6

Sophie MOURLON - Notre première séanceplénière va traiter du contrôle de la sûreté deséquipements sous pression nucléaires. Nousallons tenter de dresser un panorama généraldes pratiques et des approches en matière desûreté et de gestion du vieillissement.

Nous avons voulu des panels variés et nousavons fait en sorte que des approchesdifférentes soient représentées dans chaquepanel. L'Autorité de sûreté nucléaire françaisesera représentée à la tribune par Alain Schmitt,directeur général adjoint en charge desréacteurs de puissance à l’ASN, par LaurentFoucher, adjoint à la sous-direction chargéedes réacteurs de puissance à l’ASN et qui estégalement ancien chargé d'affaires du BCCN,par moi-même, et par Dominique Arnaud, monadjointe.

This first technical introduction is about controland supervision of safety of nuclear pressureequipment. I will try to present the Frenchposition on this matter.

Ageing issues for nuclear pressure equipmentare numerous. Amongthem, and veryimportant, is thedegradation of manymechanical propertiesof materials. Forinstance, thermalageing and irradiationembrittlement areissues for pressureequipment becausethey degrade theirmechanical propertiesover time. We also

have the degradation of the equipment itself,for example through stress corrosion, fatigue,and other kinds of corrosion. Very importantand not to be forgotten: the loss of skills andknow-how and the obsolescence of materials.This loss of skills and know-how is importantbecause it affects engineering teams of utilitiesand also manufacturers and sub-contractors.This is all a challenge for the safe operatinglifetime of nuclear power plants. AsMr Borchardt said, it is very important to takethis into account now in order not to meet any

problems in the future with current powerplants or future power plants.

What is specific to France? In France, 58pressure water reactors are operated by oneutility. They are all similar in design. Theywere built by Framatome. The oldest one isFessenheim – it started operations in 1977 –and the most recent is Civaux, which startedoperations in 1996. We have quite animportant fleet of reactors with similar designand very close in age, because they all startedin a period of less than 20 years. This hasadvantages because we have a large fleet tohave feedback experience. That is a goodadvantage for ageing management but it alsohas drawbacks. In particular, any problem thatmight affect one reactor might, in fact, affect allreactors at just about the same time becausethey are all so close in age. This is taken intoaccount in the French regulatory approach toageing management.

It is important to say that, in France, theregulatory approach to ageing managementdoes not set any licensing lifetime. There is nolifetime introduced in the licensing process. Ofcourse, there are hypotheses on lifetime whichare taken into account in design studies, butthe operator is responsible for maintaining thesafety of the plant and the plant may operateas long as safety is ensured. I said theoperator is responsible for the safety of theplant. It is responsible for safe operation, ofcourse, but also surveillance in operation whichincludes in-service inspection, repairs andreplacements in time, and provision of safetydemonstrations. The Nuclear Safety Authoritymay require a comprehensive review of safetyat any time and may also stop a reactor ifsafety is challenged at any time.

In fact, this comprehensive safety review isdone every ten years. It is a periodic safetyreview every ten years during decennialoutage. The utility, during this safety review, isrequired first to check that safety requirementsare still met although the reactor is ageing anddegradations are appearing on the reactor.This includes improvements in safetydemonstration if necessary. Of course,improvements in the demonstration are not

SEANCES PLENIAIRES / PLENARY SESSIONSDébats / debates and discussions

Le contrôle de la sûreté des équipements sous-pression nucléaires /Control and supervision of safety of nuclear pressure equipment

Tribune ASN France / ASN France panel : Alain SCHMITT, Sophie MOURLON, Laurent FOUCHERSophie MOURLON – ASN France, Susanne SCHULZ – HSK Switzerland, Karen GOTT – SKI Sweden,Jose Maria FIGUERAS – CSN Spain, Katsuji MAEDA – NISA Japan, Frank MICHEL – GRS Germany,

7

improvements in safety. At this point, it is onlychecking that safety requirements are still metthroughout the lifetime of the plant. Thesecond point, and very important, is that theutility is expected to implement technicalimprovements on equipment and operation toupdate reactors to state-of-the-art technologyand design. This is improvement of safety.The utility is supposed to take into account, inparticular, the new technologies that aredesigned for new reactors. The programme ofthe periodic safety review is designed by theutility but has to be approved by the FrenchSafety Authority.

