factors that are representative of typical practice. The values shown in Table 8-1 are intended to be an example of stiffness assumptions that may be appropriate for a typical structure, but given the inherent limitations of the simplified method, the tabulated values should not be considered mandatory. Alternate values may be more appropriate for a given structure. Determination of stiffness factors using rational analysis, such as the methods described in Section 8.2.4, will usually result in greater accuracy even if the detailed approach is not fully adopted for all elements. Implicit in the values from Table 8-1 is the expectation, in advance of the analysis, that some element types will be cracked, and some elements types will be uncracked. In general, uncracked elements have a modifier of 1.0 on the gross section properties—although modifiers greater than 1.0 can be justified if the reinforcement is considered in the calculation of transformed section properties—and cracked element types have modifiers less than 1.0. Shear walls are given special consideration in that different values are listed for uncracked shear walls and cracked shear walls. This requires an initial analysis in which the stresses in the walls are checked under the envelope of applicable load combinations, and if the tension stress exceeds qffiffiffiffi 0 the rupture stress (recommended to be taken as 6 f c for walls), then the walls are considered cracked. The extent and magnitude of the tension stresses should be considered when selecting a value within the suggested range. If there is only a small localized area of wall where the stress exceeds the rupture stress, then the upper bound modifier may be appropriate, whereas a lower bound modifier would be appropriate for a wall with tension stress greater than the rupture stress over the entire cross section. As stated in Notes f and g of Table 8-1, a similar approach may also be used to determine whether other element types would be considered cracked or uncracked for a given level of loading. 8.2.4 Detailed Method for Selecting Stiffness Modifiers As an alternative to the simplified method described in Section 8.2.3, the designer may choose to use a more detailed approach and select unique stiffness reduction factors for individual elements. This approach
68
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Table 8-1. Example Reduction Factors for Reinforced Concrete Elements Serviceability analysis 700 yearsb
Column axialc
1.0 Ag
1.0 Ag
Column flexure
1.0 Ig
0.7 Ig
RC frame beams
0.5 Ig
0.35 Ig
PT frame beams
0.7 Ig
0.5 Ig
RC slabs
0.35 Ig
0.25 Ig
PT slabs
0.5 Ig
0.35 Ig
Uncracked shear walls axial-flexured
1.0 Ig
0.875 Ige
0.6-0.9 Ig
0.5-0.8 Ig
Link beam flexuref
0.5 to 0.7 Ig
0.3 to 0.6 Ig
Link beam shearg
0.15 to 1.0 Ag
0.1 to 0.7 Ag
Wind load MRI
Cracked shear walls axial-flexured
a Typically, 1, 10, 25, or 50 years. Corresponds to applicable performance objective. See commentary to Appendix C of ASCE 7-16 (2017). b Per governing code. c Columns resisting overturning forces and under axial tension should use shear wall values for axial stiffness modification. d Walls with axial tensile stress greater than the modulus of rupture are considered cracked. qffiffiffiffi 0 A modulus of rupture of 6 f c is suggested for walls. Uniform cracking modifier is applied to vertical membrane stiffness of all shear wall elements in a lateral system. Selection of cracking modifier within suggested range should be based on the extent of cracking observed. e Corresponds to stability stiffness modifier fk recommended by R6.7.1.1 of ACI 318-14 (ACI 2014). f Link beams with flexural stress greater than the modulus of rupture are considered cracked in flexure. Link beams that are uncracked in flexure may be considered with flexural stiffness based on 1.0 Ig. For link beams that are cracked in flexure, lower bound values are appropriate for link beams with light flexural reinforcement, and upper bound values are appropriate for link beams with heavy flexural reinforcement. g Link beams with shear stress greater than the code-based strength for concrete alone are considered cracked in shear. Link beams that are uncracked in shear may be considered with shear stiffness based on 1.0 Ag. For link beams that are cracked in shear, ACI SP-240-5 (2006) reports modifiers between 0.1 and 0.4, with lower bound values appropriate for link beams with light shear reinforcement and upper bound values appropriate for link beams with heavy shear reinforcement. Note: For uncracked elements of all types, transformed section properties (including stiffness contribution of reinforcement) may be used in lieu of gross section properties.
STRUCTURAL MODELING AND ANALYSIS
69
recognizes that two individual elements of the same type will, in general, have differing • • • •
Cross sections, Loads and stresses, Extent of cracking, and Reinforcement.
Any one of these factors, or any combination of them, would mean that the two individual elements of the same type would have different force-displacement behavior, which implies that different stiffness modifiers are appropriate for each. Calculation of stiffness modifiers using the detailed method should be based on the actual load and reinforcement in the element and should approximate the secant stiffness based on that load and the expected force-displacement behavior (Figure 8-2). Different stiffness modifiers would usually be expected for the same element under service loads and ultimate loads. As an example of the detailed approach, the flexural stiffness of a reinforced concrete beam may be calculated in consideration of its moment and reinforcement using the well-known Branson equation from ACI 318 as follows: Ie ¼
Mcr Ma
3
Mcr 3 Ig + 1 − I cr Ma
Figure 8-2. Secant stiffness approximation.
