Iran Conference on the Transfer of Nuclear Technology [2/4]

Citation preview

Iran Conference on the

TRANSFER OF NUCLEAR TECHNOLOGY

Persepolis/Shiraz, Iran April 10-14, 1977

Volume ))

Published by 4 p c r )5 7 e 7 } 9 b c } 9 4 e )u4 p )c e c 8 )} 4 e

Publication Division

CONTENTS OF VOLUME II NUCLEAR REACTOR SAFETY Parallel Sessian

Examples of application and correlation studies for safety rules and standards in nuclear power plants and related considerations W.F. Vinck, H. Maurer, J.L. Lamy, B. Tolley, G. Van Reijen . A brief review of the approach to safety In the nuclear Industry F.R. Farmer

; 3

29

Technology transfers in matters of nuclear safety M. Monte ii, C. Allard

37

Implementing power plant nuclear safety requirements in rapidly industrializing countries D.C. Purdy

48

Current status of reactor safety research in the Japan Atomic Research Institute S. Nozawa, H. Nakamura, K. Taketani

58

Aseismlc design of nuclear power plant In Japan K. Akino

i£

Use of nuclear power plant operating experience V. Stello

rzr

Safety questions involved In nuclear technology transfer W. Marshall

rrc

Developing seismic analysis techniques and requirements for multi-country applications G.J. Bohm, C.W. Lin

r£r

NUCLEAR POWER GENERATION Parallel Session

General Electric's experience in the transfer of nuclear technology P. Cartwright

147

CANDU and the development of local engineering capability R.F.W. Guard

154

Utilization of_plant operating experience In new plant designs: a consultant's viewpoint M.A. Lintner, R. W. Griebe

ror

Operating experience of Shi mane nuclear power station: maintenance of high plant capacity and fuel Integrity M. Miki, C. Nishida, Y. Niki

169

Towards an Argentine nuclear industry R.J. Frydman, 0. Wortman, J.A. Sabato

185

Experience In the transfer of nuclear technology: the Spanish nuclear program A.G. Rodrigues

I

192

Experien£e w ith the construction of the first nuclear power plant in Yugoslavia J. Dular, D. Feretic, Z. Gabrovsek, J. Karuza, M . Copic

210

NU CLEAR SY ST EM S AND A PPLICATIONS Parallel Session

Philippine experience in technology transfer in atomic energy research l.D. lbe, R.J. Palabrica

219

Experience in transfer of control and instrumentation technology to a developing country J. lqleem, J .A. Hashmi

£cc

Development of electronics in India A.S. Rao

£••

The implantation of reactor physics codes: a case of technology transfer R.Y. Hukai, W.J. Os starkamp, M.A.V. Peluso

252

Transfer of radioisotope production technology from the developed to the developing countries M. Rahman

£~•

Technological experience in radioisotope production in Egypt E.A. Hallaba

259

Building and starting up a heavy nuclear components shop 8.0. Hultkrantz

272

ROLE OF EDUCATIONAL INSTITUTIONS Parallel Session

Role of Argonne National Laboratory in the transfer of nuclear technology R.G. Sachs

£ti

Contribution of French universities to the nuclear training of personnel, especially members of developing countries E. Bauer, M. Oria

£D•

German power plant school, its nuclear power plant simulators and its availability for training of engineers of foreign countries 0. Schwarz

czi

The role of foreign specialists in the transfer of technology within Iran P.S. Saeedi

cri

The role of education in nuclear technology transfer: a possible contribution from the JRC S. Finzio, B. Henry

c£•

Nuclear education in the Spanish polytechnical universities: present situation and perspectives for the future M.J. Ortega

cc£

ii

Utilization of training centers for staffing nuclear power plants R. W. Deutsch

342

Training of personnel to design and construct nuclear power plants J.H. Utegaard, F.E. Haskin

352

Nuclear power plant simulator training for developing countries J.M. Roth

367

TRAINING ON THE JOB Parallel Session

Manpower training in developing countries for nuclear power program M. Aslam, 1.U. Rehnam, A.R. Malik

383

Research and education - education and training on the job 0.5. Plail

401

Training on the job of nuclear power plant personnel H.B. Gutmann

408

On the job training during construction and commissioning of nuclear power plants K. Kirchweger, J. Hinterwalder Staff development for plant startup and reactor servicing activities R.J. Walker, B.J. Selig

420

424

Utilization of TAVANIR conventional power plants for training technical staff of future nuclear power plants of Iran M. Vakilian

439

Transferring concepts in nuclear technology with computer ·generated films R.B. Jones, M.A. Tavel

445

A self-consistent approach to monitor and revise procedures of training local skills Z.A. Sabri

467

iii

NUCLEAR REACTOR SAFETY

PARALLEL SESSION

Co-Chairmen: F. Farmer (UK-AEA/England) M. Farzin (AEOl/lran)

7 l 4 r h nQ7 s c 8 4 h h n)5 4 p )c e 4 e 6 5 c } } 7 n4 p )c e s p o 6 )7 s 8c } s 4 87 p b } o n7 s 4 e 6 s p 4 e 6 4 } 6 s )e e o 5 n7 4 } h c v 7 } h n4 e p s 4 e 6 } 7 n4 p 7 6 5 c e s )6 7 } 4 p )c e s

W.F. VINCK With the co-operation of

H. MAURER, J.L. LAMY, B. TOLLEY AND C. VAN RE/JEN Nuclear Safety Section Commission European Communities

P3

)e p } c 6 o 5 p )c e

With an increased number of nuclear power plants and associated equipment to be installed, in countries where the specific designs are originated as well as in countries which import nuclear plants, emphasis must be placed upon a systematic examination of the problems and their possible solutions. One major reason for this is the need for sufficient equivalence in the measures to assure public health and safety. The opinions expressed in the present paper do not necessarily reflect the views of the Commission of European Communities. L3

s c r 7 9 7 e 7 } 4 n 5 c e s )6 7 } 4 p )c e s c e 4 e )e p 7 } e 4 p )c e 4 n r 4 } a 7 p )e e o 5 n7 4 } 5 c e 5 7 h p s 4 e 6 7 i o )h r 7 e p s

Fundamental differences exist in the attitude of the different nations towards nuclear energy, depending in particular on the degree of technical development in nuclear industry. The supplier nations and especially their industries are generally interested in abolishing existing technical barriers that may hinder the further development of and market for nuclear energy. They also wish to have a sound and uniform basis for criteria, standards and guidelines not only to facilitate the export of nuclear reactors and associated equipment but also to be in a better position in any nuclear controversy. The export of nuclear installations is of economic importance as it contributes to the national balance of payments and makes use of the already available total production capacity which may be not fully occupied if the domestic ordering cycle is interrupted by economic crises. Countries not yet fully industrially developed and embarking on nuclear energy programs are primarily interested in becoming independent of oil (possibly coal) and in obtaining cheaper energy by developing their domestic industry not so much to become competitive with the supplier nations but in order to become capable of producing important components and to furnish services to their own nuclear industry. This allows them to reach a degree of independence from supplier nations, which may include uranium enrich-

;

ment and/or spent fuel reprocessing. I may quote here the example of Brazil where German industry conducted contracts for a whole system of nuclear infrastructure beginning with uranium ore exploration and enrichment plants, reprocessing plants, up to nuclear power plants., the installation of component production faci I ities and assistance in siting and component design.

This example

demonstrates of course the conflict of independence in nuclear technology with non-proliferation policy.

The motivation for nations to acquire their~ independent uranium .en-

richment and spent fuel reprocessing faci I ities can only be reduced if fuel cycle services can be provided and guaranteed by supplier nations independently of their own national political developments.

An im portant aspect to which I come back explicitly later in the

paper is the need for technical assistance by exporting countries; not only technical assist-ance in the design of heavy components, but, what is more inportant, technical assistance in safety assessment, quality assurance and periodic inspection in order to guarantee an adequate level of safety in existing power plants during the whole period of lifetime.

The

kind and level of assistance needed is certainly dependent on the degree of development of the existing institutions and may vary from country to country.

In general, such scientific

assistance should include the development of research and education centers, establishment of licensing authorities, exchange of research results concerning safety aspects, especially on quality assurance and periodic inspection techniques. The harmonization of existing criteria, codes and standards is also of a certain im portance from the point of view of educating receiving countries. I would also like to dwell for a moment upon the im portance of the costs of a safe design, which may best be demonstrated by a figure quoted recently by a power plant supplier company.

To accomplish the construction and comm issioning of a nuclear power

plant the vendor needs about 1000 man years.

One may im agine that licensing authorities

and control organizations as well as plant operators may need at least an equivalent effort to accomplish their engagements.

Such figures are im pressive and throw a spot-light on

the im portance of the money which needs to be spent to guarantee safe operation of nuclear power plants.

;3

T'-JE s o - t 7 5 p r 4 pp7 } s Z 6 7 8)e )p )c e s Z 4 5 c r h } 7 d 7 e s )f 7 s o } f 7 b

First of all one must clarify the differences and relationships between a number of notions which are frequently confused with one another. It is clear that with an increased number of nuclear power plants and the associated equipment procurement, a "pragmatic" approach will not suffice and.therefore emphasis is now placed on a more systematic way of attacking the problem. First a parenthesis is made in an effort to clarify the differences and relationships between a number of notions which are frequently intermingled or confused according to the habits (legal, regulatory and industrial within a country) and to the understanding of the terms (language effects). These notions are e.g. legal requirements, rules and regulations, general and de-

S

tailed criteria, industrial standards ("normes" in French), specifications and furthermore notions which might be called "hybrid" such as guides and codes.

All these notions could

be summarized arbitrarily under the general term "standards" (of which part can be considered as "requirements"). Just to give an example of the complexity of the problem of definitions one may refer to the term "specifications" which in one instance may mean the "operational lim its and conditions" under which an operating license is given to a plant, and at another may refer to the detailed equipment procurement conditions ("cahier des charges" in French) fixed between a utility organization and the vendor and construction firms. The substance of these notions can be developed - again depending on the national practices and habits - as regulatory (mandatory) requirements or (non-mandatory) standards and/or as so-called non-mandatory consensus standards (a). Also some countries have a tendency to include at an early stage in legal requirements (e.g. governmental decrees) technical details which may not be proven and are subject to periodic evolution and revision whilst others have intricate systems of developing sim u PF

taneously requirements and standards at various levels with varying degrees of (non-) mandatory character and varying revision provisions. Furthermore, the above requirements and/or standards can have a bearing on the overall plant concept and its siting conditions, on structures (e.g. concrete structures, such as containment) on systems or sub-assemblies (e.g. primary systems, cooling, offgas-systems, ventilation systems, reactor protection systems, power supply systems, etc) or on 1:omponents (pressure vessel, pipes, pumps, valves, electromechanical_ components, electronic components, etc) . And finally these requirements and/or standards may refer simultaneously or separately to design; fabrication and construction; assembly, testing and inspection; periodic inspection and surveillance; normal operation conditions, transient behavior and malfunction or accident conditions. It seems therefore evident that groping one's way through this jungle is a formidable task of which at present one can hardly predict the outcome. Even those who advocate, in the case of imported plant concepts and equipment, flatly using the requirements and standards emanating from the country of origin, will know the difficulties encountered, i.e.: keeping up-to-date with the developments and evolutions of requirements and standards; understanding and interpreting them correctly; correlating them with existing national requirements and standards, to the extent they exist in the receiving country;

(a) standards elaborated on a consensus basis between experts from utilities, vendor and construction firms, and licensing authorities and/or the therewith associated safety and control organizations.

D

including requirements which result from the particularities of the site (e.g. seism ic, population density, neighbouring industrial activities, etc); Technical variances with so-called "reference concept-plants".

Anyway the only way to proceed in a concerted effort seems to be in an orderly and selective manner, certainly not trying to deal with everything at the same time. So the following plan was adopted in the 1st phase and implemented or is in the process of being im plemented. 1. ;

The following were considered perequisites for further work: The setting-up of an up-to-date survey of the national legislations and the adm inistrative licensing procedures applied within E.C . countries. The setting-up of agreed definitions for a number of terms in various languages. The setting-up of an up-to-date and systematically organized collation of the various types of technical requirements and standards (rules) in existence or under development in member countries and in third countries, or on an international basis.

2.

The following areas were chosen for an inventory and inter-correlation of the (nonstandardized) practices, requirements and standards applied for a number of generally postulated accident conditions of external and of internal origin.

Those considered

so far are: a)

External accidents airplane crash (im pact considerations, probabilistic data and criteria). seismic effects. explosions (including fire). flooding.

b)

internal accidents (LWR) loss-of-coolant accident (mechanical and thermo-hydraulic effects in primary system , effects on containment, radiological consequences). fuel handling accident. turbine explosion (including internal fire). loss of power-supply (grid, emergency). steamline break outside containment.

c)

For LMFBR, safety strategies and the various types of accident conditions applied in the analyses of the present prototypes, under construction or operating, are also being compared systematically.

3.

The following priority areas were chosen to find co r-relatio n s amongst general and detailed design and fabrication requirements and/or standards for LWR's: the primary pressure boundary the emergency core cooling system the reactor protection system the power supply systems.

U

4.

Finally some countries have taken the initiative of submitting within the relevant C. E .C . working groups, draft "rules or regulations" for comments from other organizations within the E .C . before finalizing the contents on their own national basis. All member countries have been invited by the Commission to apply a similar procedure. This initiative obtained a more formal character from the Resolution of the Council of the European Communities dated 22 July 1975 which requests the Member States to notify the Comm ission of any draft laws, regulations or provisions of sim ilar scope concerning the safety of nuclear installations in order to enable the appropriate consultations to be held at Community level at the initiative of the Commission. It would seem that on all the above mentioned subjects a certain degree of convergence

will be found while a certain fraction of divergence will be more precisely defined. The sheer fact of going about this in a systematic way with all interested parties will probably itself have a harmonizing effect. It is also possible that subsequently, in some areas, as they become ready for a more uniform approach, common recommendations might be issued on a community basis. Such areas might be for instance: general design criteria for nuclear power plants of the LWR-type; design and fabrication - perhaps inspection - requirements for mechanical systems, structures and components such as pressure vessels, primary piping, containment; the redundancy requirements for vital or emergency electro-mechanical systems; the reliability requirements for the reactor core protection systems; the assumptions and calculation methods for certain postulated accident conditions (e.g. gas cloud explosions, airplane im pact considerations, fuel handling accident analysis, reference LOCA accident analysis assum ptions).

4.

TRANSPOSITION OF RULES AND STANDARDS

A few examples are given of the application of rules and standards of a country of origin of a nuclear design and perhaps of the equipment to a receiving country. The possibilities but also the lim itations and difficulties are shown, with the aid of examples such as containment structures and LOCA-ECCS problems.

Some examples are given based on cases of

specific European nuclear power plants.

4. 1 Containment 4. 1.1

For a specific Belgian PWR nuclear power plant steel and prestressed concrete

primary (inner) containments were compared. Contrary to what might have been expected it emerged, that pres tressed concrete containment has economic advantages over the steel containment if one considers that the free volume of the steel containment is not only dependent on the available energy within the primary circuit (to be considered in case of a LOCA) but also on the wall thickness for which certain obligations are laid down in Euro-

W

pean codes e.g.. steel containments with greater wall thickness have to be heat treated after welding, an operation which has to be avoided for large containment structures. Concrete structures on the contrary can be designed for higher internal pressures with the consequence of a smaller free volume.

As a result of this economic consideration the inner

containment chosen consisted of a complete prestressed internal concrete structure with a steel lined surface, surrounded by a second reinforced concrete shielding building. 'The considerations given to economic aspects did not disregard the aspects of safety. The main safety factors pertaining to prestressing applied to this type of structure are the following: Stresses induced in the wires will be the highest at the moment of pretensioning: due to the various unavoidable losses in the prestressing force, the stresses in the cables will always be less than those existing at this moment of pretensioning and even under reference accident conditions. Moreover, it should be emphasized that in case of an accident, the stress variation in the cables would be negligible: as a matter of fact, internal pressure induces decompression in the concrete which only results in a slight elongation of the prestressing cables.

4. 1. 2

For the same reactors systematic application of design standards in force in the

country of origin (in the present case the US) caused certain difficulties which were able, however, to be solved easily in most cases. We may give as an example for the containment the general Design Criterion No. 56 of the USAEC (10 CFR 50, Appendix Al concerning the isolation of the piping penetrating the containment. This criterion, formulated basically for a single containment, did not apply very well to a double containment. It was thought more logical here to place the valves in series in order to isolate the ventilation pipes in the gap instead of revising the principle applied in the elaboration of the criterion.

Indeed, the valves are protected by the two containments against missiles

caused by accidents of internal or external origin and, in addition it is sufficient for the pipes to be able to withstand an accident pressure up to the level of the outer valve in order to justify not placing either of the two isolation valves inside the prim ary (inner) containment.

4. 2 LOCA - ECCS Problems

Another example is the difficulties encountered in examining a loss-of-coolant accident for a specific reactor in respect of a reference power station of the same type. Before checking the conformity of the calculation results with the US-NRC criteria it ·• ,JS

r.ecessary to make a critical assessment of the validity and the degree of conservatism

inherent in the proposed accident sequence, code descriptors, sensibility analyses. Next, the extent to which the proposed method of accident analysis was applicable to the power station under consideration, by checking the conservatism of the input data used, had to be assessed.

1

To begin with, certain observations were made and the g_aps in the method proposed by the designer were pointed out. The critical points in the analysis were as follows: the question of heat transfer during the pressure relief phase: in transient operation, little was known about the leakage rate for a large breach, the DNB onset time', the heat exchange correlations during post-critical operation or the effect of the system ·on the heat exchange; the question of heat transfer during the reflooding phase: here again the mechanisms were not at all well understood, and there was little information available on the effect of the system on heat exchanges; the behavior of the pumps during pressure relief; the problems associated with the interaction between the steam and the emergency injection water, For all these points, we examined the validity of models tried out during tests in which the effects were separated, Would these models take due account of the accident when applied to the reactor as a whole, when the results of small-scale tests were transposed to a reactor core? A further difficulty cropped up because the effective height of the core was 8 feet, whereas the models developed by the designer and approved by NRC were based on a standard 12-foot core; it was therefore necessary to show that the standard model was applicable in this case, even though other essential parameters, such as the volume and pressure of the accumulators, differed considerably from the data for the standard reactor, Would there still be the same margin of conservatism? The analysis of the final results of the calculations in the case of the worst possible breach showed a net reduction in the maximum permissible hot point factor confronted with the initial design values compatible with the cladding temperature of 2, 200°F imposed by the criteria, Results derived from the August 1974 models gave way to those obtained April 1975, models which took account of the five modifications called for by the NRC. Of these five modifications, the introduction of a delay in the outflow from the accumulators due to the formation of steam by the hot walls in the downcomer and the revision of the pressure losses at the place where the emergency cooling water injection main is tapped into the loop were the main cause of the reduction in the permissible hot point factor or, to put it another way, the increase in cladding temperature in the event of an accident, This cladding temperature rise reduced commensurately the still present margin of uncertainty, which was thus brought to its simplest expression in the case of a brand-new core, Al I these factors convinced the experts of the difficulty, not to say impossibility, of applying at that stage a margin of error to the results and prompted them to formulate reservations on the degree of conservatism in the calculations, On the practical level, these reservations have found expressions in recommendations aimed essentially at ensuring that the reactor could at no time in its life exceed the limits laid down for the hot point factor, This imposed severe constraints on the operators as regards the reactor control made and the use of the various types of control rod,

2

5.

.C ORRELATIONS BETWEEN PRACTICES, RULES AND STANDARDS OF VARIOUS COUNTRIES

5. 1 Such Correlations Have to be Dealt with in an Orderly and Selective Manner. The Substratum for the Performance of such Correlations (e.g. within the E.C.) is: 5.1.1 An inventory of the "vehicle" on which, in each country, technical health and safety matters ride: i.e. national legislations and the administration I icensing procedures

applied (inventory now available e.g. as report EUR 5284e for the E.C., as report EUR 5525e for certain non-member states of the E .C.). The licensing procedures in the various countries have developed, and are continuing to develop, along different lines, depending on: the political and administrative structures and the laws applicable in each country; the organizational characteristics of each country and its regions. These procedures are a channel for the development and/or application of the technical practices and methods concerning nuclear safety as well as the relevant safety requirements and "rules". Annexed to this paper are diagrams ii lustrating the licensing procedures applicable in two countries - the Federal Republic of Germany and France (Appendix 1 and 2). These two countries have been selected because of their very different political and administrative structures, the one being a federal state and the other a centralized state.

5. 1. 2

The setting-up of agreed definitions for a number of terms in various languages.

In recent years various national organizations and institutions were elaborating rules in the nuclear field but there were no internationally or multinationally recognized definitions of guides, criteria, safety rules, codes, standards, specifications, etc., to indicate the extent to which a rule is mandatory or to provide an order of priority in application. The definitions drafted in 1973 by an ad hoc working party of the CEC working group on "methodology, codes and standards" gives a certain order of dependence of the indications and of the mandatory nature of the rules. The following.definitions have subsequently also been proposed to ISO/TC 85/SC 3 "Power reactor technology". a) Regulation bl

A mandatory direction laid down by an authority legally vested with such power. Criteria A .set of principles and data on which judgement or decision may be based.

Note: Criteria may or may not be mandatory. c)

Guide A collection of recommendations not mandatory which outline a method of achieving

an objective.

10

( R s& " _( " ! ( 4 & k­ >_0 ­ " { ( kj ­ ! 0k ­ +_j k_j / j +! ' k_k!" · " • 0&( k_ k!:+!( k! ·0 ­ >k_ - k:/ ' _0j j k_ " / j ' kj & "& & k& k_ - k>+!( k k! ·" j j k_ ? 0 !( 3

xR

a!0 & k! 0k_ N5 ! 0 & k! 0" R 70 _k } k0>k ° +_ 9 !/ _( j " & €k_ / _( c " & k_Z " _>" _( ( k!k! o ! & k0 ·k ' k:" ··&+( k! - kj ­ >·/ j j k ' k:" j j &? k!( k_3 4 _• k! • / _' K a ! 0& k!0 k_ • +__k_ €? 0_' k_( +( k! _0­ >& €? 0 _' k_( j k0_3

­R

}0 ­ >& ·0 _0 k s " • • ·/ _' _0­ >& €? 0 _' k_( k! 7• ·/ _' k_ €/ ! - kj ­ >! k0x/ _' k0_k! r k& >+( k :/ ! k0_ xkj & 0• • & kj u0 k·3

(R

e +!• 70 _k ( k! c ::k_& ·0 ­ >• k0&€/ ' " _' ·0 ­ >k &k­ >0j ­ >k - kj ­ >!k0x/ _' +( k! 6 k:0_0 & 0+_Z ( 0 k 0_ u/ j " • • k_" ! xk0 &/ _( 0•

70 _° k! _k>• k_ • 0&" ·0 k_ xk& k0F·0 ' & k_ h " !& k0k_ +( k! • 0 &0>! k!