For nuclear pressure equipment, there is anextended number of regulatory texts. One ofthem is very important: it is the order ofNovember 10, 1999 for Main Primary Systemand Main Secondary System of PWRs. Thistext is oriented towards ageing management.All the requirements that are found in this textare, in fact, oriented towards ageingmanagement. It uses a ‘defence-in-depth’approach with requirements on design, onsurveillance operations, on maintenance andon feedback experience that we are going todetail now.

For design and fabrication, the designer andfuture utility are expected to study materialsand their fabrication with respect to ageingissues and ageing problems. Materials have tobe chosen and manufacturing processes haveto be chosen and studied with respect toageing management. They are supposed tobe qualified to known ageing degradations.The mechanical properties that have to betaken into account for design are expectedend-of-life mechanical properties. At thedesign stage, it is expected that the safety ofthe plant be demonstrated for the expectedlifetime of the plant. It is also required that thedesigner did what is necessary to preventfatigue. This means taking into account thepossibility of fatigue in design and also in futureoperation. Measures to prevent fatigue can betaken as well for the design of the equipmentas for the operation of the plant. It is alsoexpected, it is required, that the designer didwhat is required to favour in-service inspection.At the design stage, the reactor must bedesigned so that the utility may performthorough in-service inspection. No limit onin-service inspection should be introduced atthe design stage.

About design and fabrication, there are othertexts. For example, there is a 1974 order onconstruction and fabrication of power plants.There is now a new order in preparation on this

matter to take into account a recent EuropeanDirective on pressurised equipment. Thepresent reactors in operation in France werebuilt under the 1974 order. The French SafetyAuthority has also issued technical rules forconstruction that are to be applied for newreactors, in particular the Europeanpressurised water reactor (EPR) that iscurrently being designed in France and built inFinland. For France, these technical rules forconstruction that were issued officially by theFrench Safety Authority have to be taken intoaccount.

The utility, EDF, and the designer Framatomein association wrote the RCCM code to codifythe regulatory requirements. The FrenchSafety Authority has examined the RCCM codein its first version and is looking at the newerversions of the code. We do not approve it butby looking at it we check that what is written inthe code helps in meeting the regulatoryrequirements. When instruction is finished, weissue a letter to give a decision, to say that theFrench Safety Authority is okay with the RCCMcode and this should be the reference of thedesign and construction of nuclear pressureequipment.

In the defence-in-depth approach there aremany important features for operation; inparticular, a good surveillance of the plantshould be done by the utility. It has to monitorthe relevant parameters – pressure,temperature, chemistry – and to do a transientbook-keeping to check that operatingconditions are consistent with designhypotheses, and also to perform in-serviceinspection to detect degradations and flaws. InFrance, we require the utility to do its in-serviceinspection in order to detect the flaws beforethey challenge integrity and before they lead toa leak. This is to check that materials behaveas anticipated. They are designed withhypotheses on operation that have to bechecked during operation. The designconclusions on the behaviour of the materialsduring operation are to be checked throughoutoperation of the plant.

8

The order of 1999 requires that, for eachdegradation mode, there be an ageingsurveillance programme that is designed by theutility and that has to be approved by theSafety Authority. The degradation modes, forinstance, are irradiation embrittlement for thereactor pressure vessel, and thermal ageing.Any degradation mode that is linked to ageingand that is identified is supposed to beaddressed through a specific surveillanceprogramme.

To design in-service inspection programmes,the utility is expected to analyse expecteddegradations and to adapt NDE techniques toflaws. We require performance demonstrationsystems, called qualification, that we willpresent tomorrow during another plenarysession on ISI. The NDEs have to be adaptedto the flaws that are expected – known orexpected. Also, the definition of the frequencyof the examinations has to be defineddepending on the expected degradation and onthe growth rate of the expected flaws. But notonly that – we consider that it is not enoughand that sample checks should be performedon top of those examinations that are done tosearch for expected degradation and expectedflaws. As I said in my introductory address:nature and physics are always moreimaginative than men and engineers. To tryand meet nature and physics, we require thesesample checks to find degradations that werenot expected in case they happen. Also, ofcourse, we require the utility to take feedbackexperience into account for its in-serviceinspection programmes. In-service inspectionprogrammes are expected to be revised atleast every ten years to take into accountfeedback experience – national andinternational.