70
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
The following equations for the shear stiffness of a reinforced concrete beam, in consideration of shear cracking, are recommended in ACI Committee 375’s SP-240-5 (2006) based on the work of Dilger and Abele (1974): GAv ¼ Gross GAv ¼ GAv cr ¼
Gbw d for V ≤ V c 1.2
ρv 1 + 4nρv − VVc
Es bw d
where Vc = Concrete shear strength V = Applied shear ρv = Shear reinforcing volume ratio = bAwvs Av = Shear reinforcing area s = Spacing of shear reinforcing n = Modular ratio = EEcs Es = Modulus of elasticity of reinforcing steel Ec = Modulus of elasticity of concrete bw = Width of rectangular beam d = Effective depth of beam For tall reinforced concrete buildings, shear walls often provide most of the structure’s lateral stiffness, and the lateral stiffness of the shear walls is sensitive to axial-flexural cracking. It may be prudent to scrutinize the stiffness of these elements more closely by using the detailed approach, while using the simplified approach for the other, less significant contributors to the lateral system. One approach for evaluating the stiffness of a cracked reinforced concrete wall is presented by Rahimian (2011) and includes consideration of tension stiffening behavior, that is, the stiffness provided by the concrete in between crack locations. Using Rahimian’s explicit method, the flexural cracking modifier for any wall element may be calculated based on the concrete stress-strain curve in tension, the wall’s vertical reinforcement ratio, and the wall tensile stress observed from a linear analysis. This calculation may be performed for each cracked wall element, and a unique cracking modifier may be assigned to each. Iteration is required, because the extent of cracking will usually increase once the stiffnesses of the initial cracked elements are reduced. For greatest accuracy, the described iterative procedure would be performed independently on different versions of the model—one for each load combination being considered. Alternatively, the wall stress used in the calculation may be conservatively based on the envelope of load combinations.
CHAPTER 9 WIND OPTIMIZATION PROGRAM
9.1 INTRODUCTION A building’s aerodynamic shape plays a significant role in the design of tall and supertall buildings. The goal of the aerodynamic optimization process is to determine a building geometry that provides near-optimal wind response, while also meeting all other design requirements, including architectural constraints, cost targets, and construction considerations. Aerodynamic optimization may therefore be classified into two categories: aerodynamic modifications and aerodynamic design. Aerodynamic modifications are relatively minor changes to building shape or massing owing to architectural constraints that limit the modifications to enhance wind performance. Hence, corner treatments, such as chamfering, slotting, and roundness are common approaches in this category. Aerodynamic design, on the other hand, is an approach that is integrated with the architectural design from the early stages of the project. The building shaping, massing orientation, and geometry are permitted to be influenced by results of wind studies. A series of wind tunnel tests are usually required to achieve the required number of design iterations. 9.2 BUILDING ORIENTATION The first four links in the Davenport wind loading chain (the commonly used method to determine wind-induced responses of buildings) are wind climate, influence of terrain, aerodynamic effects, and dynamic effects (Davenport 1977, 1982). Once a detailed wind climate and wind tunnel study have been completed, each of these parameters will have been 71
72
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
determined on a directional basis. Optimizing a building’s orientation aligns these directional distributions such that the wind load effect of interest is minimized. When considering wind optimization through-building orientation, it is recommended that the combined effects of wind climate and terrain be considered. Combining the wind climate model, which is natively in a reference open terrain, with the terrain and topography model will produce more valuable directional insight for the design team. Changing the orientation of the building at its base requires a flexible site location, where the building footprint is relatively unconstrained by adjacent streets and buildings. Additional flexibility is typically available at higher elevations, where a twist or change in orientation may be introduced. This can allow for the upper portions of a tall building to be more favorably oriented with respect to the directionality of the wind climate, in addition to the potential reduction in crosswind loading. The magnitude of reductions available from building orientation optimization depend on the directionality of the wind climate, wind load effects, and the flexibility of the site location. In some circumstances, this optimization could yield reductions upward of 25% for base building overturning moments.
9.3 BUILDING GEOMETRY The architectural form of tall slender buildings tends to be heavily influenced by the impact of wind, which is dependent on the geometry (i.e., shape) of the building itself. The dynamic wind response of tall structures is governed by a number of factors including shape, stiffness, mass, and damping. It is not uncommon for modern tall slender towers to suffer from dynamic vortex-shedding issues at relatively frequent recurrence intervals. The vortex-shedding characteristics of a tower are dependent on the basic tower shape via the nondimensional quantity known as the Strouhal number. Aerodynamic modifications of the tower shape can help to control dynamic motions and reduce the risk of excessive dynamic wind response. Potential modifications include recessed, slotted, or chamfered corners, horizontal and vertical through-building openings, porous tops, tapering footprints, and staggered (dropping off) corners, all of which have been shown to significantly reduce the wind-induced loads and responses in tall buildings. Figure 9-1 shows examples of corner modifications. Chamfers, twisting, and tapering of the building reduce the potential for a tower to exhibit dynamic instability, as illustrated in Tanaka et al. (2013). Aerodynamic modifications achieve this most effectively by reducing the correlation of the vortex shedding along the height of the tower.
WIND OPTIMIZATION PROGRAM
73
Figure 9-1. Aerodynamic modifications to a square building shape. Source: Adapted by Kijewski-Correa from Kareem et al. (1999).