' k_k!k··k_ u/ j & • 0• • / _' :kj & ' k·k' &/ _( ° +_ k0_k! " _k! • " __& k_ s & k··k ' k_k>• 0 ' & ? +!( k> 0j & 3 kR s !k_ ' k_/ ' k_ • / j j Z ' k' kxk_k_:" ··j k0 _j ­ >·0 kj j ·0 ­ > ( k! d k!j & k··/ _' j F/ _( a +_& !+··• k& >+( k_3 6 " kj 0•

f k!k0_0 ' & k_ a +_0' ! k0­ > • k0_k ' k_" • k 7 _& j / _' ( kj 4 / j ( !/ ­ • j Aa +( kA

' 0x&F ( k! s 0 __' k>" ·&k_& j &( k! - k!k0­ >_/ _' Af +!j ­ >! 0:& A +( k! ( k! - k!k0­ >_/ _' Ae +!• A F j +··& k ( k! 4 / j ( ! / ­ • Aa +( kA 0•

0_& k!_" & 0+_" ·k_ u/ j " • • k_>" _' • k0_k f k!? k_-

( / _' :0_( k_3 0_ 8!k_­ > "{

} k' ·k• k_& "& 0+_ 60 ! k­ & 0° k ( k ­ " ! " ­ & k!k +x·0 ' "& +0 ! k " ! ! k& kk _ ccn s t r uc t Lon crcit:··• ~ >/* ial ocr;:.1tJ ~>p~rat103 n0 c e n j kR

---~-,•-

~ ± ± ±

-- - - -

(cntinuoua op erati.o o

license) e\K;G%V°V>\

PPF YF &F 0F F

PF ! FF F & F !F & F

)

Hbi,_tc;-~1-

=- -

J.uthorities To Pc

Trial

dF FB, FFFFY ) 3

Inter,or

Jte.1ctoc Safety Coc-ai s s i on

33•••3

XEdFJ

y

L1ccnarn:, Aut~orlty-

'

!

qqqqq )

Pr~;,,rative \,;1Jr\c Not trxc:_,vativn fP.uitdings. - r~~.;•.1irin& a License - (~,r1~.\ .• n3Z · · AAAE -•~.fys.te:::s , · -· _....,_ ~- --- - - - -

C..1

n

J

y

J

& FFFE &F FB,I FFFFYFSF IF I I

In+se rv i e c B_E°K;;*VrL_°

t 3F 3F x FnF & F tF B, FFnFF

Supervi s i o .. 1

I 41

t

t

53( ·Z·Z±3t F F F·E FF F F F B, F·FF F F FF F F F F F ·F F F F

!

F F FI,

Ti,ue

_ :, J:, C X

±

4 h h 7 e 6 )l

L

Rbv-JF dvIPF mR hbmJF• MbFe vKKXV;H*V>_ °\wYV**kW *> UV_V°*/µ / B_W\°*/µ H_W bk°kH/;· ± hH°°kW *> >*·k/ UV_V°*/Vk° concerned / V_/YH*V>_ H_W ;>YYk_*EJ>YYk_;kYk_* >< X>;HX k_j\V/µ 5V< _k;k°°H/µ7Eh/kXVYV_H/µ °H/* presented w4LE *·k e*H_WV_Q J>YYV**kkE v_HXµ°V° >< K/kXVYV_H/µ °H/* wµ *·k JFvE F+HYV_H*V>_ wµ *·k e*H_WV_Q J>YYV**kkE mK VE_V >_ H_W /k;>YYk_W H* V>_°E • /H%HX wµ *·k UV_2V°*/µ / NkHX*·E h\wXV;H*V>_ >< *·k 2H\*>/V°H*V>_ Wk ;/kH*V>_2

8

oTr£ Y>_*·°

.8.TTTTTTTTTTTTTTTT4

r£Tri Y>_*·° 5°>Yk*VYk° X>_Qk/7

I

rJ>_°*/\7*V>_ TTTTTTTTT~ G >< °*H*V>_ .J, 6 • ­ _& ·° wk/k h/>%V°>_HX safety /kK>/* °\wYV**kW *> UV_V°*/µ / B_W\°*/µ H_W bk°kH/;· X>HWV_Q v_HXEµ°V° >< K/>%V°>4r4_HX °H/* wµ4 *·k JFvE F+HYV_H*V>_ wµ e*HXWV_Q J>YYV**kk. >KV_V>_ H_W /k;>YYk_WH*V>_°E vKK/>%HX wµ *·k UV_V°*/µ P>HWV_Q >< FF F F F F FF F➔Z /kH;*>/ E. h>$k/ /\_T\K , Z *k°*° 4 J, ->* XH*k/ *·H_ U RV_HX °H/* °\wYV**kW *> UV_V°*/µ for Y>_*·° wk/k B_W\°*/µ H_W bk°kH/;· k+KV/H*V>_ >< *·k l WkHWXV_k °k* wµ *·k v_HXµ°V° >< T E E i . , /V°H*V>_ Wk F+HYV_H*V>_ wµ *·k e*H_WV_Q J>YYV**kk. >KV_V>_ H_W /k;>YYk_WH*V>_°E ;/kH*V>_2 ZP3 • kHWXV_k °k* wµ UV_V°*k/VHX HKK/>%HX / °*H/* >< />\*V_k *·k m/Wk/ / >Kk/H*V>_E 2H\*>/V°H*V>_ Wk ;/kH*V>_2E

t

r-----

LS

'3

4 h h 7 e 6 )l

;

JXH°°V_ eµ°*kY B__k+ >< Ji*kQ>/Vk°

rEc

dk/Y° H/EW JXH°°V_ dk/Y° V_ -\;XkH/ e;Vk_;k H_W dk;·_>X>Qµ -\;XkH/ bkH;*>/ JXH°°V_ eµ°*kY° J>YKVXH*V>/~ H__ PV°*° >< -\;XkH/ e*H_WH/>°

£E £Er £E£

Ak_k/HX XXVGXk° v;*°. bkQ\XH*V>_° A\VWk°n J/V*k/VH

3.

eV*k Ak_k/HX eV*k Ak>X>QV; H_W ekV°YV; • k°VQ_ Uk*k>/>X>Qµ an:, NµW/>X>Qµ

1. rEr

rE£

~Er cE£ 3.3 •E •Er 4,2

•Ec 5.

5.1 ~E£ 5-.5

~E•

~E~ oE

oEr oE£ o • E~ 6.4

oE~ oEo

bHWV>H;*V%k fH°*k°n h>XX\*V>_ H_W J>_*/>X bHWV>H;*V%k fH°*k° fH*k/ h>XX\*V>_ H_W J>_*/>X. UkH°\/kYk_* vV/ h>XX\*V>_ H_W J>_*/>X bHWVH*V>_Th/>*k;*V>_ Ak_k/HX IV>X>QV;HX e·VkXWV_Q F_%V/>_Yk_*HX H_W F°VYk*/µ hk/°>__kX U>_V*>/V_Q vWYV_V°*/H*V>_. mKk/H*V>_. hXH_* h/>*k;*V>_ vWYV_°*/H*V%k J>_*/>X mKk/H*V>_ d/H_°K>/*H*V>_ H_W hH;&HQV_Q / bHs4XV>H;*V%k Materials a\HXV_ H_W ekXk;*V>_ >< hk./°>__kX B_W\°*/VHX ek;\/V*µ RV/k h/>*k;*V>_

25

4 h h 7 e 6 )l

tE

tEr tE£ 7-3 tE•

;

N5 +_& 0_/ k ( R

eHXH*V>_

APPENDIX 3

(Continued)

12. 12.1 12.2 12.3 12.4

bkH;*>/ h/>*k;*V>_ h/>*k;*V>_ eµ°*kY bkH;*>/ B_°*/\Yk_*H*V>_ vXH/YTeµ°*kY° B_;VWk_* H_W h>°*Tv;;VWk_* U>_V*>/V_Q

13.

NkH* bkY>%HX. FYk/Qk_;µ J>/k J>>XV_Q

14.

yk_*VXH*V_Q H_W vV/ JXkH_V_Q Systems A\VWk°. eHYKXV_Q. eµ°*kY° dk°*V_Q >< eµ°*kY° RVX*k/° H_W vw°>/wk/°

14.1

14.2 14.

3

15. 15.1

15.2 15.3 15.4

15. 5 15.6 16. 16.1

16.2 16.3

17. 17 .1 17.2

17._; 17.4 17. 5 17.6

NH_WXV_Q. e*>/HQk. h/>;k°°V_Q >< R\kX H_W bHWV>H;*V%k /TBH*k/VHX° J/V*V;HXV*µ eH\*°VWk bkH;*a/ PV/HQk RH;VXV*µ h/>;k°°V_Q UH*k/VHX J>_*/>X. B_%k_*>/µ Uk*·>W° / v_HXµ°V° FXk;*/V; Fj\VKYk_* H_W eµ°*kY° FXk;*/V; eµ°*kY° h>$k/ e>\/;k° B_°\XH*V>_ H_W >*·k/ Rules dk°*V_Qn B_°Kk;*V>_. Surveillance Ak_k/HX h/k°°\/k yk°°kX° H_W I>VXk/° J>YK>_k_*° and J>>XH_* J>_*HV_Yk_* H_W B_°*HXXH*V>_° B_Tek/%V;k e\/%kVXXH_;k H_W B_°KkE;*V>_ ~ rEk;*/X; Fj\VKYk_*

27

4 h h 7 e 6 )l

;

N5 +_& 0_/ k ( R

riE riEr riE£ riEc

UH*k/VHX e*H_WH/W° B/>_ H_W e*kkX ->_TRk//V; Uk*HX° H_W >*·k/ Materials JH°*V_Q°. R>/QV_Q°. J>YK>·k_*°

rDE rDEr rDErEr rDErE£ rDErEc rDE£ rDE£Er rDE£E£

UH*k/VHX dk°* Uk*·>W° ->_Wk°*/\;*V%k dk°* UkT*·>W° UHQ_k*V; hH/*V;Xk B_°Kk;*V>_ Uk*·>W bH>V>Q/HK·µTUk*·>W° MX*/HTe>_V; Uk*·>W° m*·k/ dk°* Uk*·>W° Uk*HX dk°* Uk*·>W° m*·k/ dk°* e*H_WH/W°

£i

4 - } )7 8 } 7 f )7 v c 8 p d 7 4 h h } c 4 5 d p c s 4 8 7 p b )e p d 7 e o 5 n7 4 } )e 6 o s p } b

FRANK REGINALD FARMER, 0.8.E. Safety Advisor to the Authority United Kingdom Atomic Energy Authority

ABSTRACT p >0j 0_• 0_'

0_ " & +• 0 ­ k_k ! ' ¢3 s " :k& ¢ ­ +_j 0( k ! -

"& 0+_j ( +• 0_" & k( & >k k" ! ·¢ ( k ° k ·+k ! k­ +! ( +: &>k :0! j &9 k _k° " 5 +_:k! k _­ k +_ &>k h k " ­ k:/ · o j k j +: 4 & +• 0­ 7 _k ! ' ¢ k ! k >" ° k x k k_ • " _¢ " & &k• +j k :+! ­ +_& 0_/ +/ j +! ·+? ·k ° k · k\ " ° k x kk_ ! k" j +_" x ·¢ ? k ·· " ' ! k k( 0_& k ! _" & 0+_" ··¢Q &>+j k !k ·" & 0_' &+ " ­ ­ 0( k _& j &+ _/ ­ ·k " ! 0_j & " ··" & 0+_j +! !0 j • j 0& / "& 0+_j >" ° k j & 0 ·· &+ xk ! kj +·° k( 3 f " ! 0+/ j ­ ! 0& k ! 0" >" ° k xk k _ k j k 3 )&? 0 ·· k_­ +/ ! " ' k " ! " & 0+_" · " k ! k 0j j +• k j • " ·· ! 0j • +: " ­ ­ 0( k _& j 0_ • +j & 0_( / j & ! 0" · " _( j +­ 0" · " ­ & 0° 0& 0k j Q " & +• 0­ k_k ! ' ¢ 0j _+&/ _/ j / " · 0_ &>0j ! kj k ! 0_( / j &! 0k j &>" _ >" j xk k_ " ··k ' k( x¢ &>+j k ? >+ ° 0k ? &>k _/ ­ ·k " ! 0_( / j & ! ¢ 0_ 0j +·" & 0+_3

)_ &>0 j >" ° k 0_:·/ k _­ k( &>k ( k ° k ·+k ! k >" ° k x kk _ • " _¢ " & &k• +j k :+! ­ +_& 0_/ +/ j +! ·+? ·k ° k · k\ " ° k x k k_ ! k" j +_" x ·¢ ? k ·· " ' ! kk( 0_& k! _" & 0+_" ··¢Z &>+j k ! k ·" & 0_'

&+ " ­ ­ 0( k _& j &+ _/ ­ ·k " ! 0_j & " ··" & 0+_j >" ° k j & 0)) &+ x k ! kj +·° k( 3

p >! kk &>k • k j >" ( xkk _ ? k ·· kj & " x ·0 j >k( x¢ P2DDZ " _( &>kj k ) 0_& ! +( / ­ k =/ 0& k x ! 0k :·¢K N" R Ac _k +: &>k ­ / ! ! k_&( 0::0­ / ·& 0k j 0_ k° " ·/ " & 0_'

! k" ­ &+! >" €" ! ( j 0j & >0 j ·" ­ • +: k\ ! k" ­ &+! " ­ ­ 0( k _& j A N·R 3 A4 ·· +& >k ! k_' 0_kk ! 0_' &k­ >_+·+' 0k j >" ° k " ( ° " _­ k( _+&+_ &>k x " j 0j +: &>k 0! j / ­ ­ kj j kj Z x / &+_ & >k x" j 0 j +: &>k 0! :" 0 ·/ ! k j 33 3 4& +• 0­ k_k ! ' ¢Z >+? k ° k ! Z • / j &:+! k' + &>0j " ( ° " _L &" ' k +: k x" j 0 j +: • _+? ·k ( ' k ' " 0 _k( x¢ :" 0 ·/ ! k j A N R 3 p >0j _k? j 0& / "& 0+_ F &+ 0 j ­ +• k / j k +: ( 0 ::k! k_&­ ! 0& k! 0" :+! ­ >k • 0­ " · " _( ! " ( 0+" ­ & 0° k

£D

poisons, and I will refer to it later. (c) The potential for harm to people and property is very great for nuclear installations. Parker(J) estimates a full release from a 1,000 MW(th) reactor could kill 200-500 people in a region of population density 200-500 per square mile. WASH-740 estimates that there could be a remote possibility of killing 3,400 people(4 l The proposition that we have no experience of accidents and we cannot afford to have any in view of the unusual hazard has played an important part in the development of the engineering, operation and regulation of atomic energy and is, in part, the cause of the nuclear debate today. Reverting now to the theme of accident anticipation, it is quite significant that in 1957 Hinton drew attention to these new conditions under which atomic energy was to be developed. The ideas expressed reflected the mood among the leaders of atomic energy at a time when reactor safety analysis and reactor safety criteria were in a formative stage. Hinton, in his Axel Ax: Johnson lecture in Stockholm, writes: "All other e_ngineering technologies have advanced not on the basis of their successes, but on the basis of their failures. The bridges that have collapsed under load have added more to our knowledge of bridge design than the bridges which have been successful; the boilers which have blown up have added more to our knowledge of boiler design than those which have been free from accident, and the turbo-alternator rotors which have failed have taught us more of turbo-alternator design than those which have continued in satisfactory operation. Atomic energy, however, must forego this advantage of progressing on the basis of knowledge gained by failures". Faced with this seemingly new situation, various ideas were put forward as aids to safety, such as safety by siting, by containment, by the consideration of and protection from the maximum credible accident.

It seemed, on reflection, that we sought safety by

rule, longing for a deterministic solution as if by proceeding one way we would be safe some other way unsafe. This is well illustrated by our search for siting criteria in the UK. Sites for the first phase of the British nuclear power program were chosen against population limits stemming from the Marley and Fry accident analysis, Geneva, 1955; namely that in any 10° sector there should be less than 500 people within

H

miles, less

than 10,000 people within 5 miles, or less than 100,000 people within 1 c miles (5 ). The difficulties in the use of precise limits became more and more apparent and this led to changes which were first introduced i~ an IAEA panel discussion in November 1961 (5 ) 7 and at a CNEN Conference in 1962( ). In the following years the UK adopted a graduate.d measure of assessing populatiop devised to find sites relatively sparsely populated near the reactor. Meanwhile, in the USA from 1959 onwards, siting criteria were based on dose limits to the population which were assessed as a result of a maximum credible accident. The

30

calculation model was prescribed and included the benefit of containment having a leak rate of one part per one thousana per day and assumed the release of fifty per cent of the volatile fission products from the fuel. These assumptions seemed largely independent of the real state of the reactor or its containment and supported the belief that safety was achieved by siting. During this period in the development of sarety ideas, the concept of a maximum credible accident played a major part, even if in a somewhat confused fashion N- R K

"Thus, the maximum credible accident is defined as the upper limit of hazard, I.e. fission product release, against which the features of the site must be compared". There was a move to bring reactors closer to population; this required an amendment of the maximum credible accident to become more realistic and required more confidence in the functioning of safety equipment: PP • • • consequences-limiting safeguards~ have a very high degree of dependability". "In summary, 'high level dependabil!ty' in safeguards means systems designed with unreserved commitment to the concept that their effective performance at any time, under any conditions, when they may be called on, is a matter of life and death". There is, at this stage, a move away from the simple concept of "safe" and "unsafe"; a slow move into the recognition of risk or conversely the recognition of the need for a high degree of dependability or reliability, but still a great reluctance to be in any way specific in regard to the reliability criteria or the acceptance of risk. During the ten-year period 1955-1965 we, in the UK, were also seeking absolute standards for reactor safety starting with the requirement that the Magnox reactor should survive a double duct fracture. A later realization that this may not be the worst accident led us to assume complete flow stagnation for thirty seconds - a critical period for temperature equalization in fuel and can - and to assume remedial measures then to come into operation. Various ideas were explored', such as the search for inherent safety - a requirement that any perturbation imposed on the reactor would lead it to take up a safer position. This we found an impossible target - even large negative power temperature coefficients can be dangerous. For a time in our accident analysis we used the device of tracing the effect of two coincidental faults - but not three. We soon found the two coincident fault criteria nad little merit; single faults could range from the trivial to the catastrophic, and for any pair the logic of their coincidence could be questioned. By 1964 we were moving slowly and hopefully into the development of realistic safety criteria. Part of the realism led to the acceptance of some risk: "The objective of the temperature control criterion is to avoid widespread can melting duri!"g the temperature transient following bottom duct failure. It is assumed acceptable to have some low probability of ignition of a single channel, or small number of channels ... (9)". I have quoted various statements which have referred to risk - in general they have

31

shed little light on the problem and often revert to a circular argument, as for example in 1966, in Part 100 of its Regulations, "AEC has in effect defined undue risk by establishing general guidelines for evaluating the safety of reactor sites" - by implication, it is safe if it is sited. We have a problem with the word 'safe'. The safety of a plant - or its hazard depends on the framework within which it is presented. A discussion of safety in atomic energy arouses weighty emotion - let me illustrate the difficulty in using the word 'safe' in a nonnuclear industry. Consider the record of vapour explosion from petroleum products, as reviewed by various writers. . (1 O) There have been 108 known cases of vapour explosion over forty years : Up to 1950

Damage averaged less than d r per year with the exception of the accident in Cleveland when an explosion of liquid natural gas in 1944 led to 136 dead, 77 missing, and a cost of several million dollars.

1950-1964

Damage averaged less than $1M per year and since then there has been an increasing number of events and it would appear in recent times that the cost is running as high as several tens of mill ions of dollars per year. Failure of a 10 inch valve led to the release of 4,000 gallons of isobutylene;

1967

the vapour cloud explosion killed 7, with damage estimated at $35M. 1968

Accident at Pernis cost $46M.

1959

Freight car accident - liquid petroleum gas killed 23.

1962

7,000 gallon truck led to a vapour explosion - killed 10.

1966

Propane explosion killed 17, whereas 10 tank cars in Illinois caused major destruction but did not kill anyone.