As I said, nature is more imaginative thanengineers so we require sample tests and alsohydraulic tests that are performed every tenyears on pressurised equipment. Thishydraulic test is a global test that allows us tofind – if it should happen – importantdegradations that were not expected. It proveduseful in 1991 when, during a hydraulic test atBugey 3, the reactive pressure vessel headshowed a small leak. This started a whole setof issues about reactor pressure vessel headsthat we will talk about in one of the workshopsthis afternoon.

For us, for the French approach, it is veryimportant for sample tests and hydraulic teststo be performed. Although we are trying,through feedback experience, throughanalysis, through studies, through research, toknow what is going to happen, we do not know

everything. At some point, we must haveanother type of test, of global test, just to checkthat nothing else that we did not expect isactually happening in the plants. This showsalso the importance not only of nationalexperience but also of international experiencebecause the age of the reactors from onecountry to another may be different. Theoperation of the reactor may be different. Alsoit allows a bigger statistical set if we look atinternational experience to find whatdegradations are appearing here and there.

I must also insist on the influence of operationprocedures: the same reactor, operateddifferently, may develop different degradations.In France, 80% of the electricity is producedwith nuclear power plants. This means thatsome of the power plants follow the electricitynetwork so they are not operated on a basicoperation but there are fluctuations to followthe electricity network. This means thatoperation in some of the plants is a bit differentand may create new degradations or differentdegradations or accelerate the process ofdegradation.

Maintenance: a defence-in-depth approach isalso very important. The utility must makesure that repair techniques and repairequipment are available. This means thattechnical skills have to be created andmaintained, that they have to check that thecontractors will be available with the right skillsat the right time, and the right number ofcontractors – this is very important in Francewith 58 very similar reactors – and that theindustrial capacity is still there for repairequipment. The utility is expected – it is in theorder of 1999 – to repair cracks as soon asthey are detected. Also, we often require thatutilities perform research on replacedelements: when equipment is taken away fromthe reactor and replaced by other equipment, itis a good opportunity to perform research onthe equipment that was taken away.

With respect to ageing management, theposition of the ASN is as follows. The ASNconsiders that operation of the French reactorsfor 30 years is possible with adequatesurveillance in operation and consideringrelevant safety cases, of course. With what weknow, what we have looked at, and what theutility is doing, we consider that operation for30 years is possible. Beyond, a thoroughanalysis is required. The condition of the plantwith respect to ageing phenomena has to beaddressed and the demonstration has to bemade that operation may proceed safely for tenmore years. The utility is submitting operation

9

continuation aptitude cases – in French, DAPE– and a very comprehensive reviewprogramme will have to be performed at thethird decennial outage when the reactor hasreached 30 years. Only after that will we knowif operation may go on beyond 30 years.

In conclusion, for the French Safety Authority,ageing management requires good design,thorough surveillance and in-service inspection– because design and fabrication are veryimportant but not every degradation can beprevented at the design stage and we have tocheck that everything is happening the way wethought it would happen – and also thecapacity to repair and replace in time.Managing ageing is one thing; more interestingis to anticipate. Anticipation requiresexperience feedback, requires sample checksand hydraulic tests to find the degradationsthat were not expected, and, of course,research that can find new degradations thatwill happen or may happen on the reactors,and international experience. Thank you.

Ulla EHRNSTEN, VTT Finland - I have twoquestions. I do not know whether they shouldbe put to you or somebody else in thisaudience. My first one is that you said that notall French plants operate on steady-stateoperation with full power all the time. Is thatseen in the degradations, in that the more youhave fluctuations in capacity, the more youhave degradation?

Sophie MOURLON - This is under study. Ofcourse, we think right away about fatigue butfatigue degradations have shown that they arenot linked only to this kind of operation. One ofthe main examples of fatigue in France is theCivaux event, when a crack appeared at thevery beginning of operation at Civaux. Thiswas not linked to non-steady operation of theplant. The question is under study. Maybe theutility will tell us a bit more about that tomorrow.