Figure 9-2. Rate of twist shape optimization (the Chicago Spire). Source: Courtesy of TT/RWDI. In general, modifications to the building corners such as slotted or chamfered corners typically need to be greater than approximately 5% to 10% of the building width to be beneficial. Modifications that increase the projected area or the breadth of the building may improve crosswind responses, but can result in increased along-wind loads and responses. Similarly, twisting or tapering need to be significant to be substantially beneficial (Figure 9-2). Aerodynamic modifications are usually most effective when applied over the top third of the building height. The approximate effectiveness of some of the measures, in terms of the reduction of the dynamic loads or accelerations from a standard baseline building shape with a square footprint, is given in Table 9-1. 9.4 HOLISTIC OPTIMIZATION It is relatively straightforward for wind engineers to suggest aerodynamic improvements to buildings to reduce loads and responses;
74
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
Table 9-1. Approximate Effectiveness of Aerodynamic Modifications Aerodynamic modification
Dynamic load reduction (%)
Dependent on
0
—
Rounded corners
10–25
Shape and size of rounding
Chamfer
10–25
Shape and size of chamfer
Taper
0–25
Degree of taper
Twist
0–50
Degree of twist
Combination of modifications
25–75
Extent of modifications
Baseline square
however, to be an effective contributor to design, these modifications need to be considered holistically within the total design environment. The first consideration is site and building design and aesthetic constraints. The second constraint is the program and floor area requirements for the project. The third constraint for many locations is maximum height restrictions. Other considerations include the efficiency of floor plate and view corridors. In most cases, optimization is conducted to control crosswind responses, which requires a combination of aerodynamic form and structural optimization. Control of along-wind responses is more dependent on aerodynamic shape and orientation relative to critical wind directions with the structural properties typically being less significant. If a building is already at the maximum permitted height, this limits the potential for aerodynamic shaping, which usually implies a reduction in usable space on modified floors, unless there is room to increase the floor plate size. In very dense cities, this is often not possible. In contrast, on larger open sites increasing plan dimensions may be possible. The effects of aerodynamic optimization also need to be considered in relation to the effects on the structural performance of revised architectural forms, because changes in dynamic properties also affect the wind loads. Such changes in properties must be considered with regard to the critical design responses, whether these are accelerations, serviceability deflections, or strength-level design loads (Denoon et al. 2012; Tse et al. 2006a, b).
WIND OPTIMIZATION PROGRAM
75
Usually, aerodynamic optimization occurs after a set of initial wind tunnel tests have been conducted to establish a baseline against which alternative designs can be compared. This allows for the identification of the responses that are critical for design and the most appropriate form of optimization, whether aerodynamic, structural, or a combination of both, to then be determined. Aerodynamic shaping options that increase the building height have the potential to increase the along-wind responses, but if these do not control the loads or accelerations, this may be acceptable. In addition, the shaping may result in sufficiently improved response such that the need for supplemental damping can be reduced or altogether eliminated.
CHAPTER 10 CONCLUDING REMARKS
10.1 DESIGN VALIDATION As discussed in Chapter 2, the design process for tall buildings starts with the establishment of performance objectives and a preliminary structural design. This is followed by a wind climate assessment and wind tunnel study that provides more accurate wind loads for use in design and evaluates the building accelerations and movement under wind loads. The results of the wind tunnel study are then used to develop a final design that meets the performance objectives. When there are significant changes in building geometry, orientation, surroundings, or dynamic properties, it is recommended that the wind tunnel consultant review the changes and advise the design team on the implications of the changes on wind loads and building acceleration. Often, the wind tunnel consultant can extrapolate new building wind loads and dynamic response based on the original wind tunnel test. However, when the building geometry or environment changes significantly, retesting in the wind tunnel may be required. The final wind tunnel results are used for completing the structural design and producing final design documentation to be used for construction. 10.2 PEER REVIEW For tall building design, some jurisdictions may require a peer review of the structural design as part of the approvals process when certain conditions related to height, area, or slenderness ratio are triggered. However, even if a peer review is not required by code or the local 77
78
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
jurisdiction, owners often consider engaging a third-party peer reviewer to review the wind design of the primary structure for additional assurance of the quality of the design and for potential value engineering benefits. Owners may consider a peer review when the building height is greater than 400 ft (approximately 120 m) or for slender or buildings with otherwise unique attributes such as unusual or complex geometries or foundations. An independent peer review brings a different perspective and often helps identify aspects of a building design that warrant special attention. If a code peer review is required, the code or local jurisdiction likely prescribes the requirements for the review scope and process. For ownerinitiated peer reviews, the peer review may occur near the completion of the design, when the owner seeks to verify the safety of the design. If the owner is looking for value engineering benefits, a multistage peer review may be implemented to obtain input at various stages of design. This can help identify and resolve potential issues as well as allow for value engineering earlier in the design process when it is easier to incorporate design changes. It is recommended that the structural peer review be performed by a qualified independent structural engineer with experience in tall building design. The structural engineer of record should provide the peer reviewer with the drawings and specifications and structural reports, as well as the wind tunnel and geotechnical reports. The owner may also request that the engineer provide the structural analysis model to the peer reviewer or ask that the peer reviewer develop an independent analysis. The peer reviewer should provide written comments to the structural engineer, owner, and, if required, to the building department having jurisdiction. The structural engineer would then provide written responses to the review comments. Sometimes, multiple rounds of comment/response may be required to resolve issues.