Do I conclude the oil refineries are safe or not safe? They supply a need. I estimate 2 the risk of a major accident to any large refinery site to be around 1 o- per year, but the 4 chance that I might be killed living near a site to be far less than 10- per year (these estimates are illustrative). This is a risk level to which I am already exposed in many occupations and social activities. The UK average for the working population is 5 x 10-S per year, and accidents at home are very age dependent, but over many years the average is also 5 x ,o-5 per year. Hence I might well conclude that the petrochemical industry is just about safe. However, the assumptions made in accident model I ing and assessment are vital to this issue. On 2 June 1970 in Illinois, 10 tank cars of 33,000 gallon capacity were derailed; there was a vapour cloud explosion starting fires and subsequent explosions that destroyed the business section of the town, but there were no fatalitjes. Tl'lis was du~ to the sequential explosion of the tank cars; had there been a single explosion of nearly 1,000 tonnes of propane, it is possible that 1,000 people or more could have been killed. To estimate in advance the result of explosions or escape of toxic gas is very difficult,

c£

and - as with the assessment of nuclear accidents - the result depends on the way in which the accident develops; the population distribution; the wind and weather; local topography, etc. Accidents involving explosive substances have, on an average, killed about 1 person per tonne of explos-ive, but it is possible that the result could be lower or higher by at least a factor of ten. Similar numbers might apply to the release of toxic gases, with the reservation that the lethal range could be much greater, but might on some occasions be compensated by evacuation. Comparisons between nuclear accidents and toxic gases were made by Marley and Fry (S)

in 1955 leading to a rough equivalence. 100 tonnes Cl 10 tonnes phosgene 100 MW (th) Today we would assume that a reactor accident might lead to the release of the gaseous and volatile fission products. I make the further assumption that the release often per cent of the volatiles - iodine, tellurium, caesium - would, under most circumstances, give rise to the most severe consequences. A larger release would almost certainly be associated with a thermal uplift from the self-heating of the cloud which would reduce the down-wind concentration (as referred to in the Report of the Royal Commission (l l)). With these provisos, I would now give a rough equivalence of: 100 tonnes Cl 10 tonnes phosgene 15,000 MW (th) It is generally supposed that radioactive poisons are very much more hazardous than chemical. McCullough(l) gives a factor of 106-109. The basis for comparison is important, as with the model for accident analysis. If we consider that quantity of "poison" which has a >SO per cent chance of causing death, then we find a ratio of 10: 1 by weight for Cl: Pu. If we consider the dose which might - on a linear hypotheses - give a O. 1 per cent chance of death after many years, then we arrive at a lower figure for Pu, by a factor of 1,000. We should not forget that there is a 1 per cent probability of accidental death in coal mining and O. 3 per cent as an average of all industrial activities. Comparing different "poisons" or the hazard of different industries is difficult, but several attempts have been made to assess the relative harm of the different ways of producing electricity. Sir Edward Poch in has assessed the population exposure from nuclear power production and other radiation sources (l 1). From radiation exposure to those at work and the public from the complete nuclear cycle, Including also occupational risk of death from accidents, he concludes that the fatality rate might be O. 8 per 1,000 MW (el y. Of this figure 80 per cent arises in construction and mining; the contribution from reactors and reprocessing plants is trivial. He refers to other studies assessing the health hazards from coal, oil or gas power networks but finds a wide disparity in the effect of air pollution at distances of up to 8,0 km. Whereas radiation damage is assumed to be linearly related to dose, no such basis exists on which to estimate the effect of low concentrations of sulphur-dioxide, polycyclic hydrocarbons, nitrogen-oxides, etc Poch in does discuss the alleged difference in type of harm between radiation and other hazards: "The difference in type of harm with the induction of malignant disease and of genetic

J

J

J

J

33

injury as compared with accidental deaths and injuries, is understandably thought of as giving a worse character to the hazards of nuclear as compared with other forms of power production, and this increases the need for proper perspective as to the frequency with which these effects might occur.

In fact, however, the induc-

tion of malignant disease by chemical factors in the working environment is, unfortunately, becoming recognized in a number of industries or production processes, and carcinogenic chemicals are commonly also mutagenic.

Indeed, the risks of

fatal malignancies involved in some forms of industrial exposure to chemical agents have been shown to involve up to 10.to 30 deaths per 10,000 workers per year, as compared with a probably maximum estimate of less than 1 per 10,000 per year for industries involving radiation exposure". In this discourse, I am not attempting to justify risk at any level nor to justify one risk by comparison with another of different type.

I seek rather to clarify the nature of

the risk we run in an industrial society - a society in which the risk of accidental death decreases annually and in which life expectancy increases annually. There is a small risk of serious accidents which we should endeavour to anticipate by paying greater attention to the more frequent and lesser events which occur.

It might be thought that the amount

of effort employed to prevent accidents should be related to their predicted severity and inversely to their predicted frequency. However, the world is not such an orderly place and we have only limited skill in prediction.

I do, however, regret the excessive attention

sometimes paid to highly improbable accidents; this diverts effort into long-term research and into difficult and expensive engineered devices which may or may not offer the advantage originally sought. Through my association with the assessment of hazards in nuclear and non-nuclear industries, I have found the need to have some target or framework with which to describe the hazard.

In 1967, I put forward a proposal at an IAEA Conference suggesting

limiting criteria having an inverse relationship between frequency and severity. I suggested that the nuclear industry might explore the possibility of designing to a target 3 such that the release of 10 curies of 1-131 and other associated gases and volatiles might Tc o be not more frequent than 10 per reactor per year and the release of 10 curies should 7 6 be less frequent than 10- to ,o- per reactor per year{l3) There has been considerable argument, often not very profitable, about the regression index and there have been other suggestions even that the emergency reference level {ERL) Tt should not be exceeded on a frequency of 1 O per year. As an exposure to an ERL is not fatal - in fact the ERL is only a precautionary index carrying a risk, for the most part, of 4 7 less than 10- per year of later incidence of harm - then a target of 10- per year for an. 11 ERL implies a risk per person of less than ,o- per year, far less than the risk from lightning, meteorites, black holes or other cosmic events. Similarly, I regret the apparent rigidity of safety analysis as exemplified by the recommended pattern of reactor safety submissions. In the US applicants are required to show 34

that reactors can withstand eight classes of accident applying certain assumptions on fission product release rate, emergency core cooling, and containment capability. This procedure does not make room for accidents beyond Class 1 which are relegated to the incredible Class 9. I assume that the reactor is deemed safe enough if it meets the relevant criteria. although the fission product release currently assessed for the design basis accident (WASH-1250) (l4 ) would only carry a risk of 1 in 100 of causing 1 fatality in the 10-20 years after the major accident - hardly an event requiring such an elaborate approval procedure. The more recent publication of the Rasmussen report (WASH-1400) (l5 l has considerably extended the range of accident study and has ascribed probabilities of occurrence to the conseqvences of the more important accidents. The report has been widely debated, has received favourable and unfavourable comment, but is accepted by most people as making an important contribution to reactor safety. Tnere are clear signs that ideas are changing. Even Class 9 accidents are now studied and it appears from WASH-1270 (l 5) that the US Regulatory Authority may be moving toward a new criterion, namely: "The safety objective is that the I ikelihood of al I accidents with significant consequences not included in the design basis envelope should not be greater than one chance in one ,o-6 per year".

million per year, i.e .• should not o.ccur with a failure rate greater than This-is entirely in line with my own thinking and thatofVinck(l7 ):

6 "I do not believe target value lower than 10- events/reactor-year are practical, nor are they necessary. taking account for instance of the severity with which the nuclear activities are handled". Finally, I wish to encourage a rational approach in which we recognize the risk of possible hurt to people through accidents in atomic energy installations and also in many other industrial and social activities and that we share our experience and our efforts to reduce risk.

REFERENCES 1.

McCullough, C.R. and Others. "The Safety of Nuclear Reactors", Proceedings of International Conference on the Peaceful Uses of Atomic Energy, Geneva 1955, p. 79, Vol. 13.

2.

Hinton, C. "The Future for Nuclear Power" Axel Ax: Son Johnson Lecture, Stockholm. 15 March 1957.

3.

Parker, H. M. and Healey. J. W. "Environmental Effects of a Major Reactor Disaster". Proceedings of International Conference on the Peaceful Uses of Atomic Energy. Geneva 1955, P. 106, Vol. 13.

4.

WASH-740 "Theoretical Possibilities and Consequences of Major Accidents in Large Nuclear Power Plants" USAEC, March 195i.

5.

Marley, W .G. and Fry, T .M. "Radiological Hazards from an Escape of Fission 35

Products and the Implications in Power Reactor Location", Proceedings of International Conference on the Peaceful Uses of Atomic Energy, Geneva 1955, Vol. 13, pp. 102-105. 6.

Proceedings of IAEA/ISO Panel on the safe Siting of Nuclear Reactors, Vienna,

7.

Farmer, F.R. "The Evaluation of Power Reactor Sites", Proceedings ofCNEN Confer-

31 Oct - 3 Nov, 1961, pp. 7-10. ence Problemi di Sicurezza degli lmpianti Nucleari, VII Congresso Nucleare, Rome, 11-17 June 1962, pp. 39-45. 8.

Beck, C.F. "Engineering Out the Distance Factor - A Progress Report on Reactor Site Criteria", Annual Convention of the Federal Bar Association, Philadelphia, Pennsylvania, USA, 25 September 1963.

9.

Hughes, H.A. and Horsley, R.M. "Application of Safety Limitation against Depressurization to the Calder and Chapelcross Reactors", Proceedings of BNES Symposium on the Safety of Magnox Reactors, London, 11 November 1964, pp. 198-203.

10. Strehlow, R.A. (1973b) "Unconfined Vapor-Cloud Explosions - An Overview". 14th Symposium (international) on Combustion. The Combustion Institute, 1189-1200. 11. Royal Commission on Environmental Pollution, Sixth Report, Nuclear Power and

the Environment, September, 1976, Chapter VI, para. 263, Cmnd 6618, HMSO(UK). 12. Pochin, E.E. "Estimated Population Exposure from Nuclear Power Production and other Radiation Sources", NEA/OECD, January 1976, p. 37. 13. Farmer, R.R. "Siting Criteria - A New Approach" IAEA Symposium on Containment and Siting, Vienna, 3/7 April 1967, SM-89/34. 14. WASH-1250, "Safety of Nuclear Power Reactors (Light-Water Cooled) and Related Facilities" USAEC, July 1973. 15. WASH-1400, "Reactor Safety Study - An Assessment of Accident Risks in US Commercial Nuclear Power Plants". USNRC, October 1975. 16. WASH-1270, "Technical Report on Anticipated Transients without Scram for WaterCooled Power Reactors", USAEC, September 1973. 17. Vinck, W. and Reijen, G. Van. "Quantitative Risk Assessment, the Promised but not the Sacred", ANS/ENS International Conference on World Nuclear Power, . Washington D.C., 14-19 November 1976.

36

p 7 5 d e c nc 9 b p } 4 e s 8 7 } s )e r 4 p p 7 } s c 8 e o 5 n7 4 } s 4 8 7 p b

MARCEL MONTEil CHRISTIAN ALLARD Departement de Surete Nucleaire (IPSN) Commissariat a l'Energie Atomique France

ABSTRACT Technology transfers in matters of nuclear safety for the benefit of countries assisted by foreign firms in building nuclear installations must be directed at the various levels concerned: governmental authorities which take decisions, a technical safety orqantzatlon which prepares the decision, the employer, responsible for construction. This transfer must enable each of these levels to ensure that all necessary measures have been taken in the design, construction and operational phases, to prevent accidents and to limit their consequences, with due consideration of possible local constraints. The extent of the assistance, which appears necessary for this purpose, is merely intended to support the authorities at the different levels to improve their capacity for judgment and decision. )e p} c 6 o 5 p)c e The coming years will undoubtedly witness an ever growing number of countries which will rely on nuclear energy to spark or accelerate their economic development.

In newly deve-

loped countries, with infrastructures which are still incomplete but which are being geared for future possibilities, this means the building of a steadily increasing number of power plants and nuclear research centers with research reactors and various laboratortes , The necessary equipment has already been or is being supplied by industrialized countries, the promoters of this new source of energy, whose companies have gained undeniable expertise in carrying out the construction and commissioning of these installations. The intrinsic share of newly developed countries in their national nuclear achievements will only grow gradually, with their increasing acquisition and circulation of technical and technological know-how in several fields, and with the correlative growth of basic industrial structures. On the other hand, in the area of operation, new teams will have to be trained rap Idly to handle the running and management of these very expensive tools entrusted to their hands.

In terms of safety, it may be stated, without being paradoxical, that the teams 37

should be installed before the beginning.

In effect, the authorities shou Id be able to select

the most suitable sites, the most appropriate installations, for the development of the country, not only from the economic standpoint, but also in human terms. Moreover, from the outset, in order to safeguard the future, they must be capable of providing a guarantee against any incident jeopardizing life and property which may result from the special risks incurred by nuclear energy. However, the question arises as to whether a nuclear risk exists which is significantly different from other risks, and if it does, what should be done to evaluate, reduce and finally to control it? Like any other human enterprise, necessarily imperfect, the use of nuclear fission and its related products does present a risk. Hence it requires precautions, even if this risk has sometimes been exaggerated and presented in an apocalyptic light for purposes often widely different from the declared humanitarian concerns. Very briefly and schematically, the opponents of nuclear energy claim that the very high concentration of energy culminates In uncontrollable nuclear fission, and that the perenniality of radioactivity creates a permanent hazard for all time to come. Although the population of newly developed cou_n tries, through its closer and more daily experience of the close relationship between risk and action, is less sensitive to these arguments, it is nevertheless wise to take care that some levels of their societies do not adopt an attitude of blatant opposition. Irrespective of our position, we must therefore combat this attitude in its irrational aspects, by intensifying our knowledge of the potential causes of the risk, so as to pinpoint it better and to devise increasingly effective means of prevention. This reveals the & ! kmendous importance which must be attached everywhere to the evaluation of nuclear risk and to the determination of precautions to be observed to contain it within limits which are not absolutely ideal, but nevertheless acceptable. 1.

a e c v n76 9 7 5 c e 5 7} e )e 9 pd 7 s o - t 75 p

To limit the scope of the subject and to focus on the most important example, and the one most likely to bear significance today for most countries represented at the Conference, that of the safety of nuclear power plants, this discussion will be restricted to this type of installation. With respect to research reactors and laboratories, as the problems raised and the knowledge required to devise an 'acceptable' solution are not substantially different, the transposition is easy to make. 1.1 Basis The first question to be raised concerns the content of the knowledge indispensable for an evaluation of the degree of safety of a power plant, determination of the methods to be implemented for the best use of this knowledge, and the technical and administrative codifications necessary for their practical application, so as to reach the desired degree of safety. When it is remembered that safety represents all measures taken, from design, through 38

construction t!' operation, to prevent accidents and limit their consequences, It may be claimed that the knowledge necessary to achieve this embraces everything that must be known to design, build and operate a nuclear power plant. This represents a good deal, yet it is Inadequate, because it must be remembered that a power plant is not an isolated entity and hence cannot be considered by itself.

It must be viewed in the natural and

human environment with which it interferes. Beyond considerations of nuclear technology, this knowledge must therefore extend to the fields of biology and medicine to account for possible consequences of radioactivity release on the human environment. However, this spreads beyond the specific attributions of nuclear safety. The latter is, in effect. limited to consideration and control of radioactive sources proper. Despite this limitation the field remains vast, especially since progress in many sciences and techniques must be considered, to achieve constantly better results, subject to constant improvement." The scope and variety of the parameters to be accounted for explain the need to resort to a logical method to analyze the different situations of the Installation and the various interactions between the latter and its environment.

Starting with the very simple prin-

ciple that an attempt should be made to confine radioactivity to the minimum, and to interpose a number of obstacles between radioactive sources and the exterior, the so-called 'barrier' method immediately comes to mind. This consists of the systematic logical analysis of the behavior of a number of shields and containments, in normal operation or following an incident, and to check whether the 'in-depth defense' system thus established performs the role assigned to it.

By applying this method suitably, It should be possible

to determine whether the three keys to safety, prevention, surveillance, emergency action, are suitably respected. Yet the possibility remains that the final barrier will be crossed. Hence the circumstances in which such an event may occur must be analyzed, its frequency determined, and its consequences assessed. This demonstrates the need for probability studies of logical sequences of events and incidents or equipment and system failures. This consideration of random events of low probability serves to broaden even more the extent of knowledge required, and complicates matters.

It raises many questions, and much re-

flection is still necessary before the day when the frequency and magnitude of the risks wi II be correctly estimated. 1. 2 Regulations The foregoing discussion shows that if care is not taken, safety· could soon develop into purely speculative resea·rch. This is why all available knowledge must be codified, from both the technical and administrative standpoints, to provide the basis for action.

In. order

to convert it into criteria, codes and technical or administrative regulations, knowledge of a new order appears to be necessary. From the technical standpoint, safety requirements must be finally reflected by specifications concerning material quality and the quail-

39

:0 ­"& 0+_ +: &>k • k _ ! k =/ 0 ! k ( &+ ( k j 0' _Z x / 0·( Z &k j &" _( +k j k ! k =/ 0! k • k _&j Z ? >0·k ! k • " 0_0_'

­ +• k ¢ " ! k k ::k ­ & 0° k ·¢ " _( ­ +! ! k ­ & ·¢ " k ­ +• k ­ !k" & 0+_ +: _k? !k' / ·" & 0+_j " _( k k• !k' " !( &+ " / & >+!0& 0kj 0j " _k­ kj j 0& ¢3 p>0j k_& 0!k " j k xk& ? kk_ &k­ >_0­ " · " _" ·¢j 0j " _( " ­ & 0+_ " k " ° k xkk_ 0 j 0j " ·0 ° 0_' j & !/ ­ & / !k ? >0­ > • / j &&" • k ( / k ­ +_j 0( k!" & 0+_ _+&+_·¢ +: j ­ 0k_& 0:0­ " ( ° " _­ kj Z x/ &" ·j + +: &>k +? ' " 0_k( 0_ ­ +_j & !/ ­ & 0+_ " _( +k :+!•

+: 0• k_ :" ­ k( ? 0 & > &>k j k >k·< +: =/ " ·0 :0k( j 0' >·¢ " ( " kj k j & !/ ­ & / !kj ? 0 ·· k_" x·k ­ +_j & !/ ­ & 0+_ +: &>k :!" • k? +!• +: &>k !k0_:+!­ k• k_&+: _k? +!' " _0€" & 0+_j ? >0­ > _k? ·¢ ( k° k·+ &+ ­ + &>k k ­ ·0k_&­ +/ _& !¢Z x/ &" x+° k " · ) +: k " ­ & 0+_ &+ xk &" • k_ 0_ 0j +!' " _0 €" & 0+_ 0j " ·j + ' k_F 40

erally responsible for checking that the employer, who is responsible for the·safety of the

installation undergoing construction, observes all the ethics of safety, from both the technical and procedural standpoints. The employer must therefore have his own safety organization, or must call on the above organization in order to make sure that the builder exercises due observance of safety regulations. This schematic breakdown may, according to need and possibilities, be adapted to the situation, but it is important to make a clear distinction in safety matters between the three types of activity: decision, surveillance and action. ; 3 ACTION 8c } pd 7 - 7e 78)p c 8 9 c f 7} e r 7e p4 n 4 o pd c } )p)7s

In view of the inherent responsibi I ities of the Government authorities in the safety area, action for their benefit appears to be necessarily highly varied, both in magnitude and scope. At their request, assistance may be provided in the area of general safety organization, taking special care to separate the three functions mentioned above. Moreover, substantial· information concerning administrative regulations in force in the supplier country must be provided to them. These regulations often exhibit particular features which must be explained and replaced in their context to avoid misunderstanding. In passing it should be stated that with respect to sites, a problem sometimes occurs at the outset, concerning compatibi I ity with the regulations of the exporting country, related to the fact that the choice of the site is often made before that of the supplier of the power plant. This is a matter which must be clarified. Should the authorities so desire, the regulations of the supplier country, or international regulations, may be adapted to the specific conditions prevailing in the purchasing country. These special conditions may, as a first approximation, be of four different types: intense specificity of climatic and meteorological conditions, demography, administrative and legal particularities. Owing to the very sensitive implications of cooperation in the areas mentioned above, the foregoing indications are only given as an example of what can be done. This cooperation is a possibility which must be offered, but, as stated above, it must be clarified for each specific case. S3

4 5 p)c e 8c } pd 7 - 7e 78)p c 8 pd 7 s 4 87pb c } 9 4 e )u4 p)c e

In this case the assistance provided is of a technical order, and therefore less delicate and of wider scope. 4. 1 Organization To start with, if the need arises, assistance may be provided relating to the organization. In view of the experience gained and the faults and deficiencies observed in its own house, the safety organization of the supplier country is perfectly capable of providing advice on SP

the manner in which its counterpart in the client country should be established and strengthened.

By assigning to the client country a few engineers who are thoroughly familiar with

the trials, tribulations and growing pains of such an organization, the latter will be able to avoid repeating past errors and to achieve satisfactory effectiveness with novice engineers fairly rapidly. lacking.

It must nevertheless be remembered that errors are possible if patience is

It may be disastrous for the organization established to be in reality totally in-

effective, and to become the scene of internal stresses which are entirely unacceptable, because it goes against a number of widely accepted principles in the client country. 4. 2 Participation in Assisting the Decision-making Process The first task of the new organization will be to prepare the decisions of the Governmental safety authorities. To do this, it will have to be in possession of the administrative and technical regulation of the supplier country, even if some of its members are already familiar with a specific set of regulations, or with international regulations. We shall not discuss administrative regulations, which were dealt with in the previous chapter. Technical regulations will include general requirements, regulations, codes and guides in use in the supplier country. All regulations must be explained to avoid the risk of hindering attitudes. Many points wi II doubtless have to be adapted to account for certain local contingencies or constraints. From the operational standpoint, if the final choice of the site has not yet been made, assistance wi II start with help in preparing the decision on this matter.