Ulla EHRNSTEN - My second questionconcerns the internals for the EPRs. Thedesign basis is 60 years. You said that for allthe ageing modes that you might have, youneed a surveillance. How are the internals ofthe pressure vessel going to be surveyed forthe EPR?

Sophie MOURLON - I am sorry, I cannotanswer that. BCCN only deals withpressurised equipment : the internals are not inour scope.

Alain SCHMITT, ASN France - Maybe justone complementary remark about the influence

of load variations. In fact, up until now wehave not seen any effect on ageing of loadvariations compared to baseload production.The licensee has decided to study this issue inmore depth and to concentrate on baseloadproduction on some reactors and loadfollowing on the others. Maybe, in some time,we will have new things to say about this topicbut, for the time being, we have not seen anyinfluence on the ageing of pressurisedequipment.

Dominique ARNAUD, ASN France –Susanne Schulz, physicienne, inspecteur del’Autorité de sûreté nucléaire suisse HSK, vanous présenter le programme de surveillancedu vieillissement ainsi que les documents émisdans ce cadre en Suisse.

Susanne SCHULZ, HSK Suisse - Thank youvery much for giving me the opportunity tointroduce the Swiss Federal Nuclear SafetyInspectorate’s guidelines for ageingsurveillance formechanicaland electricalequipment incivil structuresin nuclearinstallations. InSwitzerland,we have fournuclear power plants with five reactors thatcover about 40% of the electric energyproduction in Switzerland. We have three quiteold reactors : an old Westinghousepressurised water plant with two reactors atBeznau and an old General Electric boilingwater reactor at Mühleberg. Somewhat neweris a Siemens pressurised water reactor atGösgen and then another General Electricpressurised water reactor at Leibstadt. Thethree oldest blocks have accumulated alreadyover 250 000 operating hours on the net. If welook at the world statistics of nuclear powerplants we see that our reactors are all in thesecond half.

The history of the Swiss Ageing SurveillanceProgramme dates back to the early 90s whenalready some damage had been found. In1991, a letter was sent to the Swiss nuclearpower plants with the requirement to establishageing surveillance programmes. In responseto that, a working group of the Swiss nuclearpower plant operator, GSKL, has developed abasic programme for that task, that wasacknowledged by HSK. Since about ten years,the elaboration of plant-specific AgeingSurveillance Programme procedures anddocumentation is underway. Some of the

10

documentation has already been revisedseveral times. At the end of last year, HSKissued guidelines on ageing surveillance ofmechanical and electrical equipment in civilstructures in nuclear installations. It has 51guidelines. It is only in German because all ofour nuclear power plants are in theGerman-speaking part of Switzerland. Atnearly the same time, the Federal Law onNuclear Energy in Switzerland was renewed.The new Federal Ordinance on Nuclear Energynow has a separate article that requires ageingsurveillance, Article 35. You can see it on theinternet in German, French and Italian.

Some recent ageing issues in Swiss nuclearpower plants are, for instance, the core shroudcracking in Mühleberg. This was found in 1990and Mühleberg decided not to replace the coreshroud but they installed reinforcement. Inrecent years, there were efforts to slow downthe crack growth and crack initiation bymodification of the primary water chemistry ofthis boiling water reactor but it has not beensuccessful so far. This will be an issue in thefuture. The stress corrosion susceptibility ofthis Inconel 600 penetration that we alreadymentioned is also an issue, although luckily nocracks have been found in Swiss nuclearpower plants up to today. Beznau nuclearpower plant, with the Westinghouse reactors,recalculated the reactor pressure vessel headtemperature last year and found it was higherthan they had thought before. Thisreassessment led to an enhanced ISIprogramme in the last few years. I think thenext test will be in fourteen days or so.