10.3 CONCURRENT RESEARCH AND FUTURE DIRECTIONS 10.3.1 Monitoring Monitoring of building properties and responses can be used both during and after construction to validate and/or refine wind tunnel predictions. Measurements of natural frequencies and (estimates of) inherent structural damping can be used to provide updated predictions of building responses, particularly to help tune any supplementary damping requirements. Longer-term measurements of building accelerations can be used during occupancy to ensure that the building is performing as expected and to address any serviceability issues that might be associated with building movement. More common monitoring of field performance of completed buildings, with results published in the public domain, can only
CONCLUDING REMARKS
79
assist advances in structural engineering to increase future design efficiency and building performance. 10.3.2 Performance-Based Design Performance-based design (PBD) procedures have been developed in recent years for use in seismic design. Performance-based wind design is at an early stage of development and research, with multiple teams working on the advancement of appropriate methodologies. One of the first outcomes from this is the ASCE/SEI Prestandard for Performance-Based Wind Design (2019). This Prestandard describes a basis for how to use PBD for both structural and envelope design for buildings. It is applicable for all buildings and provides guidance on performance goals and minimum requirements that must be met in design to demonstrate that these have been met. 10.3.3 Computational Wind Engineering Computational wind engineering, commonly using computational fluid dynamics, is sometimes promoted as an alternative to wind tunnel testing to determine wind effects on buildings and structures as is done in the aerospace and automobile industries. However, ensuring the accuracy and reliability of CWE procedures requires highly experienced analysts with wind engineering backgrounds. Loads and responses based on CWE need to be carefully validated, and the reliability of CWE remains unproven for structural engineering applications. At the time of publication, there is no ASCE-accepted standard that allows the use of CWE for structural design, and there are limited validated studies that show that the use of CWE can meet the structural reliability requirements of ASCE 7. CWE can, however, be used for a range of other studies of tall building design related to serviceability issues such as pedestrian wind comfort or the investigation of air flows within the building. 10.3.4 High-Performance and New Materials Material advancements in the use of now readily available ductile high-strength steels, as well as high-strength and high-modulus concretes, will drive industry and design of tall buildings. Their constructability advancements will also be critical in their further adoption into the tall building design industry. New lightweight materials, such as engineered mass-timber may also be considered in the design of tall buildings and thus warrant consideration. Special attention and research should be dedicated to investigating mass-timber tall buildings with lower-mass aerodynamic behavior, increased inherent damping, and their system durability and robustness.
80
DESIGN AND PERFORMANCE OF TALL BUILDINGS FOR WIND
10.4 CLOSING REMARKS As design and construction techniques rapidly evolve, so will the building industry’s ability to fulfill societal ambitions for taller and more complex structures. The consideration of wind effects on tall buildings will become increasingly important as new structures continue to push the bounds of what has been previously designed. This Manual of Practice provides the framework with which tall buildings can be evaluated for wind effects. Given the continuous innovation in the field of tall building design, it is critical that structural and wind engineers keep current on new developments.
REFERENCES
ACI (American Concrete Institute). 2006. Performance-based design of concrete building for wind loads. ACI SP-240. Farmington Hills, MI: ACI. ACI. 2010. Report on high-strength concrete. ACI 363R-10. Farmington Hills, MI: ACI. ACI. 2014. Building code requirements for structural concrete. ACI 318-14. Farmington Hills, MI: ACI. ACI. 2016. Specifications for structural concrete. ACI 301-16. Farmington Hills, MI: ACI. AIJ (Architectural Institute of Japan). 2004. Guidelines for the evaluation of habitability to building vibration. AIJES-V001-2004. Tokyo: AIJ. AISC (American Institute of Steel Construction). 2016. Specification for structural steel buildings. AISC 360-16. Chicago: AISC. ANSI (American National Standards Institute). 1972. Building code requirements for minimum design loads in buildings and other structures. ANSI A58.1-1972. New York: ANSI. ANSI (American National Standards Institute)/ASCE. 1988. Minimum design loads for buildings and other structures. ANSI/ASCE 7-88. Reston, VA: ASCE. ASCE. 1999. Wind tunnel studies of buildings and structures. ASCE MOP 67. Reston, VA: ASCE. ASCE. 2012. Wind tunnel testing for buildings and other structures. ASCE/SEI 49-12. Reston, VA: ASCE. ASCE. 2017. Minimum design loads and associated criteria for buildings and other structures. ASCE/SEI 7-16. Reston, VA: ASCE. ASCE. 2019. Prestandard for performance-based wind design. Edited by Scott, D. R. Reston, VA: ASCE. AS/NZS (Standards Australia Limited/Standards New Zealand). 2011. Structural design actions, Part 2: Wind actions. AS/NZS 1170.2:2011. Wellington, NZ: AS/NZS. 81