In view of the complexity

and difficulty of this problem in certain cases, this help may be very substantial and include not only documentary surveys, but also active participation in coordinating a wide variety of surveys in the field. The supplier country must therefore be capable of sending in specialized teams in a wide variety of disciplines and techniques, or of strenthening local teams in areas where gaps may emerge. The potential duration of these surveys should not be minimized, in the event that certain basic data are lacking on meteorology, tectonics or seismology, for example. 4. 3 Contribution to the Training of Specialists Simultaneously, collaboration should be established for the training of specialized safety engineers, if the number of engineers already familiar with safety problems proves insufficient. Training can begin by the assignment of some of these engineers to work in the safety organization of the supplier country. They can thus gain fami I iarity with the technical problems raised by the power plant, as well as with the methodology employed to systematize the safety analysis so that no item is neglected. This will also provide the opportunity for some of these engineers to become familiar with safety investigations and tests, or with methods of probabllity analysis. 42

In addition to the value of the training itself, this exchange

offers the " ( °" _& " ' k +: k_>" _­ 0 _' & >k • / & / " · / _( k!j & " _( 0 _' +: " ° k & + ? +!• & +' k& >k!3 p>0 j +!+/ ' > " ? " !k_kj j & >!+/ ' > " ­ & 0° k k :/ _­ & 0+_j ? >0 ­> & >k¢ ? 0 ·· j / xj k=/ k_& ·¢ >" _( ·k & >k• j k·° kj Z & >k¢ ? 0 ··Z 0 j & !" 0 _0 _' ­ +• kj k" !·¢ k_+/ ' >Z xk !k" ( ¢ & + k0 ! " k °" !0 +/ j 0_° kj & 0' " & 0+_j " _( " _" ·¢j kj 3 s +• k k\ " • kj k " !k • k_& 0+_k( xk·+? Z ? 0 & >+/ &_k­ kj j " !0 ·¢ ­ +_j & 0& /& 0_' " _ k\ >" / j & 0° k ·0 j& 3 p>k +/ ·( xk " j j kj j k( 0_ !k·" & 0+_ &+ & >k j " :k& ¢ !k' / ·" & 0+_j " ( +k ­ ·0 k_&­ +/ _& !¢3 p>0 j k°" ·/ " & 0+_ ? 0 ·· 0_­ ·/ ( k " _ " _" ·¢j 0 j +: & >k +°k!" ·· ( kj 0 ' _Z " &+: & >k :/ k· k·k• k_& Z +? 0 _' &+ 0& j 0• " &+: & >k " ( " 0 ­ > " !k ­ +( kj " _( !k' / ·" & 0+_j Z k­ • k( " & & k_& 0° k·¢3 p>kj k & ? + 0 ' >·0 ' >&& >k ' !k" &j 0' _0 :0­ " _­ k ? >0 ­ > • / j &xk " & &" ­ >k( &+ & >k ' / " !" _& kk +: & >k =/ " ·0 & ¢ +: " ·· ­ +• k =/ " ·0 & ¢ " j j / !" _­ k j ¢j & k•

:k" & / !kj " &" ·· ·k° k·j ' / " !" _t Kkkj +:

­ +• 0 < xk& ? kk_ & >k j " :k& ¢ +!' " _0 €" & 0+_ " _( & >k +!' " _0 €" & 0+_ !kj kj k & " j • j Z k::k­ & 0 °k " j j 0 j& " _­ k j >+/ ·( xk k k\ " • 0 _" & 0+_ +: j " :k& ¢ !kk ( 0 ::k!k_&j & " ' kj +: ­ +_j & !/ ­ & 0+_ x¢ & >k 9 +°k!_• k_& "· "/ & >+!0 & 0kj 3 )_ ' k_k!" ·Z & >k " ( • 0 _0 j& !" & 0 ° k !kk • " 0 _ j& k+!0 €" & 0+_ :+! ·+" ( 0 _' +! j & " !& F/ k k • k j / k j " :k &¢ +! ' " _0-

&>k +! ' " _0€" & 0+_ 0& j k ·:3 p >k :+! k 0' _ k \ " ° k &+ ! k _( k !

" ( ° 0­ k +_ &>k • +j &k ::k ­ & 0° k • k & >+( j :+! ( k " ·0 _'

? 0& > &>k k • j k ·° k j &>k " ­ & / " · j " :k &¢ 0_° k j & 0' " & 0+_j Z x ¢ • " • 0_' 0&k •

+/ &0_ &>k :" ­ 0·0& 0k j +: &>k 0! +? _ +! ' " _0€" & 0+_Z +!

0_ " _¢ +&>k ! :" ­ 0·0& ¢ ( k k • k ( j / 0& " x ·k :+! &>k k _ >" ° k &+ k ! k ·k ° " _&­ +_­ ·/ j 0+_j :+! &>k 0_j &" ··" & 0+_3

S3 1 5 +_& ! 0x / & 0+_ &+ )_:+! • " & 0+_ +: &>k h / x ·0­

p >k :+! k ' +0_' ( 0j ­ / j j 0+_ ! k ·" &k j &+ &>k " j j 0j & " _­ k ? >0­ > :+! k 0' _ k \ k +·+' 0­ " · ·k ° k ·Z &>k ° 0&" · 0• k ¢ ? 0·· k ¢ • " ¢ k j " :k & ¢ " _( k _° 0! +_• k _& " · 0·k ( k • +_j & !" & 0_' &>k ­ +_­ ! k & k j +·/ & 0+_j &+ &>k j k k ¢ • " ¢ " ·j + " j j 0j &&>k k j " :k& ¢ +: &>k 0_j & " ··" & 0+_ ? >0­ > >k >" j x/ 0 ·&" _( ? >0­ > >k ? 0 ·· +k x/ !( k_ +: :/ !_0 j >0_' & >k k\ " &&>k j " :k& ¢ +: >0j k ·k° k· +: " ­ & 0+_3 d k_­ k ? k j >" ·· • k!k·¢ j & !kj j & >!kk 0­ > ? k ­ +_j 0( k! 0• k k !0j • j +: k!!+! " &&>k ( 0::k!k_&j & " ' kj +: k ­ +• " &+! ' " _0 €" & 0+_j

+: :+! k0' _ k\ k _" & 0+_" · +!' " _0 €" & 0+_j " !k &+ j / xj & 0& /& k :+! &>k ·" & & k! 3 p? + ­ +_& !" !¢ !k" ­ & 0+_j • " ¢ !kj / )& :!+•

&>0 j j& "& k +: • 0 _( Z x+& > >" ! • :/ · :+! &>k

:/ & / ! k +: _/ ­ ·k" ! k_k!' ¢ 0_ _k? ·¢ ( k° k·+k :0! j &!k" ­ & 0+_ ? +/ ·( xk " !k:/ j " ·Z 0_ & >k :k" ! +: &>k !0 j • +: ·+j j +: _" & 0+_" · j +° k! k0' _& ¢3 p>k j k­ +_( ? +/ ·( xk &+ xk·0 k° k &>" &_+ k::+!& 0j !k=/ 0 ! k( 0_ &>k " !k" j ­ +_j 0( k! k( " x+° kZ j 0 _­ k &>k :+! k0' _ +!' " _0 €" & 0+_j ? 0 ·· ( + " ·· &>" & 0j _k­ kj j " ! ¢3 )&0j kj j k_& 0" · & >" &&>kj k _+& 0+_j xk ( 0 j ­ " !( k( Z xk­ " / j k &>k¢ " !k :" ·j k3 c ! ' " _0 €" & 0+_j +: :+! k0 ' _ k\ k " / & >+! 0& 0kj " &&>k ( 0 ::k! k_&·k° k·j ­ +_­ k!_k( ? 0 & > j " :k& ¢ &+ +x& "0 _ ~/ j & 0 :0k( !k­ +• • k_( " & 0 +_j 3 p>kj k !k­ +• • k_( " & 0+_j • / j &xk k\ k•

"j

&+ ­ +_j & " _& ·¢ 0• k0 ! 0­ > ? 0 ·· >" ° k &+ ( ! " :& &>k !k­ +• • k_( " & 0+_j ·k" ( 0 _' &+ :0_" · ( k­ 0j 0 +_j Z 0: & >k¢ " ! k &>k &k­ >_0 ­"· "/ & >+! 0 & 0kj Z " _( &+ ( k­ 0( k &>k• j k·° kj Z 0: &>k¢ " ! k & >k 9 +° k!_• k_& "· "/ & >+! 0& 0kj 3 p>k¢ ? 0 ·· " ·? " ¢j >" ° k & + (+ & >0 j Z " _( ? 0 ·· xk " x·k &+ " ­ ­ +• & >k0 ! & " j • 0_ & >k :/ ·· • _+? ·k( ' k +: &>k :" ­ & j " _( ? 0 & > ! k" &&+ & >k kj j k_& 0" · \°k

IkXQV\Y

d/V_> yk/;kXXk°k

£tz

rDo•

fk°*V_Q·>\°k

B*HXµ

J·>;GS=

£t~

rDot

fk°*V_Q·>\°k

R/H_;k

rDoD. rDt£

fk°*V_Q·>\°k

efV*=k/XH_W

hBvY4

IbTc

-4E7O

co•

roz

rDoi

fk°*V_Q·>\°k

eKHV_

bV_Q·HX° £

ioz

rDt~

fk°*V_Q·>\°k

efkWk_

Jk_*/HX ;H>/°>

io£

rDtt

AE FE

B*HXµ

rDtt. rDti

fk°*V_Q·>\°k

eKHV_

£

X

Dcz

v_Q/H B

o~t

rDti

fk°*V_Q·>\°k

I/H=VX

Jm° vX*> PH=V>

3ZZ

X

L>°k JHw/~ /H

vXYH/H=

&

£

2

X

Dt~

rzzD

2.

DEVELOPMENT OF REQUIREMENTS

A country planning a nuclear-power program theoretically has the initial choice of adapting existing requirements of some other country, independently developing their own, or adopting a combination of these techniques.

Practically, the last-named course of action

is expedient if for no other reasons than to save cost and time and to avoid others' demonstrated mistakes. Countries which export reactors (e.g. United States, West Germany, Canada) have established requirements. These standards have been developed in parallel with the systems themselves, starting at the beginning of nuclear energy development. The standards are still evolving to some extent as the regulatory commissions acquire more experience regulating these power plants, and their understanding of them deepens. From our present viewpoint, it is evident that the development of nuclear safety requirements is never an easy task. Prodigious resources in time and money have been expended; total costs will never be known. One part of the effort, the R C D programs under-

taken by the regulatory authorities, can be measured by the budgets devoted to this purpose. Also, the cost of writing the explicit requirements themselves could be measured by the budgets of the authorities. These are but a small fraction of the total expenditures. A further unmeasureable effort that enters the evolution of requirements is the accumulation of licensing cases. Safety requirements are·never developed in one orderly, linear effort, but rather in the context of a continuing dialog between regulators and designers, a continuing progression via proposal by the designer and review by the authorities. In the United States, this dialog can be traced by examining the succession of safety analysis reports for various projects. The format that developed was the submittal of a report or amendment by the reactor owner_, followed by questions by the AEC/NRC. When the regulators' questions had all been answered satisfactorily, permission to construct or operate the reactor was granted. Thus, the medium for development of requirements was basically question and response. The expression for regulatory requirements in the United States was essentially satisfactory answers to questions. Explicit requirements followed the earliest reviews of reactor designs. TID-14844 which established the first siting criteria was published in 1962, following start-up of Yankee Rowe, Dresden I, and Indian Point I. More specific siting criteria were later enunciated in Regulatory Guides 1. 3 and 1. 4, published in 1970. Other aspects of safety requirements had a similar history, but occurred even later. In the meantime, developing answers to questions of the regulators involved reactor manufacturers, and plant owners and their engineers in extensive, largely-unrecorded study and research and development expenditures. It can be seen that the cost and effort of developing safety requirements in reactor exporting countries has been great. It would not be prudent for a reactor importing country to develop safety requirements, ignoring the experience of the reactor exporting countries. It would be difficult, moreover, to obtain proposals from foreign manufacturers 50

to fit unfamiliar and to some extent incompatible domestic rules, unless a large market potential existed in the importing country. An alternate would be to adapt existing requirements from one country. There are objections to this procedure also.

First, each country has its own custom and institations

with which safety requirements and regulatory procedures must conform; a specific example is considered in section 4, below. Second, under these circumstances reactor manufactures from the country selected as source of regulations would have an unfair competition advantage, increasing cost to the importing country. United States manufacturers have developed a product suited to United States rules, and West German manufacturers have developed a product suited to West German rules, which are different. Table 11 shows the countries which have imported reactors from more than one country. In most cases, these countries permitted the reactor vendors to use the safety requirements of their own country. This procedure made it easy to get reactor proposals, and resulted in reactors which, judging from experience to date, are sufficiently safe. At the time many of the import-decisions were made, this procedure was the only feasible one, since the importing countries did not have independent safety standards of their own. This procedure has the disadvantage that nuclear power opponents can point to the inconsistencies between the requirements used for the different reactors and ask the government of the importing country why, for each plant feature, the most strict requirement of each exporting country should not be used.

Table 2.

Countries that have imported or contracted for reactors from more than one country

BYK>/*V_Q J>\_*/µ

F+K>/*V_Q

J>\_*/Vk°

B*HXµ

M_V*kW e*H*k°. A/kH* I/V*HV_

LHKH_

M_V*kW e*H*k°. A/kH* I/V*HV_

eKHV_

M_V*kWTe*H*k°. R/H_;k. fk°* Ak/YH_µ

B/H_

R/H_;k. fk°* Ak/YH_µ

e$V*=k/XH_W

M_V*kW e*H*k°. fk°* Ak/YH_µ

IkXQV\Y

M_V*kW e*H*k°. R/H_;k

I/H=VX

M_V*kW e*H*k°. fk°* AkG/YH_µ

v/Qk_*V_H

fk°* Ak/YH_µ. JH_HWH 51

;3

p d 7 o e )p 7 6 s p 4 p 7 s s b s p 7 r

8+! & >k purposes +: &>0j k! k " ! k &? + ­ " & k' +! 0kj +: j " :k& ¢ " _( ·0­ k_j 0_'

A! k=/ 0 ! k• k_& j A 0_ &>k o _0& k( s & "& kj j ¢j & k• 3 c _k k• x ! " ­ kj &>k k\ -

k +& >k! ­ " & k' +! ¢ 0_­ ·/ ( kj

&>k ­ +_& 0_/ +/ j ·¢ k° +·° 0_' !k=/ 0 ! k• k_& j ( k° k·+0j 0_:+! • " & 0+_3 d +? k° k! +x& " 0_k( Z o _0& k( s & "& kj !k=/ 0 ! k• k_& j " ! k 0_ &>k k o _0& k( s & " &kj j ¢j & k•

0j &>k ! k·0 " _­ k k 4 • k! 0­ " _ e / ­ ·k" ! s +­ 0k& ¢ N4 e s R Z &>k )_j & 0&/ & k +: 7 ·k­ & ! 0­ " · " _( 7 ·k­ & ! +_0­ 7 _' 0_kk! j 0­ " · 7 _' 0_kk! j

N)7 7 7 R Z " _( &>k 4 • k! 0­ " _ s +­ 0k& ¢ +: r k­ >" _-

N4 s r 7 R 3 p >kj k j & " _( " ! ( j " ! k :! k=/ k_& ·¢ !k:k! k_­ k( 0_ & >k ' +° k! _• k_&

( +­ / • k_& j ­ 0& k( " x+° k3 p >k¢ " ! k " ·j + !k" ( 0·¢ " ° " 0 ·" x ·k :! +•

& >k j +­ 0k& 0kj ? >0­ >

&>k• 3 c _k +: &>kj k j & " _( " ! ( j Z &>k 4 s r 7 Boiler " _( h ! kj j / ! k f kj j k· 5 +( kZ ' +° k! _j &>k ( kj 0' _ +: " ·· 0j ­ +( k k\ 0j & k( " _( ? " j / j k( :+! xk& &k! j " _( k _/ ­ ·k" ! k! " Z " _( ? 0& > • " _¢ +: &>k o _0& k( s & "& kj Z 0&>" j &>k :+! ­ k +: ·" ? 3 v 0& > &>k ( k° k·+" _' k! j R ­ +_& " 0_• k_& j Z +/ ' > x+& > ?0 ·· k k! k\ < ekK*E rDto7 RV°;HX (kH/

bk°kH/;· B*kY

rDto

rDt~

rDti

rDtt

rDtD

)

rE bkH;*>/ R\kX eHYK\*k/ J>Wk° / e*kHWµ e*H*k H_W d/H_°Vk_* B//HWVH*V>_ *k°*° RV°°E h/>WE Ik·H%V>/ h>$k/ J>>XV_Q UV°YH*;· 5RX>$ J>H°* • >$_7 J>_°*/\;*V>_ >< N>* PHwE / *·k °\/%kVXH_;k >< J>YYk/;VHX R\kX°

J>Wk Wk%kX>_Yk_* H_W %k/V_ LUdb NHXWk_ bkH;*>/ H_W B_*k/T/HYK h/>Vk;* mfPTr X>>_ k+>E F+T>VXk *k°*°E J>_°*/\;*V>_

£E B_*kQ/V*µ >< h/k°°\/k I>\_WH/µ B//HWVH*V>_ FYw/V**XkYk_* >< hE yE J/H;& h/>KHQH*V>_ 5e*H*V; H_W Jµ;XV;7 hVKV_Q b\K*\/k Uk;·H_V°Y H_W Lk* R>/;k UV°°VXk H_W hVKk f·VKKV_Q h/>*k;*V>_

LUdb V//HWE H_W *k°*V_Q dk°*V_£ dk°*V_Q J>_°*E >< WV_Q IX>$ • >$_ R\kX JXHW Ik·H%V>/ 5)/TfH*k/ bkH;*V>_ H_W k*;E7 bkH;*>/ J>_*HV_Yk_* eKH/µ H_W RVX*/H*V>_

5MbB7 u bmevTc bmevT£ m_k WVYk_°V>_HX *k°* 5N>* $HXX kW F+Kk/VYk_* J>_°*E >< _ k+Kk/VYk_*

eE ekV°YV; F/ e*/\;*\/k°

e\/%kµ

mKk/H*V>_ I

)

------- FFFFFF·

F+Kk/VYk_* I

oE v_HXµ°V° H_W F%HX\H*V>_ >< Pfb eHWk° • k%kX>KYk_* hk//YH_;k e*\Wµ >< F%HX\H*V>_ J>Wk°

UkWV\ °V=k w/kH& H_W >*·k/° PmJvTFJJe hk//YH~ ;kT*k°* H_W F+k;\*V>_

tE F_%V/>_Yk_*HX bHWVH*V>_ Pk%kX F%HX\H*V>_

P>$ Xk%kX /HWE Wk*E H_W k%HX\H*V>_ °µ°*kY Wk%kX>KYk_*

iE bHWV>H;*V%k fH°*k UH_HQkYk_* • k%kX>KYk_*

dk;·_>X>Qµ Wk%kX>>Yk_*X H_W °H_ k+KE )

)

p

7 _° 0! +_• k _& " · j " :k&¢ j & /(0 k j " _( ? " j & k • " _" ' k • k _&( k ° k ·+k 0' > &k• k k :/ k · j " :k& ¢ ! k j k " ! ­ > • / j & &>k ! k:+! k x k k :/ k ·

x k >" ° 0+! 3 5 +_j 0 ( k ! 0_'

&>k " x +° k Z &>k :+··+? 0 _'

" ! k x k 0_'

" ° 0+! ! k ·k " j k ( :! +•

SR

8/ k · x k>" ° 0+! / _( k ! " ¢( ! 0( k :" 0·/ ! k Z h 5 r )Z :/ k · ­ k_& k! &k• k

0_ t 4 7 } )3

t 47} ) E j 0! ! " ( 0" & 0+_ &k j & j 0_ &>k d " ·( k _ } k" ­ & +! h ! +~ k ­ &>" ° k x kk_ ­ +_& 0_/ k ( j 0_­ k P2UWZ " _( j + :" ! L; &kj &:/ k · " j j k • x ·0 k j >" ° k x kk_ 0! ! " ( 0" & k( 3 s k ° k ! " · :/ k ·j " ! k :/ ! & >k ! &+ x k 0! ! " ( 0" & k( 3 p >k d " ·( k _ } k " ­ & +! h ! +~ k ­ & >" j " _ k\ ­ k ··k _&0_F­ +! k 0_j & ! / • k _& "& 0+_ &k­ >_0=/ k Z ? >0­ > >" j • " ( k " j 0' _0 :0­ " _&­ +_& ! 0x / & 0+_ &+ t 4 7 } ·E j :/ k · ! kj k " ! ­ >3 8+! k\ " • k ( 0 " • k& k! j +: &>! k k :/ k · ! +( j

61

can be measured simultaneously under power conditions, so that more local and direct

data can be obtained for the PCMI studies. Fuel densitication studies have been pertormed, using JMTR irradiation facilities, since 1973, and so far four capsules and one loop specimen have been irradiated and are being examined. One example is shown in Fig. 1 where the densification is represented by the decrease of pore ratio (ratio of the theoretical density to the smear density) as a function of the grain size. Also the PCMI is studied at the JMTR using three capsules. One of the unique features of this experiment is that two kinds of pellets are provided for the test: one is solid and cylindrical in shape and the other with eight segmented cross sections. Using these pellets the relocation phenomena can be examined, which are of key importance in the PCMI. +0.2

i. ~

----.----r--r---.--r----r---, •

0 •..

F

Cl>

3 .AA° ±

±

J-o.4 . .

0

~ o -0.6t-

iu:

( JJm )

•

•

'-0.2tu

Groin Size

F

±±F· Z

F

4)

/

-0.8t-

-1

I

0

:

: •;. TD }JAERI

A

89

Chubb ( 1975 l

I

I

I

I

I

2

3

4

5

6

F

Fig. 1. Density Increase [Fractional Porosity Decrease) as a Function of Grain Size

H

JAERI has participated in the Inter-Ramp Project hosted by AB Atomenergi, Sweden, where the failure mechanism of BWR fuels under operational transient conditions are studied. Subsequently the Over-Ramp Project for PWR fuels is to be started and JAERl's participation is now under procedure. Data obtained from these irradiation tests will provide a basis for the verification of the computer codes being developed in JAERI. 3. 2 Development of Computer Codes for Fuel Behavior' Analyses The computer codes under development for the fuel behavior analyses may be classified into two categories corresponding to the principal fuel design criterla under normal conditions. The one is mainly to predict the temperature distribution corresponding to the criteria on the maximum temperature and the stored energy in the fuel rod. The other is 62

to analyze the stress and strain of the fuel rod corresponding to the cladding derorm atton criterion.

The computer codes under development are listed in Table 2.

As seen in the

table, these codes FREG-3, FREG-4 and FREC-3, provide an integrated thermal and mechanical analysis of a reactor fuel pin in terms of its path-dependent power history.

As an

exam ple, in Fig. 2 the fuel center tem perature predicted by FREG-3 is com pared with the experim ental data of IFA-224 irradiated in the Halden reactor.

As burnup is increased,

the prediction tends to become higher than the experimental data, thus providing a conservative evaluation. In addition to the above described studies, the release and subsequent behavior of the fission products from defective fuels under a postulated main steam line failure accident of a BW R has been investigated using the OW L-1 loop at the JMTR.

So far six runs have

been carried out. Also an out-of-pile thermohydraulic experim ent on the fuel behavior under a PCM condition is being carried out of which data will be analyzed with reference to the in-pile data obtained at the PBF in USA. Post-irradiation exam ination of fuels used in comm ercial reactors is of im portance in obtaining data on actual fuel performance.

However, so far in the world such examinations

can be made in a very lim ited number of facilities.

A large scale fuel examination facility

is now under construction at JAERI which could provide comm ercial fuel data.

The compa-

rison of these data with various separate tests data will further increase reliability of fuels. The facility is designed to be capable of exam ining a total of about twenty fuel assemblies a year from current PWR, BW R and advanced thermal reactors, including plutonium enriched ones.

Six concrete and three lead cells are provided for the beta-gamma line,

and two concrete and ten lead cells for the alpha-gamma line.