Another issue is steel containment corrosionthat resulted from temporary leakages. It hasbeen found in recent years and occurs mostlyon inaccessible areas of the containment and alocal loss of wall thickness of more than 10%has already been detected. We think it is not avery severe condition at the moment but, in thefuture, there shall be additional inspections andexaminations in order to stop these leakagesthat cause that corrosion. A single ageingissue was the finding of cracks in a safe end ofa reactor pressure vessel nozzle in Mühleberg,that was caused by thermal stratification. Itwas interesting because this nozzle was in theageing programme and the mechanism ofthermal stratification was mis-judged becausethe temperature distribution was calculatedwith the design flow which was not true anymore for operation. There is a lesson to belearned from this.

Our guidelines give a definition of ageing, thatis, cumulative time-dependent change in

physical or chemical properties. Theguidelines deal only with material ageing. Wedefine ageing surveillance as all measures oftimely recognition, evaluation and mitigation ofthe condition of ageing. The ageingsurveillance programme is the systematicprocedure to do so and close gaps if theanalysis shows gaps in ageing surveillance.

The whole service life of our nuclearequipment is covered by several guidelines.The ageing surveillance guidelines cover thewhole service life from design to removal of thecomponent. The ageing surveillance shall notonly take into account normal operation andageing by normal operation but also singledamaging events and flaws from fabricationthat are left in place and may influence theprocess of ageing.

The basic requirements of ageing surveillanceare the identification of the ageing mechanismswith the help of catalogues of ageingmechanisms; the component-specificidentification of possible ageing mechanismsand the documentation of these assessments;the inventory of existing methods of ageingsurveillance; evaluation of inspection methodsand techniques and, if necessary, lists ofsupplementary actions; and, of course, aninterface regulation between the differenttechnical departments so that no orphancomponents between the technicaldepartments are left over in ageingsurveillance.

There are several requirements for thesystematic procedure that is to consider allknown and possible ageing mechanisms, tocheck the qualification and application ofageing surveillance methods, to identify andtreat possible deficiencies and open questions,to evaluate trends from maintenance andoperating experience, and to evaluate theknowledge from research and technical andindustrial experience. Then we have the tasksto determine the values of risk relevance forcomponents – I will come to this later – with thehelp of probabilistic safety assessment studies.Last but not least, to document the results andproofs of ageing surveillance.

The component-specific evaluation of ageinghas to take into account, on one hand, thegeneric ageing information and the specificlocal data of the component or system, such asmaterial information, water chemistry,environment transients and so on. Theassessment shall end with a list of positionswhere possible ageing mechanisms have beenidentified, the main method that is

11

implemented as an ageing surveillancemethod, and references to establishedprogrammes and, if a gap is found, then areference to the action plan that is necessaryto close that gap.

In our guidelines we have specific instructionsfor civil structures but because of the lack oftime I will skip them. Then for electricalequipment and mechanical equipment let memention that we look mostly at the classifiedequipment; of course, all the pressure retainingboundary of the primary coolant system and soon. There may be important components forthe safety of the nuclear power plant that arenot classified; we try to cover thesecomponents by the evaluation of risk valuesfrom probabilistic safety assessment studiesbecause these studies are done independentlyof the safety classes.

I think we have already heard lots of these withthe previous speaker so I come to HSK’ssupervision of ageing surveillance, thatconsists in: assessment of the ageingsurveillance programme documents,catalogues and technical reports; the review ofregular and event reporting of the plantoperators’ inspections; plant walk-downs;technical and regulatory meetings to discussageing issues; if necessary, requirements forfurther assessments; and, like in mostcountries, the assessment of ageingsurveillance activities as part of the periodicsafety review every ten years.

Let me come to the conclusion. With ourageing surveillance programmes, we now havea systematic procedures established todetermine the current state of ageing of ourcomponents and support the planning andmaintenance in this respect. Although ourageing surveillance aims at material ageing, ithas useful side effects because the historicalinformation is made available from the archivesduring the assessment so it counteractsdocument ageing. Young people learn aboutthe history and operating experience of the oldcomponents and systems so it counteractspersonnel ageing. The recent periodic safetyreview reflects successful experience with ourageing surveillance programmes. There havebeen periodic safety reviews for Gösgen, forMühleberg, and for both blocks of Beznau inthe last few years. The next one will be forLeibstadt. The slogan is ‘Ageing undercontrol?’, but there is a question mark becauseit is a permanent struggle. Thank you.