82
REFERENCES
Burton, M., K. C. S. Kwok, and A. Abdelrazaq. 2015. “Wind-induced motion of tall buildings: Designing for occupant comfort.” Int. J. HighRise Build. 4 (1): 1–8. Charney, F. A. 1990. “Wind drift serviceability limit state design of multistory buildings.” Wind Eng. Ind. Aerodyn. 36 (1–3): 203–212. Charney, F. A., and J. Marshall. 2006. “A comparison of the Krawinkler and Scissors models for including beam-column joint deformations in the analysis of moment-resisting steel frame.” Eng. J. 43 (1): 31–48, AISC. Davenport, A. G. 1977. “The prediction of risk under wind loading.” In Proc., 2nd Int. Conf. on Structural Safety and Reliability. Munich, Germany. Davenport, A. G. 1982. “The interaction of wind and structures.” In Engineering meteorology, edited by J. Plate. Amsterdam, Netherlands: Elsevier. Denoon, R., K. Strobel, and D. Scott. 2012. “Challenging paradigms in wind engineering design of tall buildings.” In Proc., 9th World Congress of Council on Tall Buildings and Urban Habitat. Shanghai, China. Dilger, W. H., and G. Abele. 1974. “Initial and time dependent shear deflections of reinforced concrete T-beams.” In Deflections of concrete structures. 487–514. Farmington Hills, MI: ACI. FEMA. 2018. Seismic performance assessment of buildings. FEMA P-58. Washington, DC: FEMA. Gamble, S. L., R. J. Miltenburg, M. D. Cicci, and M. Accardo. 2001. “Prediction of local exterior wind pressures from wind tunnel studies using a time history analysis approach.” In Proc., 1st Americas Conf. on Wind Engineering. June 3-6, 2001. Clemson, SC. Griffis, L. 1993. “Serviceability limit states under wind load.” Eng. J. 30: 1–16. Harris, R. I., and D. M. Deaves. 1980. “The structure of strong winds.” In Proc., CIRIA Conf. London. . Holmes, J. D. 1990. “Directional effects on extreme wind loads.” In Civil engineering transactions, 45–50. Sydney, Australia: Institution of Engineers. Holmes, J. D., and S. A. Bekele. 2015. “Directionality and wind induced response–Calculation by sector methods.” In Proc., 14th Int. Conf. on Wind Engineering (ICWE14). Porto Alegre, Brazil. Irwin, P., R. Denoon, and D. Scott. 2013. Wind tunnel testing of high-rise buildings. London: Routledge. Irwin, P., J. Garber, and E. Ho. 2005. “Integration of wind tunnel data with full scale wind climate.” In Proc., 10th Americas Conf. on Wind Engineering. May 31-June 4, 2005. Baton Rouge, LA. Irwin, P., and B. Myslimaj. 2008. “Practical experience with wind-tunnel predicted tall building motions.” In Vol. 17 of Proc., IABSE Congress Report, International Association for Bridge and Structural Engineering (IABSE), Chicago: Zurich. 296–305.
REFERENCES
83
ISO. 1984. Guidelines for the evaluation of the response of occupants of fixed structures, especially buildings and off-shore structures, to low-frequency horizontal motion (0, 063 to 1 Hz). ISO 6897:1984. Geneva: ISO. ISO. 2007. Bases for design of structures–Serviceability of buildings and walkways against vibrations. ISO 10137:2007. Geneva: ISO. Isyumov, N., M. J. Mikitiuk, P. C. Case, G. R. Lythe, and A. Welburn. 2003. “Predictions of wind loads and responses from simulated tropical storm passages.” In Proc., 10th Int. Conf. on Wind Engineering. Copenhagen, Denmark. Kareem, A., T. Kijewski, and Y. Tamura. 1999. “Mitigation of motions of tall buildings with specific examples of recent applications.” Wind Struct. 2 (3): 201–251. 10.12989/was.1999.2.3.20110.12989/was.1999.2.3.201. Lepage, M. F., and P. A. Irwin. 1985. “A technique for combining historical wind data with wind tunnel test to predict extreme loads.” In Proc., 5th US National Conf. on Wind Engineering. Lubbock, TX. Li, Q. S., J. Q. Fang, A. P. Jeary, and C. K. Wong. 1985. “Full scale measurements of wind effects on tall buildings.” J. Wind Eng. Ind. Aerodyn. 74–76: 741–750. NRC (National Resource Council). 2015. National building code of Canada. Ottawa, ON: NRC. Rahimian, A. 2011. “Lateral stiffness of concrete shear walls for tall buildings.” Struct. J. 108 (6): 755–765. Riad, J. 2016. “Conceptual high-rise design—A design tool combining stakeholders and demands with design.” M.S. thesis, Dept. of Applied Mechanics, Division of Material and Computational Mechanics, Chalmers Univ. of Technology. Simiu, E., and D. Yeo. 2019. Wind effects on structures: Modern structural design for wind. 4th ed. Oxford, UK: Wiley-Blackwell. Steckley, A. 1989. “Motion-induced wind forces on chimneys and tall buildings.” Ph.D. thesis, Faculty of Graduate Studies, Univ. of Western Ontario. Tanaka, H., Y. Tamura, K. Ohtake, M. Nakai, Y. C. Kim, and E. K. Bandi. 2013. “Aerodynamic and flow characteristics of tall buildings with various unconventional configurations.” Int. J. High-Rise Build. 2 (3): 213–228. Tse, K. T., K. C. S. Kwok, P. A. Hitchcock, C. M. Chan, and R. O. Denoon. 2006a. “Building economics of wind-engineered tall structures—Part 1: Empirical response formulae of corner-modified buildings.” In Proc., 12th Australasian Wind Engineering Society Workshop. Queenstown, NZ. Tse, K. T., K. C. S. Kwok, P. A. Hitchcock, C. M. Chan, and R. O. Denoon. 2006b. “Building economics of wind-engineered tall structures—Part 2: Construction cost and returns on investment.” In Proc., 12th Australasian Wind Engineering Society Workshop. Queenstown, NZ. USG Corporation. 2014. The gypsum construction handbook. 7th ed. Hoboken, NJ: Wiley.