This facility is expected to

be in operation in the later half of 1979.

4.

RESEARCH ON THE COOLANT PRESSURE BOUNDAPv

Based on the philosophy of "defence in depth", the requirements for the reactor coolant pressure boundary are firstly to maintain its integrity and secondly to keep the consequence to a minim um in the event of that integrity being violated.

The research in the

pressure boundary is to ensure this philosophy in reactor plants.

The reactor coolant

pressure boundary includes such components as pressure vessel, primary cooling system

pipings, steam generators and other associated components. Studies to confirm the

safety limits for these components under normal and accident conditions are being carried

out in JAERI, as shown in Fig. 3. 4. 1 Pressure Vessel Studies

The studies on the pressure vessel consist of the irradiation embrittlement tests of pressure vessel materials and the cyclic pressure tests of the pressure vessel models.

oc

p " x ·k L3

5 +• s &!kj j

e k/ & !+_ )! !+( ·+& 0+_

5 ¢­ ·0­ n+" ( 0_' +: ·_&k!_+E h!kj j / !k " _( 7\ &k!_" · ·33++(

e k/ & !" · ·_!/ Kk( 7• x!·&&0 k• k_&

h} 7s s o } 7 f 7s s 7n

} o hpo } 7 s po 6 b c e h} )r 4} b 5 )} 5 o )p h)h)e 9

c ­ ­ / !!k_KKk +: High s &• j j

v k·( 0_'

s k_j 0& 0€" & 0+_

5 !+­ • h!+" ! )• P

d 0' > h !kj j / !k v "& k! s / X\Yk° h/k°°\/k yk°°kX 5V_;XE V_*k/_HX°7 YX YX bk;V/;\XH*V>_ ;V/;\V*°

rEcz zEzo~

°*kHY W>Yk

YX

zE•~r

\KKk/ W>$_;>Yk/

Yc

zErz£

X>$k/ W>$_;>Yk/

Yc

zErzi

\KKk/ KXk_\Y C °*kHY °kKH/H*>/

YX

zErtt

;>/k

Yc

zEr£r

X>$k/ KXk_\Y

YX

zE£oD

IfbTo ~Dc £DEt £zo

izEt ozEz r£c

£

£

/k;V/;\XH*V>_ X>>K

4

i•i

< J>YK>_k_*°

J>/k *·k/YHX K>$k/

&$

NkH*kW Xk_Q*· •

rEii

cEto

J>>XH_* ;>_WV*V>_° K/k°°\/k

&QC;YLQ

t£Ei

;>/k $ /H*k

&QC°

coE•

°*kHY $

&QC°

/k;V/;\XH*V>_ K\YK $ /H*k

&QC°

k k\ " j x k k_ ­ +_& 0_/ k ( since May rDt~E ( / !0 _' x ·+? ( +? _ / < &+ &>k +­ ­ / ! ! k _­ k +: 6 e -

)_ &>0j &kj &&>k >k" & &! " _j :k! k kj & " x ·0 j >• k_&

+: " >k" & &! " _j :k! • +( k · &+' k& >k ! ? 0& > &>k } c s 4 ( " & "3

D3 L } k j k " ! ­ > +_ } k:·++( 0_'

p + 0_° kj & 0' " & k &>k &>k ! • +>¢( ! " / ·0 ­ j ( / ! 0_'

! k :·++( 0 _' Z " _( &+ ( k ° k ·+< &>k ! k :·++( 0 _' ­ +• -

k P6 k\ Z ( 0 ! k­ &>k" & 0_= x / _( ·k Z " _( &>k >k" & k! j / ! :" ­ k &k• " j k - ? 0& > j 0• / ·" & k( ( +? _­ +• k ! " _( k ·" ! ' k j ­ " ·k ! k :·++( 0 _' &kj &0j 0_& k_( k ( &+ x k " _ 0_& k' ! " · ( k • +_j & !" & 0+_ &kj &/ j 0_' " :" ­ 0 ·0& ¢ ?0 & > " j 0€k :! +•

? >0­ > ­ +_­ ·/ j 0+_j ­ +/ ·( x k k\ & k_( k ( &+ ·" ! ' k ­ +• • k ! ­ 0" · h v } E j 3

7 \ ! +/ ' > &>0j &kj &? 0 ·· x k / j k( :+! &>k ° k ! 0:0­ " & 0+_ +: &>k ! k :·++( 0_' ­ +( k j / _( k ! ( k° k ·+k +x ~ k­ & 0° k +: &>0j &kj & 0j &+ 0_° kj & 0' " & k j / ­ > k::k­ & j "j • / ·& 0F( 0• k _j 0 +_" · :·+? 0_ " ·" ! ' k ­ +! k Z &>k k::k ­ & 0° k >k" ( 0_ &>k ( +? _­ +• k ! Z " _( >¢( ! " / ·0 ­j 0_ &>k x ! +• k _ " _( 0_& " ­ & ·++k " \ 0" · ( 0• k _j 0+_ +: &>0j :" ­ 0·0 & ¢ 0j +: &>k :/ ·· j 0€k +: " PXXX r v k ­ ·" j j h v } Z ? >0 ·k &>k >+! 0€+_& " · ­ ! +j j j k ­ & 0+_ +: &>k j 0• / ·" & k( ­ +! k 0j " x +/ &P.LDZ ­ +_&" 0 _0 _'

_0 _k >k" & k ! x/ _( ·k j +: PD\ PD ·" & & 0­ k 3 p >k :·+? ( 0" ' ! " •

" _( j ¢j & k•

k +x ~ k­ & 0° k +: &>0 j j& / ( ¢ 0j &+ +x & " 0_ / _( k ! j & " _( 0_' +_ &>k x k>" ° 0+! +: &>k :/ k · ­ ·" ( ( 0 _' / _( k ! nc 5 4 ­ +_( 0 & 0+_j Z " _( &+ :kk( x " ­ • & >k " x +° k • _+? ·k ( ' k &+ &>k 7 5 5 s & 9 " 0_

t2 PS ( mg/cm2)

•4I

I

PU

18

I

20

Fig. 10. Ductilities of Steam Reacted Zircatoy=u Tubes as Functions of Reacted Amounts and Reacted Temperatures (Test Temp.: 1ok • 0 j k ! 0­ " · &+< +: W•

( 0" • k & k ! " _( LX•

&+& " · >k 0' >& Z k=/ 0 0­ > ("& " >" ° k x k k _ " _" ·¢€k ( " _( / j k ( :+! &>k ° k ! 0:0­ " & 0+_ +: ! k ·" & k ( ­ +• k r " ! • F)) ­ +_& " 0_• k _&+: 9 7 & ¢" j x k k _ &k j & k( ? 0& > " P.U j ­ " ·k • +( k · ? 0& > • / ·& 0k :" 0 ·/ !k & >!kj >+·( j :+! :/ k· !+( j " _( & >k • " ' _0 & / ( k +: " _¢ :/ k·F ?"& k! !k" ­ & 0+_j ? >0 ­ > ­ +/ ·( !kj / ·&:!+• j / ( ( k_ 0_­ !k" j k 0_ k :0!j &" j k >" j xkk_ ­ +_( / ­ & k( j 0 _­ k c ­ & +xk! P2WD " _( ? 0 ·· xk ­ +_& 0_/ k( :+! " x+/ &:+/ ! ¢k" !j ? >k!k &>k ­ +!k " _( :/ k· xk>" ° 0+!j / _( k! } )4 ­ +_( 0 & 0+_j " !k j & /(0 k( x¢ k\ " • 0 _0 _' & >k :" 0 ·/ !k k k\ k ­ k_& k! k e s } } Z ? >0­ > ' k_k!" & kj °k!¢ !" k o _0 & k( a0_' ( +•

0_

1957. 7\ &k_j 0° k • +( 0:0­ " & 0 +_j &+ & >k ( kj 0 ' _ +: ­ +!k j & !/ ­ & / !kZ " ( &+ xk ­ +_( / ­ & k( & + 0_& !+( / ­ k k" !& >=/ " • k !kj 0 j& " _­ k xk­ " / j k & >k k ­ +!k j & !/ ­ & / !kZ °0 x!" & 0+_ & kj & j ? k!k ­ " !!0 k( +/ &0_ & >k o 3a3 " _( t " +j k ­ +_& !" ­ &j k " j k0 j• 0 ­ ( kj 0 ' _ +: j / 0' > x/ 0 ·( 0 _' j x/ && >k!k ? " j _+ 0_j & " _­ k +: k• x/ 0 ·( 0 _' j 3 )_ ­ +_j k=/ k_­ kZ t 4h5 E j e +3 L !k" ­ & +!Z & >k pj / !" ' " h+? k! h·" _& Z xk­ " • k & >k :0!j &!k" ­ & +! 0_ & >k ? +!·( &+ k• +j k ( " ¢j Z _k0& >k! 9 7 _+! & >k ­ +_j / ·& " _&­ +• & >k } k' / ·" & ­ !¢ 9 / 0 ( kj " _( s & " _( " !( } k°0 k? h·" _j 0_ & >k o 3s 343 >" ° k xkk_ ­ +_j +·0 ( "& k( &+ j & " _( " !( 0 €k " _" ·¢& 0­ " · • k& >+( j Z +k !k' / ·" & +!¢ " / & >+!0& 0kj " !k ­ / j & +• " !0 ·¢ " ­ ­ / • / ·" & k( 0_ t " & >k 0• " • 0 _' & " x·k ? " j kj & " x·0 j >k( " && >k xk' 0 __0 _' +: P2WU3 )e p} c 6 o 5 p)c e c _k +: & >k 0• k j " :k& ¢ +: " _/ ­ ·k" ! k °0& " · :" ­ 0) 0& 0kj

+: _/ ­ ·k " ! k k >0' > &k­ >_+·+' 0­ " · ­ >" ! " ­ & k ! 0j & 0­ j +: j +F­ " ··k ( " j k 0j • 0 ­ 0& ¢ xk­ " / j k & >k ! k >" j x kk_ _+ " ( ° " _­ k( " j " &&>k &? " & k! ! k" ­ &+! j

Nh v } Z - v } R >" ° k k _ &>k :0! j & ­ +• • k ! ­ 0" · ·0' >& ? " & k ! ! k" ­ & +! :+! t " " j x kk_ k• k k" ! & >=/ " • k ­ +/ _& ! ¢ +: t " " j x kk_ " ­ & 0° k ·¢ 0_& ! +( / ­ k( 3 4 :&k ! &>0 j Z :+! " x +/ &&k_ ¢k" ! j / < &+ &>k k k ! 0& k( ? 0& > ·0& & ·k ­ >" _' k j Z 0_­ ·/ ( 0_'

! k:0_k -

• k _& j &+ _/ ­ ·k " ! k 0' >F! 0j k x/ 0 ·( 0 _'

­ +_j & !/ ­ & 0+_j Z &>k 0• & >k " j k 0j • 0­ ( k j 0' _ +: _/ ­ ·k " ! " ° k _+&x kk _ • " _¢ 0_j & " _­ k j 0_ t " k e 0 0' " & " k" ! & >=/ " • k 3 - k ­ " / j k +: &>k 0_j / ::0­ 0k _& ·¢ " j k 0j • 0­

( k j 0' _ +: & >k +0· k k" ! & >=/ " • k ­ " / j k( ( " • " ' k &+ &>k •

­ " / j 0_' +0·j &+ :·+?

+/ &+: &>k :0! k " _( ( k j & ! +¢ >/ _( ! k( j +: k ( 0 j "j& k! ( 0( _+&! k • " 0 _ ? 0& >0 _ &>k +0 · =/ " • k 0_ " _ k_' 0_kk! 0_'

" j " _ " j k0j • 0­ ( kj 0' _Z

• _+? ·k( ' k +: & >k ­ " / j kj +: k" ! & >=/ " • k +­ ­ / ! ! k_­ k ? +/ ·( xk 0• +_ k" ! & >=/ " • k " j x¢ ( k( / ­ 0_'

? >k! k " _( ? >" &j +! &+: k" ! & >-

=/ " • kj >" ° k +­ ­ / ! ! k( x" j k( +_ & >k _/ • xk! +: & 0• kj " _( & ­ +_& 0_/ k &+ +­ ­ / ! 0_ " j +• k? >" &k\ k j 0& kR j >+/ ·(

xk k\ " • 0_k( 0_ ( k& "0 · & + k·/ ­ 0( " & k k& >k! & >k k" ··+? Z ( kk ? >" &:!k=/ k_­ ¢Z k& ­3 ): " _¢ " ­ & 0° k :" / ·& j 0_ & >k ° 0­ 0_0& ¢ +: " j 0& k j >+? j 0' _j +: " ­ & 0° 0& ¢ 0_ & >k ' k+·+' 0­ " · & k! • +: &>k =/ " & k! _" ! ¢ " " &" ­ ­ k·k! " & 0+_ ? +/ ·( xk 0_­ ! k" j k( j / ( ( k_·¢ " j & >k ( 0j & " _­ k _k" ! j " :" / ·& 3 )_ &>+j k ­ " j kj +! " ! k" j Z >0j & +! 0­ " · k" ! & >=/ " • k ( " & " ­ " _ xk " ·j + / j k:/ · &+ ( k:0_k & >k ( kj 0' _ k" ! & >=/ " • k xk­ " / j k " _­ 0k_&k" ! & >=/ " • kj ? +/ ·( >" ° k xkk_ ' k_k! " & k( x¢ :" / ·&• +° k• k_& j 0_ & >k j " • k !k' 0+_3 c : ­ +/ ! j kZ x/ 0 ·( 0 _' " _¢ j & !/ ­ & / ! kj 0• • k( 0" & k·¢ " x+° k " :" / ·&j >+/ ·( xk " ° +0( k( 3

L3 D 6 kj 0' _ 9 ! +/ _(

r +° k• k_&

v >0­ >k° k! 0j ­ +_j 0( k! k( &+ >" ° k ' ! k" & k! k" ! & >=/ " • k :+! ­ k 0j " " k ( kj 0' _ ' ! +/ _( • +° k• k_&• / j &xk ( k-

& k! • 0_k( 3 p>k • " \ 0• / •

" ­ ­ k·k! " & 0+_ +: & >k ( kj 0' _ ' ! +/ _( • +° k• k_&0j ( k& k! • 0_k( 0_ ( / k ­ +_j 0-

( k! " & 0+_ +: 0R & >k k" ! & >=/ " • k >0 j& +! ¢ 0_ &>k ° 0­ 0_0 & ¢ +: &>k j 0& kZ 00 R " • " < +: " _& 0­ 0k • " \ 0• / • " ­ ­ k·k! " & 0+_ ( k( / ­ k( x¢ 6 ! 3 a " ? " j / • 0B J " :& k! j & "& 0j & 0­ " ··¢ ­ +_j +·0( " & 0_' &>k >0 j& +! ¢ +: ( " • " ' kZ 00 0 R( 0 j& " _­ k +: kk j 0& k j >+? _ +_ " • " < 0_( 0­ " & 0_'

0_& k_j k k" ! & >=/ " • kj 0_ &>k k! j 3 4 & +& " · +: & >! kk ? " ° kj 0j k• k ( kj 0' _ ' ! +/ _( • +° k• k_& Q +_k ? " ° k 0j j k·k­ & k( :!+• +xj k! ° k( !k­ +! ( j " &" j 0& kZ " _( & >k +& >k! &? + ? " ° kj " ! k j k·k­ & k( :!+•

& >+j k !k­ +! ( k( / _( k!

j 0• 0 ·" ! j +0· ­ +_( 0& 0+_ &+ & >+j k +_ & >k j 0& k3 p>k j >" k ( kj 0' _ ' ! +/ _( • +° kF

iD

• k _& j ? >0­ > >" ° k x k k _ ( k & k ! • 0_k ( " &&>k k " j k0 j• 0 ­

t " +/ ' >

( kj 0 ' _ " _( j " :k& ¢ 0 ­ > &? " j " j & "& 0­ ( kj 0' _3 p>k & k­ >_0 ­ " · j k ­ +• • 0 & & kk 0_( 0 ­"& k( & >" &• / ·& 0k - / 0) ( 0 _' s & " _( " !( n" ? +: t " k j & "& k +: " j k0 j• 0 ­ k_' 0 _kk!0 _' 0_ t " k & 0• k ? " j &>" &" " ( ~ / j &xkk_ j & " !& k( xk­ " / j k of !k­ +' _0 & 0+_

x¢ !kj k" !­ >k!j & >" &j & "& 0­

" _" ·¢j 0 j +: " j k0 j• 0 ­ ( kj 0 ' _ • / j &xk ­ >" _' k( &+ ( ¢_" • 0 ­ " _" ·¢j 0 j Q " _( & >" &" !kj " ( ~ / j &xkk_ ­ +• • k_­ k( " j " ­ " j k j & / ( ¢Q " _( & >" &&>k / j k +: k·k­ & !+_0 ­ ­ +• kj k ­ 0 !­ / • j& " _­ kj Z & >k " ( +k!k ? " j Z >+? k°k!Z " _ +" _0 ­ " · k_' 0 _kk!0 _' · & >" & Z x" j k( +_ & >k k\ F !0j k x/ 0 ·( 0 _' 0_ t " F !0 j k x/ 0 ·( 0 _' k!" 3 c _k 0_( 0 ­"& 0+_ ? " j & >" && >k ' !k" &k" !& >=/ " • k 0_ r k\ 0­ + ­ 0& ¢ 0_ P2DW ­ " / j k( _+ ( " • " ' k &+ & >k S; j & +!k¢ n" & 0_+ 4• k!0 ­ " _" - / 0 ·( 0 _' 3 p>0 j ? " j " ' ++( k\ " • " j j k!°k( " j " _ 0_­ k_& 0° k :+! & >k ' !" ( / " · ­ +_ j +· 0( " F

2P

& 0+_ +: &++·j 0_ +/ ! ­ +/ _& ! ¢Z 03k 3 ? 0& > &>k ­ +··k­ & 0+_ +: k" ! & >=/ " • k !k­ +! ( j " _( &>k F! 0j k x/ 0)( 0_' j ? " j " • " & &k! +: &0• k 0: +_·¢ +·" ! j " _( ? " j ? " 0& 0_' :+! &>k " __+/ _­ k• k_&+: k r )p I Nr 0_0j & ! ¢ +: )_& k! _" & 0+_" · p ! " ( k " _( )_( / j & ! ¢R 3 )_ • +j &­ " j kj Z &>k :0! j &" 0 j Z • " _/ :" ­ & / ! k! j ? k! k !k=/ kj & k( &+ k o _0& k( s & "& kj x/ &? >k_ ? k ­ >k­ • k( 0_ 9 7 +::0­ k " :& k! j 0' _0_' &>k ­ +_& !" ­ & Z &>k! k ? " j _+ j & " :: ? >+ • _k? ( ¢_" • 0 ­ " _" ·¢j 0j Z " _( ·k" ! _k( &>" &&>0 j ? +! • ? " j k_& !/ j & k( &+ &>k0! ­ +_j / ·& " _&t 34 3 - ·/ • kZ " _( k° k_ &>0j +::0­ k 0­ > 0j _+? " _+! • " · 0_ " x/ 0 ·( 0_' " _" ·¢j 0j !" & >k! &>" _ +_ & >k k! k ? " j _+ ­ +_­ kk 8·++! s k o 3s 34 3 )_ &>k ­ " j k +: s " _ c _+:! k

0j Z 9 7 ? " j !k=/ kj & k( & + • +( k

r c 67 ec 3

Z

s 75 p)c e Fig. 6.

94

hPv-

Reactor Building of Takai No. 2 Nuclear Power Plant

7 n3 N• · 63.65

tt ; ,.

.IE

57.0

F 7n 57e p} c F F p48p - .. - IBARAGI --- 3Co

T,/ '

46.5

:j'

)

38.8 34.7 29.0

;4

I

Z:

E )

20.3 14.0

3

;

8.2

2.0) -4.0 -l0.5 X

'

P

3

)0 :

Cl5

4 ­ ­ k·k!" & 0+_

Fig. 7.

DX

X

1.0

1.5

)X3X

X

P3X

2.0

Shear ( x 1041)

(Q)

Dynamic Analysis Model of Takai No. 2 Nuclear Power Plant Reactor Building

4 ·• +j &&>k j " • k ! kj +? _ +_ k_° k ·+k ! k" ­ & +! k ­ k_& k! +: &>k x / 0)( 0_' Z 0&? " j ( k j 0' _k ( x " j k( +_ &>k ! kj / ·& j +: " _" ·¢j 0j ­ +_( / ­ & k( j k +? _ 0_ 80' 3 W &+ &>k • +( k · +: 80' 3 1Z ( 0 j& ! 0x / & 0+_j

+: j >k " ! :+! ­ k Z x k _( 0 _' • +• k _&" _( ( 0 j " j k 0j • 0­ ? " ·· ? k ! k ­ " ·­ / ·" & k( 3 5 +• k j >k " ! :+! ­ k +x &" 0_k( &>k ! k :! +•

&+ ( k & k! • 0 _k

( k j 0' _ j & ! k j j +: k" ­ > ? " · )3 4 j &+ &>k " ! ! " _' k • k_&+: ! kx " ! j :+! &>k " j k 0j • 0­ ? " ··j Z ! k 0_:+! ­ k• k _&x " ! j " ' " 0_j & k" ! & > ­ " ·­ / ·" & 0_'

k" ­ > +: j 0\

­ " j k j " _ k_° k ·+0_' Z j / ­ > " j " &" _• ? 0& > ­ +·/ • _

j / " _' k j 0_ &>k _/ • x k ! +: :! k=/ k _­ 0k j 3 4 _ k\ " • +? _ 0_ 80' 3 2 0j • " ( k j / ­ > &>" &&>k k :! k=/ k_­ 0k j ­ +! ! kj +? j " _ k\ " • k ­ " ·­ / ·" & k( ! k j k ·Z a 3t 3Z As =/ " • k d " €" !( j 0_ t " 0j ­ +_­ ·/ ( k j • ¢ ! k• " ! • j >k ! k &+( " ¢ +_ >+? +k j & k" &" ! k &" • k _ &+ ' " & >k ! &>k ++j k ! k" ­ & +! j _+? 0_ +" &" ­ +• k ! k( :! +•

&>k +? _k ! j +: &>k _/ ­ ·k " !