Dominique ARNAUD - Spécialiste de laphysique des métaux à l'Autorité de sûreté

suédoise SKI, Madame Gott a travaillé sur lachimie des réacteurs et possède une grandeconnaissance des différents aspects ducontrôle. Elle va nous présenter la positionsuédoise vis-à-vis des problèmes devieillissement, ainsi que la base de donnéesSTRYK qui a été élaborée pour son suivi.

Karen GOTT, SKI Sweden - Good morning,ladies and gentlemen. I am going to give ashort background, and to explain a little bitabout the regulatory situation in Sweden, whichhas a political aspect to it which you may ormay not know about. After that I will talk aboutwhat I think is a useful tool in the following ofmaterials degradation and, thus, materialsageing problems. At the end, I am going to giveyou some concluding remarks.

Bosenbeck One wasshut down forpolitical reasons. Itresulted from achange in the law,passed after the1978 referendum, tophase out nuclearpower by 2010. Asecond consequenceof this law was thatBosenbeck II wasshut down on 31 Maythis year. The currentsituation following the realignment of thepolitical situation is that nuclear power maynow be used and operated as long as it istechnically viable and it is considered safe.

SKI is a small organisation. We regulate, reactto safety, non-proliferation, and nuclear waste.We do not have any responsibilities except ona collegiate basis with our sister authority onradiation protection. The full and undividedresponsibility for safety lies with the licensee.SKI, being small and using the traditionalapproach of issuing prescriptions andregulations – not ‘cookery book’ regulations –has adopted a regulatory strategy in which weensure that the licensee has the organisation,both with respect to quality assurance,documentation and systems and also sufficientcompetent personnel, to handle theirresponsibilities, as defined both by the Act ofNuclear Activities and SKI’s regulations.Following the changes in the law, the Swedishutilities are currently planning power upgradesof up to 130% and also an operational life of upto 60 years.

Following the recent changes in the law, theprescriptive regulations that we originally

12

issued, known as SKIFS 1998 : 1, have beenreissued as SKIFS 2004 : 1. They becameeffective on 1 January this year. Theseregulations are basically editorial changes andthe major difference is that we have introduceda stipulation that the utilities must provide SKIwith an ageing management programme bythe end of this year for us to review. We havealso issued complementary regulations –2004:2 – which stipulate that utilities mustmodernise their plants so that they comply withcurrent safety and design standards. Theseregulations were issued after extensivediscussions with utilities and have theiracceptance. They are, in fact, seeing this as areason for future investment and haveextensive investment plans over the next ten tofifteen years.

SKI’s regulations may not have requiredageing management programmes as such, butfor many years they have requiredmaintenance programmes which should bebased on the results of plant-specificprobabilistic safety analyses and risk-orientedinspection programmes.

The oldest SKIFS is, in fact, that concerningmechanical components. SKI was firstformally allowed to issue regulations in 1992 –one of my first jobs at SKI was to helpformulate these regulations. We modernisedthem in 2000, to comply with the 1998regulations. The mechanical componentregulations are more specific than the overallregulations. They require inspection andtesting and other programmes to ensurestructural integrity. We have, since 1994,required that only qualified inspection andrepair procedures are permitted. Alldegradation must be reported and alsoinvestigated so that we have an assured rootcause report. If you find any degradation thenyou have to expand the inspection sample tocover 100% of similar components – similareither because they have the same material orbecause they have the same operationalsituation.

We have been using this risk-based inspectionrule for a long time and it is based on aconsequence index and a damage index. Thiscombination will give you :- 100% in the area that is designated “A” ,- a sampling in the areas designated “B”,- and in “C” you must have your own

non-reportable inspection programmes.

The damage index is assigned on acomponent-by-component, weld-by-weld basisand depends on anticipated degradationmechanisms. The utilities now have guidelineson, for example, carbon count content,temperature for thermal fatigue and such like,to designate the damage index. We also havea consequence index. This is both a moreglobal approach but it is also associated withthe PSA results. The larger the consequence,the higher the risk, the faster you go up into100% sampling for your inspection.