INDEX
Figures are indicated by f; tables are indicated by t. accelerations, 9, 15, 25–27, 61, 74–75, 78; criteria, 26f; dynamic swing, 25–26; frequencydependent, 26; limits, 27; low-amplitude, 27; peak, 25–27; static actions, 25; sustained action, 25; targets, choice of, 27 acceptance criteria, 13–27; establishing, 10; building displacements, 20–23; global stability, 13; mean recurrence intervals, 13–15; nonstructural elements, 23–25; occupant comfort, 25–27; project-specific performance, 27; stability, 16–19; strength evaluation of the lateral force-resisting system, 19–20 ACI (American Concrete Institute), 3; Committee 375, 70; SP-240-5, 70 ACI 301-16, 65 ACI 318-14, 17, 19, 65, 68 ACI 363R-10, 65 acoustical disturbances, 2 aerodynamic design, 10, 71 aerodynamic modification, 10, 71–72, 73f; approximate effectiveness of, 74t
AISC 360-16; Chapter C, 17 along-wind response, 9, 29–30, 73 ANSI A58.1, 3 ANSI/AISC 360-16, 19 Architectural Institute of Japan (AIJ), 26 AS/NZA 1170.2:2011, 30 ASCE 7, 14, 17, 24, 36 ASCE 7-05, 3 ASCE 7-10, 3 ASCE 7-16, 2, 3, 14, 20, 29, 47, 49, 62; Appendix C, 15; Appendix CC, 15; C26.5.3, 40; Chapter 27, 52; Seismic Design Requirements, 19 ASCE 49-12, 47 ASCE 7-88, 3 ASCE 7-95, 3 atmospheric boundary layer (ABL), 9 auditory disturbances, 27 axial effects, 16 base movement response, 50f bending effects, 16 Branson equation, 69 building: displacements, 20–23; drift-sensitive, 15; flexure, 21; geometry, 10, 78; height, 78; 85
86
INDEX
lean under self-weight, 6; mass, estimating, 31, 32; orientation, 10, 71–72; porosity, 10; stability, 13; sway, 27 building codes, 1, 2, 4, 13–14, 19; life-safety requirements, 13; performance objectives, 2, 13–27; regional, 3; unusual structural dynamic properties, 42 building performance: building periods, 32–33; estimation of, 30–33; mode shapes, 31, 32–33; preliminary structural analysis, 31–32; strength checks, 32 building risk category, 14 building services, 1 building vertical transportation systems, 1, 4 cantilever behavior, 18 cladding, 23, 24, 61; damage to, 22; materials, 24; panel sizes, 23; precast concrete, 63 commercial occupancy; 1-year return period, 26; guidelines, 27 computational fluid dynamics (CFD) techniques, 47 computational wind engineering (CWE), 47, 79 concrete: cracking, 20, 31–32, 66, 68–70; cracking, tensile, 65; expected strength, 65–66; force-displacement, 66; heavilyreinforced, 61; mix, 65; modulus of elasticity, 65–66; reduction factors for reinforced concrete elements, 68t; reinforcement, 32; shear, 31, 69–73; stiffness modifiers, 31, 66–70; stiffness reduction factors, 60, 66; strength, 65, 70; stress-strain curve, 70 controlled inelastic response, 19 crosswind: direction, 46; loading, 42; response, 9, 54, 74; vibrations, 30
Council on Tall Buildings and Urban Habitat (CTBUH), 47 damping, 2, 53–58; active mass damper, 58f; aerodynamic, 45–46, 53, 54; aerodynamic, estimates of, 54; for composite systems, 54; compound tank damper, 58f; direct, 55; for hybrid systems, 54; indirect, 56–58; inherent, 53–54, 79; pendulum damper, 57; piston-style fluid damping device, 55f; provided by a structural system, 53; ratios, 31; recommended critical damping ratios, 54t; supplemental, 10, 53, 54–58; supplemental, strength and serviceability, 58; tuned liquid column damper, 56f, tuned mass damper, 57f; tuned sloshing tank damper, 56f; viscous devices, 55; wall coupling/link beam polymerjoint damping device, 55; for wind induced response, 53–54 database-enabled design (DED) tools, 30 Davenport, Alan G., 35–36 Davenport wind-loading chain, 35–36, 71 deflections, 18, 74; building, 20–23; first-order, 17; second-order, 17, 19 deformations, 13, 21; flexural, 21; panel zone, 32, 60, 64; permanent, 1; racking, 22, 24; second-order, 16; shear, 21, 24, 64; static, 14 design load parameters; tabulated, 3 design process, 7–11; comparison of results to acceptance criteria, 10; establish performance objectives, 7–8, 13–27; final design, 10–11; overview, 8f; preliminary structural design, 8; structural modeling and analysis, 10; wind climate assessment, 8–9; wind-induced
INDEX