0j ( 0 j ­ / j j 0+_ +: & >k ­ +• ( +­ / • k _& 0( k _& 0:0k ( 0_ 80' / ! k L +: &>0 j k

p k ­ >_0­ " · s _/ ­ ·k " ! & >k k e } 5 >" j kj & " x ·0j >k( • " _¢ k\ ­ >" _' k " _' k +/ ·( x k j k _&&+ 6 ! 3 t +j k n" :·k / ! Z 6 k k _/ ­ ·k " ! 0_( / j & ! ¢ " j 0_k 0 ! ( k j 0' _ 0• k • +! k j 0' _0:0­ " _&0j j / k j 3 p >k j k x / ··k & 0_j " ! k x ! 0k : +/ &) 0_k j +: +­ ­ / ! ! k _­ k j &>" &>" ° k ' k _k ! 0­ 0• < ) 0­ " & 0+_j Z k 3' 3Z " :" 0 ·/ ! k +! ( k :0­ 0k _­ ¢ 0( k _& 0:0k ( 0_ k=/ 0 k ¢

" ! k " ·j + k _& k ! k ( 0_& + & >k ·+­ " · k x / ··k & 0_ " _( & >k x / ··k & 0_ ? ! " k 0_( / j & ! ¢ " j 0_k & ! " _j :k! +: _/ ­ ·k " ! &k ­ >_q+·+' ¢ :! +•

+_k ­ +/ _& ! ¢ &+

" _+& >k ! ( + _+&" ·? " ¢j 0­ > 0j "­& / " ··¢ 0_° +·° k( 0_ &>k & ! " _j :k! 3 )&0j &>k ! k:+! k ? +! & >? >0 ·k ­ +_j 0( k ! 0_'

j k • +! k ' k _k ! " · j & "& k• k_& j 3 ) k & ! " _j :k! +: _/ ­ ·k " ! _+·+' ¢3 ) ­ >+j k & >0j k\ " • " ° k j +• k :0! j & F>" _( • _+? ·k( ' k +: &>k j / x ~ k­ & Z >" ° 0_' xk k_ ­ +_­ k ! _k( ? 0& > j& / ( ¢0_'

&>0j 0' >·¢ ! k ·k ° " _&0_ & >" &0&0_° +·° kj _+&+_·¢ ­ +• • k ! ­ 0" · ­ +_j 0( k ! " & 0+_j x/ &" ·j + ° k! ¢ 0• " _( j " :k& ¢3 p >0j ·" & k ! " j k j +j k _ ? " j &>k ­ +_j 0( k ! " x ·k k c / & j k&+: +/ ! j & / ( ¢ &>" &&>k ! k ? k ! k _+ x ! +" ( ' k _k ! " · 0­ > ­ +/ ·( x k 0_° +• k( &+ ' / " ! " _& kk " x j +·/ & k ·¢ &>k j " :k& ¢ +: &>k j k ° kj j k ·j 3 } " & >k ! Z ° k ! ¢ ·+? +? j " &¢k ! 0 ­ " · &+k " ( Z " _( " ·+? k ! :0\ k( k 0_·k &" _( +/ & ·k & ­ ++·" _&_+€€·k j Z " ­ ¢·0 _( ! 0­ " · x " ! ! k · " _(

>k " ( " · ) ? k ·( k ( &+' k& >k ! 3 p >k / k ­ +_-

& ! +· ! +( ( ! 0° k • k ­ >" _0 j • j " _( &>k ·+? k ! >k " ( x ¢ j • " ··k ! 0_j & ! / • k _&_+€€·k j 3 p >k ° kj j k ·

PP;

0j ­ +_j & !/ ­ & k( +: ·+? " ··+¢ ­ " ! x +_ j & kk · ? 0& > &>k 0_j 0( k j / ! :" ­ kj ­ ·" ( ? 0& > " ­ +! ! +j 0+_ ! k j 0j & " _&" / j & k_0& 0­ j & " 0_·k j j j & kk ·3 p >k ­ +! k 0j >+/ j k( " &" ·k° k · x k ·+? &>k _+€€·k j " _( " x +° k &>k x+& & +•

>k • 0j k ! 0­ " · >k " ( 3

4 0­ > &>k • " &k ! 0" · ­"_ j/ j & " 0_ ? 0 & >+/ &0_( k :0_0& k k k ° 0­ 0_0& ¢ +: &>k ­ ! " ­ • & 0< " ! k ' ! k " & ·¢ • " ' _0:0k ( +° k! & >k 0 ! ° " ·/ k j 0_ / _:·" ? k( • " & k! 0" · " _( • " ¢ x k j + ·" ! ' k &>" & &>k :/ _( " • k _& " · ­ +>k j 0+_ +: &>k • " & k! 0" · 0j k\ ­ kk( k( 3 p >k ­ ! " ­ • &>k _ k\ & k_( j 0_( k :0_0& k·¢ " _( &>k j k ­ & 0+_ :! " ­ & / ! kj 3 p + ( 0j & 0_' / 0j > 0&:!+•

:" 0 ·/ ! k x¢ ' k_k ! " · 0­ > :·" ? j +! ­ ! " ­ • j +: ( 0 ::k! k _&j 0 €k " _( j >" k ° kj j k · ? " · )Z " _( & >k ! kj / ·& 0_'

­ +_j k=/ k_­ k k° " ·/ " & k( 3 s / ­ >

" _ " _" ·¢j 0j 0_ _+ ? " ¢ 0_:k! j &>" &j / ­ > :·" ? j " ! k " ­ & / " ··¢ k

( k " ·: x / ! 0k( 0_ " ·/ • < +: k 0j " j j / • k( ( k :k­ &• " ¢ ' ! +? / _( k ! &>k ­ +• x 0_k( " ­ & 0+_ +: j & ! kj j ­ >" _' k j 0_( / ­ k( x ¢ j & " !& 0_' &>k ! k " ­ &+! / / & & 0_' it ( +? _Z ­ >" _' 0 _' +_0=/ kj ? k F" ! k " x ·k &+ ­ " ·­ / ·" & k & >k • " \ 0• / •

j 0€k " ­ ! " ­ • • " ¢

! k " ­ > xk :+! k 0&x k­ +• k j / _j & " x ·k " _( 0_( / ­ k j " _+_F( / ­ & 0·k :" 0 ·/ ! k 3 ): &>k A­ ! 0& 0­ " ·A j 0€k N( k _+& k( x¢ &>k ( +& & k( ·0 _k j 0_ 80' 3 Lx { 0j j 0' _0:0­ " _& ·¢ ' ! k" & k! &>" _ &>k &>0­ • _k j j +: &>k • k& " · " &&>k 0 j +x ° 0+/ j ·¢ 0j " j 0& / "& 0+_ ? k ? +/ ·( ? 0j > :+! k° k ! ¢ k ° kj j k ·3 ):Z >+? k° k ! Z &>k ­ ! 0& 0­ " · j 0 €k 0j j • " ··k ! &>" _ &>k j k­ & 0+_ &>0­ • _k j j

N8 0' 3 L( { F " _( & >0j 0j / j / " ··¢ &>k j 0& / "& 0+_ 0_ " h v }

! k" ­ & +! k_ &>k ' ! +? 0_' ( k :k­ &• " ¢ k\ ­ kk( &>k ­ ! 0& 0­ " · j 0€k ? 0& >+/ &" _¢ +x ° 0+/ j " && >k " x +° k k¢ " ! k k j & kk · j k­ & 0+_j :! +• 114

? >0­ > &>k j & !/ ­ F

&/ ! k 0j • " ( k Z +! xk ­ " / j k &>k ¢ " ! k 0_& ! +( / ­ k( ( / ! 0 _'

:" x ! 0­ " & 0+_Z 0_ k 0_& k' ! 0& ¢ +: &>k ° k j j k · &>/ j ( k k =/ " ·0& ¢ +: &>k :" x ! 0­ " & 0+_ " _( +_ &>k k::0­ " ­ ¢ +: &>k 0_j _0=/ k j 3 8/ ! & >k ! • +! k Z &>k ! " & k +: ' ! +? & > +: ­ ! " ­ • j x ¢ :" & 0' / k N0_­ ·/ ( 0_' ­ +! ! +j 0+_ :" & 0' / k { 0j ° k ! ¢ 0• k ° " ! 0" & 0+_ +: ° kj j k ·

:" 0 ·/ ! k 0­ > &>k :" 0 ·/ ! k " j ·+? ·¢ ! 0 j 0_' k A0 _­ / x " & 0+_ k 0_0& 0" · ­ +·( F>¢( ! + +° k ! F+? _ 0_ 80' / ! k ; 3 ) ­ " _ / j k &>0 j (0 " ' !" •

k ° kj j k · ? >k _ 0&:0! j &k_& k! j j k ! ° 0­ k 0j ( k0 j j& " ' k 0j &>k ! kj k ° kj j k · 8" x ! 0­ " &+! 3 v >0 ·j & &>k ° " ! 0+/ j ! k' / ·" & 0+_j " _( ­ +( k j 0_& ! +( / ­ k ­ >k ­ • j " _( " / ( 0 & jZ& >k j k 0_ _+ ? " ¢ ! k ·0 k° k & >k :" x ! 0­ " &+! +: &>0 j / ·& 0• " & k ! kj k :" x ! 0­ " & +! • / j &0­ > • / j & x k k / & 0·0 & ¢ " _( &+ &>k ° " ! 0+/ j 0_j k / & 0·0 & ¢ " _( x¢ &>k ·0 ­ k_j 0_' " / & >+! 0& ¢3 4 _ 0• k kj & " x ·0 j >• k _&x¢ &>k :" x ! 0­ " &+! +: " ? k ··F( k:0_k( " _( " ­ +• • k ! ­ 0" · " _( k :" x ! 0­ " & +!

&>k =/ " ·0 & ¢ " j j / ! " _­ k +j k ! kk h +j j 0x 0 ·0 & 0kj K THRH: POSSIBILITIES,

■

)3

Nc t h ing

2.

Ductile Failure (like tin cun) St.re s e :> Yield Stress

3.

Brittle Failure (l:ke glass) a > Critical Crack Length N" ­ R

·

:· "

►

,:3Z; it

,Z

;/H;&° H°°\YkW *> wk ekYVTkXXVK*V; J/H;&°

. 33 3. 33333 0

•••

£z dVYk. * 5µkH/°7

PX

• 30

•z

3

wHwVXV*µ WkKk_W° >_G

wHwVXV*µ WkKk_W° >_G

Jm• F RvIbBJvdBm- hbmJFeeFe aMvPBd( veeMbv-JF Md B-ehFJdBm- • MbB-A RvIbBJvdBmhbFeeMbF dFed

B-BdBvP aMvPBd( mR yFeeFP • FAbv• vdBm- mR UvdFbBvPe hbmhFbdBFe yFeeFP mhFbvdB-A NBedmb( RvdBAMF JbvJO AbmfdN Pvfe B-TeFbyBJF B-ehFJdBm-

RvIbBJvdmb4e bFehm-eBIBPBd(

MdBPBd(4e bFehm-eBIBPBd(

PBJF-eB-A vMdNmbBd(

Fig. 3.

Vessel Failure Probability as a Function of Vessel Life

PPW

p " x ·k L3 8" 0 ·/ ! k h ! +x " x 0 ·0 & ¢ 0j &>k h ! +( / ­ &+: &>k 8k k_( +: 0& j ·0 :k 0j ( k 0 j& +! ¢Z 0_­ ·/ ( 0 _'

&>k :! k=/ k _­ ¢ " _( • " ' _0& / ( k +: &>k 0­ > 0& 0j k\ 0j j & " ' k 0j &>k ! kj k o & 0·0 & ¢ ? >0­ > +? _j " _( +k ) 0­ k_j k &+ +k !k" ­ & +! k / & 0·0 & ¢ 0j / j / " ··¢ " ·j + &>k +? _k ! +: &>k ° kj j k · " _( " j j / ­ > 0j ! k j k ( k j 0' _ j k

N" R &>" &&>k j & ! kj j ! kk j k ­ +• " &" ( k =/ " & k j / +! 0& ¢ !k' " ! ( 0_'

&>k

0_& k' ! 0& ¢ +: &>k ° kj j k ·3

)r h n)5 4 p )c e s 8c }

p d 7 )r h c } p )e 9 5 c o e p } b

v >" &" ! k & >k 0• < ) 0­ " & 0+_j ! k·" & 0_'

&+ &k­ >_+·+' ¢ & ! " _j :k! :+! " ­ +/ _& ! ¢ 0• k ­ +_j & !/ ­ & 0+_ " j k &>k ! kj k ° kj j k · • / j &! kj &? 0& > &>k :" x ! 0­ " &+! ? >+ 0_ • " _¢ ­ " j k j ? 0PP x k ·+­ " & k( 0_ " ( 0::k! k _& ­ +/ _& ! ¢3 4 &&>k j & " ! &+: &>k +k ! kj k j " :k& ¢ +: &>k

+k / & 0·0 & ¢ " _( ·0 ­ k _j 0_' " / & >+! 0& ¢ 0_ &>k 0• k ( k j 0' _ & ! " _j 0k _& j N/ 0­ > &>k :" & 0' / k ­ ! " ­ • ' ! +? & > ­ " ·­ / ·" & 0+_j " _( &>k ! k:+! k &>k j " :k&¢ +: &>k ° kj j k · ( k " ! " ­ & k! 0 j& 0­ xk >" ° 0+/ ! +: &>k k·k ­ & ! 0­ " · j / " _0­ j 119

N( R &+ ( k­ 0( k +_ &>k :!k=/ k_­ ¢Z k\ & k_&" _( _" & / ! k +: &>k ° kj j k· 0_Fj k! ° 0­ k 0_j k / & 0·0 & ¢ " _( &>k ·0­ k_j 0_' +! ' " _0€" & 0+_ 0_ &>k !k­ k0° 0_' ­ +/ _& !¢ ? 0 ·· ? " _&&+ ­ +_j 0( k! &>k !+·kj &>k¢ ? 0 ·· &" • k 0_ &>k k\ " • 0 _" & 0+_Z j / ! ° k0··" _­ kZ " / ( 0& 0_' " _( ! k° 0k? +: " ·· j& " ' kj +: ° kj j k· j / " _' kj 0_ &>k ­ +( kj " _( !k' / ·" & 0+_j +: &>k k\ !" & kj " _( +_ • " & k! 0" · 0­ > +/ ! 0' >·0' >& k( ? " j &>k _kk(

&+ &! " _j :k! &>0j j +F­ " ··k( A? " & ­ >( +' A &k­ >_+·+' ¢ =/ 0­ • ·¢ " _( k::0­ 0k_& ·¢ :!+•

&>k k\ k !kkj k • " & & k! j " j &>k¢ !k·" & k &+ o 3 a 3 ­ +_( 0& 0+_j >" j _+? xkk_ k( " _( ­ +_­ ·/ ( kj ? 0& > " _/ • xk! +: j _+·+' ¢ & ! " _j :k! ? >0­ > " ! k " 0_& k' ! 0& ¢ " _( !k·0" x0·0& ¢ _+&+_·¢ / _( k! _+! • " · +k k" ! ·¢ " ··Z Pj &c ­ & +xk! P2WU N" ° " 0·" x ·k :! +•

PLX

dr sc R 3

DEVELOPING SEISMIC ANALYSIS TECHNIQUES AND REQUIREMENTS FOR MUL Tl-COUNTRY APPLICATIONS G. J. BOHM AND C. W. LIN

Westinghouse Electric Corporation Pittsburgh, Pennsylvania 15230

USA

1.

INTRODUCTION

In this paper, the latest developments in the seismic design of nuclear power plants are discussed with the purpose of achieving the transfer of nuclear technology to nuclear industries in developing countries. Advancements in the seismic qua I ification methods and techniques of a nuclear power plant are rendered. Current information to facilitate a broad interchange of concepts and ideas with the prevailing trends in the nuclear industries outside of the United States are also given. The need to design nuclear power plant faci I ities to stringent seismic requirements was first brought to focus in 1963 when the then United States Atomic Energy Commission (US AEC) published the TID-7024 report, "Nuclear Reactors and Earthquakes". (l)

In

addition to discussing major aspects of world seismicity and structural design considerations of basic reactor types and potential damage due to earthquakes, the report suggested using the response spectrum technique in the form of the Housner response spectra (2) for the seismic response calculations of simple types of structures, systems, and components. A response spectrum represents the maximum response of a series of single degreeof-freedom systems for a given damping value versus the natural period of the system when subjected to base motions resulting from a specific earthquake. To be meaningful for design purposes, the maximum response of such a simple structure would have to be made for several different recorded earthquake accelerograms. The accelerograms for moderate distances from the epicenter of large magnitude earthquakes, however, may be treated as a stationary random process. By using four strong motion ground accelerations with two components each (N-S and E-W directions), Housner constructed a set of relatively smooth response spectra. There is, thus, a low probability that the maximum response of a structure at a natural period would exceed its response spectrum value. These smoothed response spectra have been used as the design basis for a number of nuclear power plants. A subsequent development in response spectrum curves has led to the use of Newmark's response spectra, (3) for many plants. Although a number of plants in the eastern United States have used less severe design response spectra recommended by the US AEC, further modifications suggested by Newmark (4) resulted in an increase in the response in higher· frequency range (between 20 to 30 Hz) . Recent pub I ications ( 5 • 6' 7) have recommended

121

Regulatory Guide 1.60, ( 5 ) on the subject of design response spectra, is tailored to New m ark's recom m endation. ( 7 )

new design ground response spectra.

The design response spectra are defined at the free field of a site.

Scavuzzo, in 1967, ( 9 )

suggested that the presence of a m assive structure such as a nuclear power plant may have a profound effect on the motions of the soil im m ediately surrounding the foundation mat.

This

effect was further studied by Lin, (lO) w ith the structural dam ping included and using a deconvolution technique.

From the data of these studies, it w as concluded that when free

field ground response spectra are used as the design basis, over-design is to be expected. This phenom enon has been extensively researched using the im pedance analysis technique, 17 181 (ll, 12 · 13 ) finite elem ent model, (l 4 , l S) and a com parison of both. (l 5 , • As a result, soil structure interaction is now the accepted tool to determ ine the base mat motion of a structure, unless the foundation soil m aterial can be considered as rigid. In conducting a soil structure interaction study, the design ground response spectra will have to be converted to equivalent tim e history motions.

The m ost com m on m ethod is

to generate the synthesized tim e history motion using spectra raising and suppressing tech9 niques. (l l W ith data from a real earthquake as input, spectral raising is accom plished by adding to the original tim e history a function at the frequency of interest w ith a phase angle such that the response spectral value wi II be increased to a desired am ount.

The tim e w hen

the m axim um vibration occurred w i I I be the sam e.

In this w ay, the characteristics of the

required tim e history w ill only be slightly altered.

Spectral suppression can be carried

out by passing the tim e history through a I inearly dam ped osci I lator connected in a series w ith a second dam per.

This dam ping arrangem ent w ill be able to reduce the response

spectra value, locally, at the natural frequency of the oscillation to the desired am ount. W ith the tim e histories so generated, the motions at the base m at can be determ ined by conducting the dynam ic response analysis with a sim ple structure and detailed soil model. Such m otions, in turn, becom e the input to a refined building m odel. Either alone, or coupled w ith the reactor coolant system s, they generate either floor response spectra or are used to conduct a nonlinear tim e history analysis for the qualification of system s and 2 com ponents. ( 1) Such a coupled analysis has been shown to produce the most realistic loads. For instrum entation and control equipm ent w here analysis using a m athem atical m odel is not feaslbteb r cause the design displays strong nonlinearities such as friction parts and 22 gaps, tests m ay have to be conducted in lieu of analysis. To this end, IEEE-344( ) has been accepted by the industry as the standard docum ent to be follow ed. of the IEEE-344 requires that a narrow band input be used.

The 1971 version

The 1975 version required,

on the other hand, a broad-band multi-directional input such that the test response spectra w ould envelop the required floor response spectra at the equipm ent support location.

23

This developm ent has produced new er and better test machines. B

t

Input values have been related to the design ground response spectra consisting of one response spectrum curve for each damping value. The magnitude of a response spectrum at a given frequency is, except in the rigid frequency range where no spectral am-

122

plification is noted, inversely proportional to a function of dam ping value.

For lack of

data and for the sake of conservatism , extrem ely low dam ping values were used for sys24 25 tem s and com ponents until the sam e results w ere published iri 1973. ( • l Additional

test data accumulated since that time indicates that even the damping values recommended 26 in Regulatory Guide 1. 61, ( l which are largely based on early testings, are too low. In addition to the effort of developing more acceptable damping values so that system and structure response can be determined accurately, it is essential that serious considerations be given to factors affecting other design loads which may have to be combined with the seismic loads in the design of nuclear power plants.

2.

SEISMIC INPUTS USED FOR THE DESIGN OF NUCLEAR PLANTS

The following discussion reviews the most important considerations influencing the design of a nuclear power plant when the seismic loads need to be established for the qualification of the plant.

2. 1 Design Ground Response Spectra The response spectrum developed for a historical earthquake consists of peaks and valleys. Shown in figure 1 is a 2 percent damping response spectrum for the El-Centro 1940 earthquake E-W component, normalized to 0.2g maximum ground acceleration. The peaks and valleys of the response spectrum .are different for each different earthquake. Sin.ce each ground motion may be viewed as a sample from a population of random functions, it is meaningful to obtain

a

statistical average of the response spectra for a number of strong

earthquakes. Generally, such an average response spectrum is a smooth curve. One of 2 the earliest attempts in creating such spectra was made by Housner. C l The response spectra generated by Housner were averaged over four strong earthquakes using two components of each (that is, a total of eight seismic accelerograms) . Housner's response spectra have been used as the design basis for a number of nuclear power plants. Because they are the average of four strong earthquakes that have occurred in the Western United States, however, there are concerns that they may not be adequate to represent all typical strong earthquakes, which may occur in the future. Consequently, an additional study was carried out. (3) A set of response spectra generated using the El-Centro earthquake was later recommended for the nuclear power plant design. These spectra appear to be close to the upper bound of the El-Centro response spectra. Because of this severity, it was decided that they would not be suitable for the eastern United States. A modified set was recommended subsequently by the US AEC for certain applications.

123

,oi •

10:

•

r-----,------------

l--------t----- i.-:-7,--------1

E W D

. 10-1 I 10-

1

11

l / I

I

2

456789100

3

till 2

111111

I

I

3

456789101

I 2

3

11 ltll

456189102

847c o 7e5 b ]dLt

Fig. 1.