loads and responses, 9–10; wind optimization program, 10 design validation, 77 diaphragms, 10, 62–63; axial forces, 63; belt truss levels, 63; flexible, 62; ground level floors, 63; large openings, 54; large reentrant corners, 63; modeling assumptions, 63; outrigger, 63; rigid, 62; semirigid, 62–63; setbacks, 63; transfers, 63 differential shortening, 6 displacement, 14–15; first-order, 19; limits, 14; second-order, 19; vertical, 22f drift, 31; criteria, 3, 13–15, 23; horizontal, 24; limits, 25, 27, 32; nonstructural elements, 23; performance objective, 31–32; story, 18, 20–22, 24; story ratio, 21f, 22 drift damage index (DDI), 24 drift measurement index (DMI), 20–23; recommended, 23 dynamic: effects, 71; mode shapes, 19; properties, 59; response, 4, 10, 72 elevators, 4; shaft alignment, 25; shaft sizes, 23; temporary shutdowns, 25 ESDU (formerly Engineering Science Data Unit), 38 extreme value analysis, 39–40; annual maxima, 39; method of independent storms, 39; monthly maxima, 39 façade, 24 FEMA P-58, 25 final design, 10–11; peer review, 11 finite-element models, 64 floor: diaphragms, 60; mechanical, 62; plate, 20
87
foundation: flexibility, 60, 63; modeling, 63; pinned bases, 63 gravity: actions, 19; elements, 31; in-plane relative movements, 24; load, 17, 19, 31, 62; load-resisting elements, 60 height and slenderness ratio, 42 holistic optimization, 73–75 irregular plan forms, 42 ISO 6897, 15, 26, 27 ISO 10137, 15, 26 Krawinkler joint model, 64 lateral system, 29, 32; deflections, 1; direct load path, 55; linear buckling analysis, 18; load, 10, 64; load-resisting, 56, 61–63; mass, 10; movement, 16; stiffness, 10, 16, 31; story stiffness, 17 linked tower structures, 42 live load (LL), 61; code-prescribed, 61; in-service, 62 main wind force resisting systems (MWFRS), 42, 51 mass, 61–62; assumptions, 61 mean recurrence interval (MRI), 2, 5, 13–15, 39, 46; ∼0.1-year, 25; 1-year, 15, 25–26, 60; 5-year, 15; 10-year, 14–15, 23f, 25–26, 60; 25-year, 3, 15, 60; 50-year, 3, 15, 26, 60; 100-year, 3, 15; 300-year, 3; 400-year, 23f; 700-year, 3; 1700-year, 3; accelerations and motion perception, 15; drift and displacement, 14–15; DMI criteria for, recommended, 23f; foundation and lateral system, 14; large, 24; risk categories, 14f;
88
INDEX
ultimate, 3 member end releases, 61 meteorological data sets, 8 modeling assumptions, 3 modulus of elasticity (MOE), 65–66; concrete, 65–66 modulus of rupture, 68t moment frames, 18 monitoring, 78–79 Monte Carlo method, 37 motion predictors, 15 NatHaz Aerodynamic Load Database (NALD), 30 natural design analysis, 7 nonparticipating elements, 60–61 nonstructural elements, 1, 23–25; acceptable drifts for, 23; components and cladding, 24; damage to, 8; façade, 24; interior partitions, 24–25; vertical transportation, 25 Notre Dame University, 30 NRC, 30 occupant comfort, 3, 4, 8, 15, 25–27, 41; accelerations, 9, 15, 25–27; acoustical disturbances, 2, 27; improving, 27; visual disturbances, 2, 27 overturning, 31 P-Delta (second order) effects, 10, 16–17, 31–32, 60, 62; amplification, 17; analysis, 17–18; capturing, 16; stability evaluation, 17–18 panel zone deformations, 32, 60, 64; beam-column joints, 60 partitions, 23, 61; audible creaking, 25; damage to, 22; detailing, 23; interior, 24–25; materials, 25; visual cracking, 25 pedestrians, 5, 47, 79
peer review, 11, 77–78; code, 78; structural, 78 performance-based design (PBD), 79 performance objectives, 13–27; building displacements, 20–23; life-safety, 19; mean recurrence intervals, 13–15; nonstructural elements, 23–25; occupant comfort, 25–27; project-specific performance, 27; serviceability, 7–8; stability requirements, 7, 16–19; strength evaluation of the lateral force-resisting system, 19–20; structural strength requirements, 7 preliminary structural design, 29–33; analysis, 31–32; code-based prescriptive methods, 8; geometric layout of the structure, 8 project-specific performance, 27 quasi-static (background) loads, 45 residential guidelines, 15, 27; 1-year return period, 26 scissor joint model, 66 second-order effects, 16f seismic design, 2, 6, 19, 41 serviceability, 3, 5, 8–9, 13, 21, 54; accelerations and motion perception, 15; drift and displacement, 14–15; -level analysis, 60; wind, 14 slender moment column frames, 17 stability, 7, 16–19, 21; acceptance criteria, 19; coefficient, 18–19; evaluation with P-Delta analysis, 17–18; global, 18–19, 31; P-Delta (second order) effects, 16–17; story, 18–19; story stability coefficient, 17; torsional, 19
INDEX
stakeholders, 3–5, 7, 13, 27; architects, 4; building occupants, 4–5; façade consultant, 4; general public, 5; neighbors, 4; owners, 4; pedestrians, 5; structural engineers, 4; vertical transportation consultant, 4; wind engineering consultants, 4 stiffness, 2; estimating, 31, 59; modifiers, 61, 64; modifiers, selecting, 67–70; reduction, 20, 32, 60, 66; secant stiffness approximation, 69f; serviceability considerations, 59; tension, 70 story drift, 18, 20–22; amount of, 24; ratio, 21f, 22 story stability coefficient, 17 strength design, 1, 13, 21, 54; checks, 32; foundation and lateral system, 14, 19, 32; level analysis, 16, 60 Strouhal number, 72 structural analysis: preliminary, 31–32; ultimate, 41 structural bay, DMI for a, 22f structural modeling and analysis, 10, 59–70; building mass, 61–62; detailed method for selecting stiffness modifiers, 67–70; diaphragms, 62–63; expected strength and modulus of elasticity of concrete materials, 65–66; foundation flexibility, 63; P-Delta (second order) effects, 62; panel zone deformations, 64; primary lateral load-resisting system and nonparticipating elements, 60–61; simplified method for selecting stiffness modifiers, 67; special considerations for reinforced concrete structures, 65–70; stiffness modifiers and behavior of cracked reinforced concrete
89
structures, 66; strength-level and serviceability analysis, 60 superimposed dead load (SDL), 61 surface wind intensity, 2 terrain: channeling, 38, 42; far-field effects, 38; influence of, 38, 73; near-field effects, 38; topographic effects, 38; turbulent wakes, 38 Torre Shyris, 44f twist shape optimization, 73f ultimate limit states, 46 ultimate load: combinations, 59, 62; conditions, 19 USG Corporation, 25 visual disturbances, 2, 27 vortex shedding, 30, 42, 46; dynamic, 72 wind: along-wind response, 29–30; crosswind response, 30, 42; data models, 9; directionality, 5; excitation, 26, 41, 44; nature of, 5–6; preliminary estimates, 29–30; return period, 5; “rose,” 5 wind climate, 71; analysis, 15, 29–30; assessment of, 2, 6, 7, 8–9, 35–40; assessment, site-specific, 9; assessment, upper-level, 9; data sources, 36–37; Davenport wind-loading chain, 35–36; design criteria, 40; directionality, 29, 35; extreme value analysis, 39–40; modeling, 2; storm types, 36–37; strength, 35; terrain, 38 wind design: nonlinear dynamic analysis, 2; performance objectives, 1–2 wind-induced loads and responses, 9–10, 38; published data on, 9–10
90
INDEX
wind-induced response, 2, 35, 45; damping levels for, 53–54; resonant component, 44 wind loads/loading, 4, 21, 29, 41, 61; code-estimated, 29, 41; criteria, 3; design for, 3; determination of, 2; horizontal, 17; modeling, 2; service-level, 31; strength design; requirements for, 3, 58; strengthlevel, 58; ultimate, 14, 19 wind optimization program, 10, 71–75; aerodynamic design, 71; aerodynamic modifications, 71; building geometry, 72–73; building orientation, 71–72; holistic optimization, 73–75; wind speeds, 38; alternative minimum design, 9; definition of, 5; fluctuating component, 5; power spectral density, 5; probability, 3, 8; probability density function, 5; reference, 3; standard deviation, 5; ultimate wind speed maps, 3 wind tunnel tests, 2, 4, 7, 8–10, 29, 31, 33, 41–52, 71; aerodynamic-type model test, 43–44; benefits of, 41; aeroelastic method, 45–46; computational wind engineering, 38, 47; data and climate, 48–49;
data, simulated, 37–38; design evolution, 51; elastic model tests, 42; exclusions, 48; high-frequency balance (HFB) approach, 42–46, 48; high-frequency pressure integration (HFPI), 42, 44, 46, 48; inclusions, 48; laboratory, 37; loads, 32–33; minimum thresholds, 51–52; multisector method, 49; nondirectional, 49; physical, 47; predictions, 15; pressure integration, 45; pressure taps, 44–45; procedure, 47; reduced scale, 38; reports, 38; required input information, 48; rigid pressure-tapped model, 44; sector methods, 49; shielding and influence from surrounding buildings, 51; static (rigid) model tests, 42; storm passage, 49; timeline and type, 47–48; triggers for, 41–42; types of, 42–46; typical outputs, 49–51; ultimate, 1; up-crossing, 49 windstorm types: hurricanes, 37; synoptic low-pressure systems, 36, 39; thunderstorms, 36, 39; tornadoes, 36; topographic or thermally driven winds, 36; tropical cyclones, 36, 37, 39