El Centro E-W Time History Response Spectrum for 20% Damping

Other changes by Newmark resulted in the increase of the amplification at frequencies 4 between 30 Hz and 20 Hz, for a 2 percent damping spectrum. ( ) These changes were sub7 stantiated in part in a proposal by Newmark, ( ) which was based on the studies made by Newmark and Blume. {S,6) Regulatory Guide 1.60(8) is tailored to Newmark's recommendations. (7)

Figure 2 shows the comparison of the Regulatory Guide 1. 60 spectra, the 2 Newmark spectra, (4) the previous AEC minimum criteria, and the Housner spectra. ( )

All these spectra are for 2 percent damping and a normalized maximum ground acceleration of 1. 0g. Only a minor difference exi s ts between recommendations. (S' 81 Large differences do exist, however, between the response spectra recommended by Housner and those recommended by Newmark and Regulatory Guide 1. 60. While the Housner spectra were averages of four strong earthquakes with two components each, the Regulatory Guide 1. 60 spectra were obtained using the one sigma plus the mean value for a total of 33 recorded earthquakes. Consequentlv , the latter corresponded to a much higher confidence level, that is, their probability of exceedance is much less. In fact, it can be calculated that the probability for future earthquakes not to exceed the Regulatory Guide 1. 60 is approximately 84 percent rather than Ej + percent for the Housner spectra. Since its development, the shape of Regulatory Guide 1. 60 response spectra has been fol lowed fairly closely even in generating the design response spectra for nuclear power plants outside the United States. Shown in figure 3 is such a typical set of response spectra. By comparing figure 3 with the regulatory guide spectra shown in figure 2, the difference occurs only at and below 2. 5 Hz. The response spectra for figure 3 have not peaked until 1. 7 Hz. The nuclear power plant building design usually possesses a high degree of filtering.

124

p >k k_k! ' ¢ ­ +_& k_&+: &>k ' ! +/ _( • +& 0+_ 0j ' k_k! " ··¢ 0_ &>k :! k=/ k_­ ¢ !" _' k ·+? k! &>" _ &>k " ° k :" 0 ! ·¢ ­ +• k :/ _( " • k_& " · x/ 0 ·( 0_' _" & / ! " · :! k=/ k_­ 0kj xk xk·+? L3 D d €3 8/ ! &>k! • +! kZ k=/ 00' > _" & / ! " · :! k=/ k_­ 0kj 3 (i t 5 +_j k=/ k_& ·¢Z &>k ( 0::k! k_­ k 0_ :0' / ! k L 0_ &>k ! kj k k::k­ &+: j +0 ·F j& !/ ­ & / !k 0_& k!" ­ & 0+_3 p>kj k " ! kK PR 0• 3 p>k / j k +: &>k 0• k j 0 • k j +0 · &+ &>k / j k +: :!k=/ k_­ ¢ ( kk j k·k­ & 0+_ +: &>k j & >" &&>k :!kk :0k·( • +& 0+_ N& >k j k ° " ! 0+/ j ·k° k·j +: &>k x+/ _( " ! ¢3 4 :0_0& k k·k• k_&" _" ·¢j 0j ­ " _ & >k_ xk ­ " ! ! 0 k( +/ &&" • 0 _' 0_& + ­ +_j 0 ( k! " & 0+_ &>k ( ¢_" • 0­ !kk j +0 · • "& k! 0" · k ° " !¢0 _' & 0• k >0 j& +!¢ 0_k ( " • k / j k +: &>k :0_0 & k k·k• k_&j +0 · j& !/ ­ & / ! k " _" ·¢j 0j 3 8+! 0_j & " _­ kZ & >k!k 0j +_k k / j k +: k! ? >0­ > ( 0j ­ / j j kj &>k j 0• ! kk ( 0 • k_j 0+_" ·

• +( k· ? >0­ > 0j x" j k( +_ &>k ° 0j ­ +/ j x+/ _( " ! 0kj &+ j 0• / ·" & k &>k ? " ° k • +& 0+_j k ( 0 ! k­ & 0+_ +: &>k 0_k x" j k • " &&+ :/ ! & >k! " _" ·¢€k &>k ­ +• k ­ +• " ! " ­ &k! 0j & 0­ j +: &>k ­ +• k =/ " ·0 :0­ " & 0+_ ­ " _ xk ( +_k x¢ " _" ·¢& 0­ " · &++·j +! x¢ & kj & 0_' k • +j &!k·0 " x ·k =/ " ·0 :0­ " & 0+_ &k­ >_0=/ k 3 ±3 P s & ! / ­ &/ ! kj Z s ¢j & k• j Z " _( 7 =/ 0kj ­ " _ xk &" • k_3 p >k !kj :+! &>k =/ " ·0 :0­ " & 0+_ " _ +_k >/ _( ! k( h ! kj j / ! 0€k( v " & k! } k" ­ & +! Nh v } R _/ ­ ·k" ! &>k ·k° k· +: +_k j& " _( " ! ( ( k° 0" & 0+_ k • k" _ " • k ­ +_& "0 _• k_&x/ 0 ·( 0_' 0_& k! 0+! j& !/ ­ & / ! k3 p" x ·k P j >+? j &>k x/ 0 ·( 0 _' " • k k j =/ " ! k !++&+:

&>k !kj +_" _&! kj 0 j& +! ¢ • +& 0+_j " &" ·· " kj k k·k° " & 0+_j 3 4 j " !kj / ·&+: &>k x/ 0 ·( 0 _' :0·& k! 0 _' k::k­ & Z& >k :·++! !kj +? j j +• k j & "& 0 j& 0­ " · ("& " !kk h v } k &" x·kZ &>k • k" _ ° " ·/ k +: &>k

k=/ 0 k x/ 0 ·( 0 _' " • " & & >k • +& 0+_j " !k ·kj j !" _( +•

" j 0_ &>k ­ " j k +: & >k ' !+/ _( • +& 0+_j 3 8+! &>k " j xkk_ !k­ +• • k_( k( &+ ­ +• k _kk( k(

( +­ / • k_& "& 0+_ " _( >" j xkk_ ­ +! ! k­ & ·¢ ° k! 0:0k( " _( =/ " ·0 :0k( 3 )_ q orne ­ " j kj Z ( / k &+ " ­ " j kZ &>k / j k! ? 0 ·· xk +! kkF( 0 • k_j 0+_" · 0_ _" & / ! k3 p >k k" &­ +_( / ­ & 0+_ " _" ·¢j 0j Z " _( j & k" ( ¢j & "& k >¢( ! " / ·0­ " _" ·¢j 0j 3 p+ ­ +• k v 7 5 4 e k =/ " • k ' ! +/ _( • +& 0+_j Z j k° k! " · :k" & / ! kj +: &>k ­ +( k ­ " _ xk / j k( ( kk " _" ·¢j 0j +: ·" ! ' k " _( ­ +• " ? " ¢ &>" &" ( ( 0& 0+_" · ( 0j ­ ! k& k Ej & ! / ­ &/ ! " · k·k• k_& j ­ " _ xk " ( ( k( &+ &>k k " _0­ " · 7 _' 0_kk! j

N4 s r 7 R " _( &>k e / ­ ·k" ! } k' / ·" -

&+! ¢ 5 +• • 0j j 0+_ Ne } 5 R !k' / ·" & 0+_j j & ! kj j &>k 0• k j k0j • 0­ " _" ·¢j 0j >" j xkk_ !k0_:+! ­ k( x¢ k\ & k_j 0° k ( +­ / • k_& "& 0+_ 0_ &>k :+! •

+: ° k! 0 :0­ " & 0+_ " _( ( k• +_j & !" & 0+_

• " _/ " ·j 3 p>k ° k! 0:0­ " & 0+_ " _( ­ " ·­ / ·" & 0+_j Z " _( .+! Q

x3

" ·& k! _" & k ° k! 0:0k( " _( =/ " ·0 :0k( ­ " ·­ / ·" & 0+_" · • k& >+( j Z " _( .+! Q

­3

!kj / ·& j +: +& >k! ° k! 0:0k( k( ( " & " " _( ­ +! ! k·" & 0+_Z " _( .+! Q

>3

!kj / ·& j +: j & " _( " ! ( +( j +: " _" ·¢j 0j 3 p>k ° k! 0:0­ " & 0+_ ( k' ! kk +: ­ +_:0( k_­ k " _( ( k• +_j & !" & kj &>" &

&>k ­ +( kj ­ " _ k j k0j • 0 ­ " _" ·¢j 0j 0­ > & >k¢ " ! k / j k( 3

; 3; 6 " • k j ¢j & k• 3 p>k k::k­ &+: & >0j

&¢" j " _ 0• k j & !/ ­ & / ! " · ( " • 0­ > 0j ­ " / j k( x¢ • +& 0+_j 0_ " :·/ 0( Q +! ( / k &+ 5 +/ ·+• x ( " • k j & !/ ­ & / ! " · j ¢j F

P; D

&k!_ ­ +/ ·( ' ! k" & ·¢ " ::k­ &&>k ! kj k• " & 0­ " · &k! • j Z " _ " _" ·¢& 0­ " · k\ k ( 0::k! k_&_" & / ! kj +: ( " • k !k" · j & ! / ­ &/ ! k ­ " _ xk " 0­ > 0_( 0­ " & k &>" &k_k! ' ¢ ·+j j ? 0)) 0_­ ! k" j k ? 0& > &>k j & ! kj j +! ( 0j k k=/ 0° " ·k_&° 0j ­ +/ j ( " • 0­ > ? 0 · ) • " • k &>k k=/ " & 0+_j +: • +& 0+_ _+_·0_k" ! 3 p + " ° +0( &>0 j 0_­ +_° k_0k_­ kZ ­ +_j k! ° " & 0° k • +( " · ( " • k! ° " ·/ kj 3 ( " • k! &>" _ &>k +_k ( k·0_k" & k( 0_ &" x ·k ; " ! k " ··+? k( x¢ } k' / ·" & +! ¢ 9 / 0( k 1. 61. )&j >+/ ·( xk !k­ +' _0€k( &>" &&" x ·k ; ° " ·/ kj " ! k kj & " x ·0 j >k( x" j k( +_ &kj &( " & " ­ +• k k" ! & >=/ " • k • +& 0+_j Z ( " • k! 3 p >0j ? +/ ·( k ­ " j k +: :" / ·& k( ­ +_( 0 & 0+_j 3

p" x·k ; 3

6 " • k :/ k· " j j k• x ·0kj Z " _( _/ • k! +/ j ­ +• k_ " j ¢j & k•

+! k=/ 0k• " & 0­ " ··¢ • +( k·k( Z j k0j • 0­ " _" ·¢j 0j

0j ­ +_( / ­ & k( 3 p kj & 0_'

0j / j / " ··¢ k• &kj & 0_' +: &>k k·k­ & ! 0­ " · k=/ 0k • +j &/ j k:/ · ( +­ / • k_&&+ xk :+··+? k( 0_ ? ! 0& 0_' " &kj & 0_'

138

procedure. The 1971 version of IEEE-344 requires that a narrow band Input such as a sinebeat input be used. Figure 9 shows such a sine beat. The sine beat generally consists of a total of five beats with a pause in between beats to insure that the equipment is not overly tested. Each beat has five to ten cycles of motion which either has the frequency of the equipment or a preselected frequency to assure a proper test. The test input magnitude is taken from the envelope of floor accelerations. This procedure assumes that the peak of the appropriate floor response spectra coincides with the predominant equipment frequency and is, therefore, a very severe test procedure. Figure 10 shows tne comparison between the response magnitudes of a single frequency sine beat test versus those from random vibration input and Taft earthquake motions. In all cases, the sine beat test is more conservative than random vibrating testing by a wide margin. N- · Contrary to a narrow band test requirement, the 1975 version of the IEEE-344 Standard requires that a broad band multi-directional input be used because of the recognition, that the concrete support structure response is not necessarily narrow band and uni-directional. In the board band case, it is feasible to use a composite time history consisting of several sine beats with sufficient frequency content to have a test response spectrum enveloping the required response spectrum. Such a composite sine beat is essentially broad band (figure 11) . On the other hand, its vibration intensity at any frequency in the composite sine.beat motion is equivalent to a narrow band. Figure 12 shows that the response spectrum obtained from this motion matches the essential portion of a smooth response spectrum. Present testing techniques require shakers of special capabilities, and the control of .the input loads is of paramount importance. In regard to mechanical shakers, there are requirements of multi-direction input and certain specifications where not only maximum acceleration and displacements are established. but also admissabie velocities are of interest. Frequently, the selection of the test laboratory involved in the qualification program is of particular importance to ensure that the input loads established by the test reports can be reached as defined. Table motion must be controlled to ensure that the inputs are as desired and to avoid incorrect information from the electronics.

Fig. 9. Sine Beat Input for Testing 139

I

SINE BEAT

·~

I

I I

S'llo BUILDING OAMPINO

) ~ frfORMAL DAMPING RANGE

I ---f

) )

) )

)

)

I 10 EQUIPMENT DAMPING ,11 IPERCENTI

Fig. 10.

Conservatism Inherent to Sine-Beat Waveform (Ref. 23)

10

20

25

JO

JS

TIME ISt1.0~0S

Fig. 17.

A Typical Composite Sine Beat Displacement Time History Motion

o _( k! k :·++! !kj k ' k_k! 0­ =/ " ·0 :0­ " & 0+_ +: &>k k=/ 0 k &0• k >0j & +! ¢ 0_k ­ +• kj k !k=/ 0 ! k• k_& j ­ " _ xk • k&x¢ " ·0 • 0& k( _/ • x k! +:& kj &·" x+! " &+! 0kj " _( / _( +/ x& k( ·¢ " ! k " ( 0::0­ / ·& ¢ &>" &? 0 ·· _kk( &+ xk +° k! ­ +• k x¢ ­ +/ _& ! 0kj j & " !& 0_' &+ xk­ +• k 0_° +·° k( 0_ _/ ­ ·k" ! k_k! ' ¢ " _( ? 0& > ( kj 0! kj &+ =/ " ·0 :¢ &>k k=/ 0_+·+' ¢ &+ ­ +• _+·+' ¢ & !" _j :k! " ° k ? +!• k( 0_ " j !k­ +' _0 €k( &>k 0_& k!kj &+_ &>k k & !" _j :k! +: _/ ­ ·k" ! &k­ >_+·+' ¢ " j " " ° k xkk_ ( k° k·+_+·+' ¢ " _' k " ' !kk• k_& Z &>k • " _/ :" ­ & / !0 _' k\ ­ >" _' k " ' !kk• k_&" _( &>k &k­ >_+·+' ¢ ( k° k·+" j k_& k!k( 0_& + _/ ­ ·k" ! j ¢j & k•

&k­ >_+·+' ¢ k\ ­ >" _' k " ' !kk• k_& j ? 0& >

­ +• k :0! j &+: &>kj k " ' !kk• k_& j ?"j j0 ' _k( ? 0 & > 4 · ·' k• k0 _k 7 ·k• & !0­ 0& "& j F9 kj k· ·j ­ >" :&4 9 pk·k:/ _• k_ N4 7 9 R Z 0_ 9 k!• " _¢Z 0_ P2US3 )_ P2UWZ j 0• 0 ·" ! " ' !kk• k_& j ? k!k !k" ­ >k( ? 0 & > d0 & " ­ >0 Z n& ( 3 " _( p+• ¢+ s >0 x" / !" 7 ·k­ & !0­ 5 +• _+·+' ¢ k\ ­ >" _' k " ' !kk• k_&? " j " · j + j 0' _k( ? 0& > s ? k( k_E j 4 s 7 4 F4 pc r

0_ P2WS3 )_ &>0 j ­ " j k ­ +_( 0 & 0+_j ? k!k j +• k? >" &( 0 ::k!k_&j 0 _­ k 4 s 7 4 F4 pc r

>" ( 0_( k" _' k " j j +­ 0 "& kj ( k° k·+k0 ! 0_0 & 0" · - v } _+·+' ¢3

There are several important characteristics of these technology exchange agreements. First of all, the agreements are completely symmetrical with regard to the aval labi lity of technical information. General Electric's technical information is made available to its associate and the associate's information is made available to General Electric. In the early stages of most agreements the flow of technology is almost entirely from General Electric to its associate. General ElectrlcPs Boiling Water Reactor technology has been developed over the last two decades through the efforts of many thousands of dedicated nuclear engineers and through the.expenditure of many hundreds of millions of dollars. Over one hundred General Electric Boiling Water Reactors are in operation or under construction in nine countries. There is, therefore, an immense technology base on which General Electric's international technology associate can draw as he embarks on the development of his own Boiling Water Reactor program.

In the course of time, however, as

the associate's technological capability increases there is an increase in the flow of technology from his organization toward General Electric. Such technology backflow is an important characteristic of General Electrlc's technology exchange programs and helps knit together all of the world's Boiling Water Reactor suppliers into an increasingly strong technical family. Basically, the technology exchange .agreement provides information on what General Electric's Bolling Water Reactor product is and how General Electric goes about designing that product. Al I the information which General Electric uses to design its reactors is made available under the agreement. There are three principal mechanisms used to bring about this technology transfer;

training, documents and consultation.

The training program consists of the assignment of our associate's engineers to General Electric's engineering headquarters in San Jose, California. Training assignments usually last six months or longer. While in residence in San Jose, the trainee is given a carefully planned series of assignments actually working in General ElectricPs engineering organization as a member of the General Electric design team. The assignments are planned by General Electric's training specialists in consultation with both General Electric's engineering management and with the technical associate's management. At the end of such a training assignment, the trainee is qua I ified to return to his company and immediately begin a responsible Boiling Water Reactor design assignment. To date, 201 engineers from our associates' companies have completed or are in the process of completing training assignments with General Electric's Nuclear Energy Divisions. The second transfer mechanism is documentation.

Included in this category are all of

the drawings, specifications, technical reports, computer programs and other written material used by General Electric in the design of its nuclear power plants. As you can well imagine, this runs into thousands of documents. A complete package today would probably comprise well over one hundred thousand individual documents. An additional twenty thousand documents are issued annually. The third transfer mechanism is consultation. General Electric' s technical specialists are available to consult with our associates to explain and interpret the technical 148

documentation and to make sure that his engineers know not only how General Electric goes about designing its reactor but also why. At this point, it is important to explain another very important feature of the technology exchange agreement. General Electric makes its technology available to its associate but it is then entirely up to that associate as to how such information is used. General Electric ties no strings to its technology. The associate is free to incorporate General Electric designs directly in his Boiling Water Reactors, or he may modify the General Electric design, or he may ignore it completely and develop his own design solutions. General Electric will consult with him on the General Electric design but does not assist in the application of the General Electric design to the associate's projects. The associate is an independent designer and is fully responsible for his own designs.

(General Electric may also enter into

joint ventures with its associates to design specific plants. These joint ventures are described below). It is therefore extremely important to General Electric that the technical associate be a competent, fully qua I ified nuclear plant designer. All of our associates' Boiling Water Reactors must operate safely and reliably. The only means we have of assuring this is to select fully qualified organizations as technical associates and then to make certain that an effective technology transfer program is carried out. 2.

MANUFACTURING EXCHANGE AGREEMENTS

Manufacturing exchange agreements provide for a flow of manufacturing technology analagous to the flow of design technology under the technology exchange agreements. General Electric's nuclear manufacturing scope includes such critical nuclear system components as control rods, control rod drives, reactor vessel internals, and hydraulic control units. General Electric also manufactures nuclear instrumentation; sensors as well as nuclear control panels and racks. Nuclear fuel cycle manufacturing technology includes UF to / + conversion, zirconium tubing, fuel bundle hardware and fuel bundle fabrication 6 2 from pelletizing through bundle assembly. Under the manufacturing exchange agreement, General Electric provides al I the technology necessary to manufacture these products. Information on manufacturing processes, production equipment, qua I ity control procedures and equipment, and so on is provided under these agreements. Again the three transfer mechanisms of training, documentation and consultation are used. Trainees are given a series of rotatingassignments within General Electric's manufacturing organizations. Most of these assignments are at General El ectr ic+s nuclear manufacturing center in Wilmington, North Carolina. So far, some 40 engineers, specialists and managers representing five companies have received comprehensive training assignments at General Electric's Wilmington and San Jose facilities. Nuclear manufacturing processes are still undergoing evolutionary change. In order to keep our associates informed of such changes as they are planned and instituted, General Electric publishes periodic information briefings and holds technology seminars in both General Electric shops and in our associates' factories.

149

9 k_k!" · 7 ·k­ & !0­ kj k " ' !kk• k_& j _+·+' ¢ k\ ­ >" _' k " j j +­ 0" & kj " ·j + >" ° k • " _/ :" ­ & / !0 _' k\ ­ >" _' k " ' !kk• k_& j 3 )_ ( k° k·+k x" ­ • :·+? +: &k­ >_+·+' ¢ 0j " _ 0• ­ +• k( 0_ 9 k!• " _¢Z t " kj k ~+0_&• " _/ :" ­ & / !0 _' ­ +• k ­ " k ­ " j k +: t " " j !k­ k_& ·¢ xkk_ !k( / ­ k( &+ +_k • " _" ' k! " _( &? + ­ +_j / ·& " _& j 3 4 :& k! &>k 0_0 & 0" · j & " !& / < kj k &>!kk ­ +• +/ j " _( :/ k· x/ _( ·kj &+ ( " & k3

;3

5 4 s 7 d )s pc } b F } 7 nc 4 6 8o 7 n 8c } s h 4 )e

} k·+" ( :/ k· :+! &>k e / ­ ·k_+! !k" ­ & +! 0_ s k &k­ >_+·+' ¢ & !" _j :k! " ' !kk• k_& j ? >0­ > 9 k_k!" · 7 ·k­ & !0 ­ " _( 0& j +° k!j ±" j " j j +­ 0" & kj >" ° k / j k( 3 7 • +j k ­ >" ! & k! 0_­ ·/ ( kj & >k ( kj 0' _ " _( • " _/ :" ­ & / !k +: :/ k· :+! s " ° k " _ " ' ! kk• k_&? >0 ­ > ­ +° k!j &>k &!" _j :k! +: - +0 ·0 _' v " & k! } k" ­ &+! :/ k· • " _/ :" ­ & / !0 _'

&k­ >_+·+' ¢ " _( !k·+" ( :/ k· ( kj 0' _ " _( ­ +!k • " _" ' k• k_& &k­ >-

_+·+' ¢3 7e o s 4 !+/ ' > " ' ! kk• k_& j ?0 & > v kj & 0 _' >+/ j kR 3 7 e o s 4 >" j " _ k\ k )_( 0" _ 6 k+? _ >k! k 3

e / ­ ·k " ! s ¢j & k• j 6 k j 0' _ 5 +_° k_& 0+_" · s ¢j & k• j 6 k j 0' _

} 4hh )

} 4 h h 11

5 " _" ( 0" _

5 " _" ( 0" _

5 " _" ( 0" _

)_( 0" _ :0! •

5 +_j / ·& " _&

/ < x¢ &>k 5 " _" ( 0" _

j k&

5 +_j / ·& " _&? 0& > " • 0_+! 0& ¢ 0_& k! k j & 3 5 +• • 0 j j 0+_0_'

5 " _" ( 0" _

)_( 0" _

5 " ·" _( ! 0"

5 " _" ( 0" _

)_( 0" _

7 _( s >0k ·( j

5 " _" ( 0" _

)_( 0" _

- +0 ·k ! j

5 " _" ( 0" _

)_( 0" _

8/ k ·

d " ·: 5 " _" ( 0" _ d " ·: )_( 0" _

)_( 0" _

p / ! x 0_k

- ! 0& 0j >

- ! 0& 0j >

8/ k ··0 _' r " ­ >0_k j

5 " _" ( 0" _

)_( 0" _

7 ·k ­ & ! 0­ " · 7 =/ 0 k _/ ­ ·k" ! j ¢j & k• j ( kj 0' _3 p >k h 4 7 5 >" j ·" 0( ( +? _ k\ & k_j 0° k 0 ­ > ? k >" ( >+k h 4 7 5 >" j j 0' _k( " ­ +_& ! " ­ &:+! &>k ­ +_j & !/ ­ & 0+_ +: " 5 4 e 6 o :/ k · • " _/ :" ­ & / ! 0_'

k ! k ·( / < x¢ " k Ax " ·" _­ k +: _/ ­ ·k" ! 0j ·" _( A x k ­ " ! ! 0k( +/ &x¢ 4 ! ' k _& 0_k :0! • j 3 p >0j !k=/ 0 ! k( &>k ­ +_­ k0­ > >" ( xkk_

157

( k° k·+ &? + ·+­ " ·

­ +• ·¢ j k +& >k! ( k" ·j ? 0& > &>k • k­ >" _0­ " · " _( k·k­ & ! 0­ " · j k! ° 0­ kj Z 0_­ ·/ ( 0_' " ·· k·k­ & ! 0­ " · k e s s s Z " _( &>k¢ >" ° k " __+/ _­ k( &>k0! 0_& k_& 0+_ &+ k k_' 0_kk! 0_' +: &>k e s s s 0& j k·:3 4 ! ' k_& 0_" 0j " ·j + ( 0 j ­ / j j 0_'

&>k " j k

+: 5 4 e 6 o :/ k· • " _/ :" ­ & / ! 0_' &k­ >_+·+' ¢ " _( &k­ >_0­ " · " j j 0j & " _­ k 0_ x/ 0 ·( 0_' " >k" ° ¢ ? " & k! _+·+' ¢ ? 0& >+/ & • k_& 0+_0_' &>k ­ / ! ! k_&k::+! & j xk0_' • " ( k &+ ·0• 0&&>k j k " ° k >" ( " j 0( k k::k­ & 0_ !k&" ! ( 0_' ( 0j ­ / j j 0+_j +_ &>k &! " _j :k! +: >" ! • ·kj j &k­ >_+·+' 0kj j / ­ > " j >k" ° ¢ ?"& k! " &_/ ­ ·k" ! ­ ++ &>k :0! j &&? + ( k° k·+k¢ >" ( xk' / _ &+ j k_( &! " ( k • 0j j 0+_j &+ 5 " _" ( " &+ j k·· 5 4 e 6 o ­ +• k _/ ­ ·k" ! + >" ° k ( k° k·+k &0• k ? >k_ &>k ( 0k " / & >+! +::k! j &>k :+··+? 0_' &>+/ ' >& j &+ " ­ +/ _& ! ¢ j k! 0+/ j ·¢ ­ +_j 0( k! 0_' kj & " x ·0 j >0_' " ·+­ " ··¢ 0_( k &>k +x~ k­ & 0° k +: +k 8" / ·&p! kk r k&>+( +·+' ¢Z ?0 & >· &>k &+< k° k_& j " j ¢j & k•

j& " !& 0_' ? 0& > &>k ( k:0_0& 0+_ +: j ¢j & k•

:" 0·/ ! k· :!+•

p " j • P3 p >k ·k° k· +: ­ +_j & !/ ­ & 0+_

0_­ ·/ ( kj +k 0_( 0° 0( / " · ­ +• 8" / ·&p ! kk 03j =/ " _& 0:0k(

/ j 0_' &>k ­ +• k / _" ° " 0 ·" x 0 ·0 & ¢ +: " j ¢j & k•

0j / j / " ··¢ ( +• 0_" & k(

x¢ PXM +! ·kj j +: 0& j • 0_0• " · ­ / &j k& j 3 - ¢ / j 0_' &>k ! kj / ·& j +: &>k 8" / ·&p ! kk 4 _" ·¢j 0 jZ &>0 j ( +• 0_" _&­ +··k­ & 0+_ ­ " _ k" j 0·¢ xk 0j +·" & k( 3 NUR 6 k:0_k v " ¢ +: } k( / ­ 0_' s ¢j & k•

o _" ° " 0 ·" x 0 ·0 & ¢3

c _­ k &>k ( +• 0_" & 0_' " ! k" +: j ¢j & k•

/ _" ° " 0·" x 0·0& ¢ 0j 0j +·" & k( ° " ! 0+/ j • k& >+( j • " ¢ xk ( k° 0j k( &+ ­ 0! ­ / • ° k_&&>k kj k ­ " _ 0_­ ·/ ( k !k( kj 0' _Z x" ·" _­ 0_' +: &>k j k\ k( / ·0 _' Z !k? ! 0& 0_' +k :" 0 ·/ ! k ( k:0_0& 0+_ :+! j & " _( x¢ j ¢j & k• j / j k( >k! k ? 0·· ( 0 ::k! :! +• &>" &+:v 4 s d FPSXX3 )_ v 4 s d FPSXXZ &>k k• " j 0j ? " j +_ / _" ° " 0 ·" x 0 ·0& ¢ " _( :" 0 ·/ ! k ( / ! 0_' • 0 j j 0+_ &0• k3 v >0·k &>0j 0j 0• 0­ > ? 0 ·· 0_:·/ k_­ k k j ¢j &k •

/ _ " ° " 0·" x 0·0& ¢3 N1R i / " _ & 0:¢ 8 " / ·&p ! k k :+! o _ " ° " 0 ·" x 0·0& ¢ 5 +_ & ! 0x / & 0+_ j 3 " _ ' k 0j =/ " _ & 0:0k ( N2R 6 k &k ! • 0• < ·k •

0_ k 5 +j & +: 5 >" _ ' k 3 k _ &" & 0+_ ° k ! j / j

NPXR )_ ­ +! k +k j ¢j & k• 3 NPPRh ·" _&s 0 j ­ +··k­ & 0+_ +: ( " & " 0j " _ +_' +0_' k::+! &" _( 0j / j k( &+ / 0j +xj k!° k( &+ :" ·· +/ & j 0( k &>k k! !+! x+/ _( j ' 0 ° k_ :+! &>k ' k_k!" · ( " & " • " ¢ 0_( 0 ­"& k >" ° k ' +_k / _( k& k­ & k( 3

ZZ / I I I )) DETERMINE

COST OF CHANGE

INCORPORATE OPTIMUM SOLUTION INTO SYSTEM

I

,,, LI

Fig. 1.

Tasks Required on System Level to Optimize the Availability and Cost of a Nuclear Plant

3.

5 c s p c h p)r )u4 p)c e s 5 d 7 r 7

; 3 P s ¢j & k•

nk° k·

p>k ­ +j &+ j ¢j & k• j ¢j & k•

0j " _ 0& k!" & 0° k k\ k! ­ 0 j k +_ &>k 0• " _' kj +_­ k &>k " ! k" ( +• 0 _" & 0_'

j ¢j & k•

/ _" ° " 0 ·" x0 ·0 & ¢ >" j xkk_ 0j +·" & k( 3

p>k j & k< J>YYk/;VHX mKk/H*V>_

I

T e;·kW\XkW m\*HQk

-..J

--.

.ts -

T e;·kW\XkW m\*HQk

-

--

±

T r°* v__\HXE UHV_*k_H_;k

-...·1

-

(J1

333

333

F qZqF ,F

333 ,F

-..J

en

T e;·kW\XkW m\*HQk

-

E T £_W v__\HX UHV_*k_H_;k

FF

F -

T R>/;kW m\*HQk T e;·kW\XkW m\*HQk

---..J

F333)

...:

171

T c/W v__\HX UHV_*H_H_;k

F

Table 2.

....,

Operating Results

AE

B*kY°

Ak_k/H*V_Q *VYk

r°* µkH/

£_W µkH/

c/W µkH/

m%k/ *·/kk µkH/°

5 LH_E r 4t•7 2T• k;E cr ) t•

5 LH_E r 4t~7 24• k;E cr '75

5 LH_E r 4to7 2T• k;E cr 4to

5LH_E r 4t•7 2T• k;E cr 4to

t.£tD

·/°E

t.zrz

·/°E

·/°E

o.oDt

£z.Dio

·/°E

c.zLo+Xzo OfN

c.zDX+Xzo OfN

£.D••+G>o OfN

D.ztX+Xzo OfN

hXH_* ;HKH;V*µ /s97

t~Ec 3

toE t 3

t£E D M

t~Ez 3

P>HW /s97

DzE t i.

D~ED i.

D~Eo i.

D~Ez 3

->_k

->_k

v;;\Y\XH*kW kXk;*/V; K>$k/

N>E >< /;kW >\*HQk

nax, /\k i w\_;\Ek k+K>°\/k

m_k

m_k

5HY/E.k_;kYk_4 >< ;>YYk/;VHX >Kk/H*V>_

5H* *·k k_W >< /k H%k/HQk k+K>°\/k 5H* *·k k_W >< E >< k_%V/>_Yk_* B T rcr ;>_;k_*/H*V>_ V_ /kH;*>/ $H*k/ m_k

->_k

->_k

IkV>$ Wk*k;*HwXk Xk%kX

IkX>$ Wk*k;*HwXk Xk%kX

IkX>$ Wk*k;*HwXk Xk%kX

IkX>$ Wk*k;*HwXk Xk%kX

2TL+XzTt µ JVC;;

2T•+V>Tt µ JVC;;

2T•+V>Tt µ JVC;;

4g.•+V>Tt µ JVC;;

24cz

µ JVC°

4g.£~

µ JVC°

24cz

µCi/s

24cz

BLJVC°

4.

SPECIAL CONSIDERATION

We applied various improvements and considerations to original BWR engineering in order to increase plant integrity of the Shimane Nuclear Power Station. The plans for high plant capacity, fuel integrity and low radiation dose which were applied to the Shimane Nuclear Power Station and the effects are summarized in Table 3. Some of the items are explained as follows. 4. 1 High Reliability of Parts and Components and Easy Maintenance In addition to the normal quality assurance (QA) the activities shown in Table 4 were carried out to parts and components of the Shlmane Nuclear Power Station to insure their reliability and opera bi I ity. Many domestic parts and components were used in the construction of the Shimane Nuclear Power Station. About 90 M of parts and components used in the construction of the Shimane Nuclear Power Station were domestically produced, This high percentage of domestically produced material has made plant maintenance easier and overall plant performance has been enhanced. 4.2 Core Management Engineering Fuel integrity is one of the most Important targets for low radiation and plant operation, Most of the avoidance of local hydride failure may be attributed to manufacturing and material control. However, the impact of Pellet-Clad Mechanical Interaction (PCMIJ may be reduced and be softened during operation so that the fuel may keep integrity for a longer duration. The impact of PCM! Is a function of maximum heat generation rate and ramp rate experienced with fuel. For this nuclear power plant. the following operating strategy was applied: (1 J Flattening the Power Shape and Reduction of Maximum Linear Heat Generation Rate (MLHGR). In spite of the design specification MLHGR 17. 5 kW /ft, the lower target MLHGR 15. 0 kW/ft is set for normal operation. This operating strategy inevitably requires flattening the power shape. For the flattening power shape in the core, control rod patterns have been generated by the optimum control rod programing code, and extra reactivity 0. 5"' 1. c %ti k has additionally been deposited to afford operating duration change and controlling power shape below MLHGR 15. 0 kW /ft. Fig. 2 shows the three year operating results of MLHGR and larger margin against the design limit. (2) Slow power ramp rate When power is raised slowly. the elongation of fuel rods is reduced. This indicates that the slower power ramp rate is the better operating condition. The power of the present BWR is controlled by means of control rod withdrawal and core flow operation. This is the 173

reason why control rod withdrawal usually induces a large local power change, because the impact of PCMI is usually smaller at lower MLHGR; it would be better to complete necessary amount of control rods at lower MLHGR as possible. Therefore a core flow operation which can control the power ramp rate smoothly should be used for power raising and for compensation of reactivity depletion at higher heat rate level. Fig. 3 shows a typical example of power raise from cold condition to rated power level. The target MLHGR below which control rod withdrawal is allowed is set at 8 kw/ft. Above 8 kw/ft a core flow operation is predominantly used. It was noted that the Xenon concentration must be accurately predicted and be utilized in order to withdraw necessary amount of control rods below 8 kw/ft. The trajectory of power on the power-flow map has been computed and generated by the three dimensional Xenon transient code. (3) Compensation of reactivity depletion Control rod patterns have been periodically exchanged for compensation of reactivity change. For a relatively short duration core flow compensation is desirable from the viewpoint of capacity factor improvement. For this reason the applied control rod pattern must be determined in such a way that rated power is obtained at a core flow rate less than its rated one under the condition of the target MLHGR 15 ·kw/ft. For the assistance and convenience of operation, the CRT visual display system has been utilized for process computer output i I lustration.

4.~ Water Chemistry and Low Radiation Level Since feedwater for a boiling water reactor plant is supplied in a direct cycle, it has been used without being treated for corrosion inhibition by chemical injection. Corrosion products carried into the reactor vessel with feedwater are activated by neutrons, causing the plant dose rate to increase, and are deposited on fuel rods, decreasing their thermal margin. Because of these adverse effects, attempts have been made to minimize corrosion products. For the purpose of reducing feedwater corrosion products a pre-filter was added at the head of the demineralizer and the oxygen tnjection method was applied. This oxygen injection method takes advantage of the corrosion inhibition effect of oxygen. After a successful oxygen injection test oxygen was continuously injected into the feedwater. The results have been that corrosion products in the feedwater have been maintained as low as 1 ppb. Fig. 4 shows the water chemistry measurement. Fig. 5 shows radiation levels on the surface of equipments in primary containment vessel which were measured 4 days after reactor shutdown at the end of fuel cycle 1, 2 and 3. For example, at the end of fuel cycle 3, the levels of the surface of the recirculation pumps and the control rod drive system are about 30 mR/h and 10 mR/h respectively.

174

p " x ·k ; 3

s

~~~-ai-o

.g'.:,~5: . ~~~n: ;·v, ~!!!.

~Q)Q.l

•••

. ::0., ~-0 i::1° ::, Ill

3

n,t1>0>n 01=: ro"n~ 01::i11Z @03:::l (I) •.• CT, ~.::f_"'0»010(I) -

Ill

8.3.n,(l)Q)(I)

::,

!O

w

CT

(I)-,>

T (B) 112

(b) T CA) 112

T (B) 112

(c) T (A) 112

"' as shown in Fig. 5. An example of this phenomenon Is poor training followed by improvement In training procedures. The recovery phase curve may be time shifted to pass through the point p. to provide a measure of performance improvement with reference to initial performd1> However, this requires determination of steady-state achievement and the time

ance.

duration of the false plateau phenomenon is described by the sequential exponential models wherein the Initial phase is represented by p(t) = pil + Pn [1 - exp (-t/t1)J. t1

:5:

t'.5 T

(3)

P (t) = pi2 + Pn L 1 - exp H/t2)]. T :5 t :5 =,

(4)

and the recovery phase model Is

When the recovery curve Is advanced by plateau length, T, (5)

where subscripts 1 and 2 are assigned to parameters of Initial curve and recovery curve respectively. From Eqs. (3) and (5) and for t = O 1 p11 = Pii + Pn [1 - exp (-T/t2)].

(6)

The plateau length Is found to be

where 15t and p t2 are estimated values of the performance at t and t respectively. The 1 2 1 steady state value can be detected when the difference between two successive values of performance is small and constant.

TIIIIE, t Fig. I/.

472

Crossman Improvement Curve.

an the case of non-nuclear countries factors other than low level of skill have to be included in the model of training achievement. Neglecting the Grossman's trial and error segment of the curve, training achievement in a dynamic system of nuclear technology transfer may be measured in terms of performance, p{t) in success per number of trials expressed as p{t) =pi +pf (l-exp {-t/t)] -phexp {-at)[l3+siny{t)]

(8)

where t is the actual on-line training time, t is learning time constant, ph is a performance hindrance factor, and a, 13 and y (t)are disturbance coefficients which depend respectively on cultural compatibility between importers and exporters of nuclear technology, inadequacy of transfer media and instability of policy and economic planning. These parameters can be evaluated from socioeconomic analysis combined with analysis of training data for a given situation. The variation of the parameter y (t) with time can be neglected if the improvement rate of interest is evaluated for a short time;

for example 1 to 5 years. The

constant t is determined by the ability to absorb new information and -by the nature of skill required and pi depends on a priori experience in high quality industries. Figure 6 shows a typical dynamic training system with a step function input taking the performance hindrance factor ideally as zero. The performance output evolves with time into a plateau as shown in Fig. 7. False asymptotic values indicate poor training procedures.

✓-✓

t-------·'1----· ~

/./

rig. 5. Sequential Exponential Model.

473

,

p(t)

•l

.,,INITIAL LOCAL SKILLJ

....____...., ilLCARN ING LAG 1-------' KALMAN

PREDICTOR

FILTER OBSERVATION

NOISE Fig. 6. Dynamic Compensatory Training System.

p(t)

.,,,,,,--­ ,,,,,.,.,,,,,.

1/

/

. .---_/

/

/

FAt.SE PLATEAU

---TIME Fig. 7. Performance Output Characteristic.

q_

ECONOMIC MODEL

The rate of investment in training I (t) can be represented by l(t) = If+ (Ii+ If) exp[-t/E(t)l

474

(9)

where Ii and If are initial and final constant investments respectively and E (tl is a function of a,13,y and ?hand it represents the expected mean time of achieving a preset level of competency for a given mode of investment. The variability of E(t) takes into account the distribution of finance over time and hence it is evaluated from the probability density function of the budgetary situation over an extended period of time. For stable investment policies E (t) is constant and may be approximately taken as the learning time constant. Generally, E (t) may be defined as the mean time measured from the initial investment until the required number of operators acquire expertise and only marginal funds are allocated for retraining. According to Grossman's model, expertise is the ability to select exactly the right method after choosing the right method and the right source of signals.

In this

sense, if there are n investment alternatives for a given training program, then

n E (t) = :E (t./i) i=1 I

(10)

where ti is the mean time for using the ith investment policy. Figure 8 shows training investment patterns in a contributing country and in a recipient country. Trial and error procedures assume oscillatory modes. The recovery curve represents the effects of adequate screening of trainees and on-line training in contributing countries. Actually, the rate of investment on training a nuclear country drops faster than in the case of a non-nuclear country. This is due to the fact that training programs have been tested for a relatively long time In nuclear countries.

-•i

I

•

iFig. 8.

TIME

In vestment Rate in Training Programs Concerning Transfer of Nuclear Technology in Contrib_uting and Recipient Countries.

475

5.

CONCLUSIONS

The model developed here can be utilized in on-line computer modules to evaluate training progress in specific situations. The parameters invotved need to be periodically updated as more observations become possible to allow for better supervision, retraining or modifications of procedures. Guidelines can then be provided for expansion policies in the domain of transfer of nuclear technology. 6.

REFERENCES

1.

Husselnv, A .A., Z .A. Sabri and R.A. Danofsky, "Quantification of Operation

'

Experience and Incident Reports for Reliability Analysis and Improvement of Task Performance", Trans. Am. Nucl. Soc._ 23 1976, 484.

2.

Sabri, Z.A., A.A. Husseiny and R.A. Danofsky, "Estimating Operator Reliabifty from Actual Reactor Human Error Experience", Trans. Am·. Nucl. Soc. 23 1976, 31'1!.

3.

Sabri, Z.A., A.A. Husseiny and R.A. Danofsky, "An Operator Model for Reliabil~y and Availability Evaluation of Nuclear Power Plants", Iowa State University Report No. ISU-ERI-Ames-76177, February 1976.

4.

Bevis, F. W., "An Exploratory Study into Industrial Learning with Special Reference to Work Study Standards" M.Sc. Thesis, University of Wales Institute of Science and Technology, Cardiff, U.K. 1970.

5.

Towill, D.R. and F.W. Bevis, "Managerial Control Systems Based on Learning Curve Models", Inst. J. Prod. Res., 111973, 3.

6.

Towill, D.R., "A direct Method for the Determination of Learning c,,rvP. Parameters from Historical Data", Inst. J. Prod. Res., 11 1973, 97.

7.

Bevis, F.W., C. Finniear, and D.R. Towill, "Prediction of Operator Performance During Learning of Repetitive Tasks", Int. J. Prod. Res .• 81969, 293.

8.

Grossman, E.R.F.W., "ATheoryoftheAcquisitionofSpeed-Skill", Ergonomics, 12 1959, 153.

9.

Glower, J.H., "Selection of Trainees and Control of Their Progress", Int. J. Prod. Res., 5 1966, 43.

10. Towill, D.R., "Cusum Monitoring of Operator Performance During Learning", Work Study and Management Services, 16 Jan. 1972. 11 . Sriyananda, H. and D.R. Towi II, "Predication of Human Operator Performance", IEEE Trans. Reliability, R-22 1973, 148.

-----476

·