In the mid-90s we realised that we hadcollected information from the beginning of thenuclear power operation in Sweden from the70s and we started to try to get this organisedinto a database. We have not limited it topiping; we have extended it to include allmechanical components that are regulated bySKIFS. To date, there are about 1 900 entries,which are set for cracks, and these cover1 300 different events. This is what theinterface looks like. I would like to point outthat when you have a database, you have to becareful how you analyse it – what you actuallyput in and what you can actually get out. Youcan only put in what people will give you asinformation. All countries suffer from the factthat reporting level requirements have changedover the years, so that older information is notnecessarily exactly equivalent to currentinformation. This means that you may noteven have information about some eventsbecause people were not required to report it inthe early days.

The time entered for an event is always thetime at which you discover the crack – it is notthe time when the crack appeared. If you havea ten-year cycle, you could have a ten-year-oldcrack or you could have a one-year-old crackor you could have a two-day-old crack: you donot know. Another problem is that systemnumbering varies between plants. This maysound like a minor detail but it is, in fact, aproblem. I have cases where not even theutility knows which system they are actuallyreporting on, so some reports have one systemnumber and exactly the same event on anotherreport will have another system number. Youhave to make a decision as to which system

13

you are going to report this event under. Thiscan also give you discrepancies when you aredoing the analysis for inspection procedures forthe inspection programme. You can see thatthere are some years in which there were alarge number of events. 1986 was a year whenRinghals I and Oskarshamm I found a lot ofstress corrosion cracking. This was followingthe events in the United-States. We had oneor two leaks and they decided to just replace alot of piping and went into the lab and looked atthis piping and these events are recorded as1986 stress corrosion cracking events. In theearly 90s – 1993 and 1994 – Oskarshamm Idid an inspection and found a major problemwith its feedwater system. That is, in fact,unique in that the feedwater system goes in atthe bottom of the reactor vessel and is, in fact,an internal pipe in the final stage of thatsystem. That resulted in a major renovation ofthe plant. They worked for about 18 months,finding more and more cracks and problems asthey proceeded and had permission to restarton the condition that they replaced their coreshroud and the header of the internals – a lotof internals have been replaced there. Thisstarted a trend in Sweden and several of thepower plants have replaced their core shroudsto avoid cracking problems in the future. ThePWR core shrouds in Sweden are bolted inplace rather than welded so it is a slightlyeasier operation.

Around 2000 we have another example of ageneric problem. Inconel X750, with the wrongheat treatment is a well-known problem forstress corrosion cracking. The improvedinspection procedures that were implementedin 1999-2000 found a large number of cracksin some internals. You also have a problemthat you may not be finding things because youhave the wrong inspection procedure: eventhough you think it was qualified it may nothave had sufficient resolution to find thecracking. This is just the same information in adifferent scale.

Since we have had this inspection programmein place, detection methods have found almost90% of cracks. The major degradationmechanisms do not differ in Sweden fromanywhere else in the world. You can use adatabase to see if your root causes are thesame as in other countries. Again, this isalways a subjective analysis, it is a personalanalysis – this is my personal analysis basedon reading as many reports as have beenmade available to me. You will notice that it isnot, in fact, weld sensitisation that is the majorproblem in Sweden but it is cold work. Coldwork due to manufacturing practices but not

least cold work due to grinding and scratcheson the internal surfaces of the components.This continues to be a problem and I think it isa problem that is not recognised sufficientlyaround the world. Cold work and sensitisationare an ongoing problem on our plants, as InterGranular Stress Corrosion Cracking is afunction of operational time. It is a continuingproblem despite the fact that several of ourreactors have been running on hydrogen waterchemistry. I do not think that hydrogen waterchemistry stops propagation and, thus, whenyou increase the inspection resolution, you aregoing to find cracks that have been there forsome time. There are also new cases ofstress corrosion cracking or fatigue. I think youhave to be careful when you are looking atageing management programmes : they areplant-specific. You cannot say, “This is ageneric type of plant, therefore we will not haveproblems. Look at Oskarshamm III – noproblems therefore there will be no problems inForsmark III.” – it is not true. You have to dothis on a plant-specific basis.

In-service inspection programmes areeffective. Many people call these ‘ageingmanagement’ programmes but we think thatan ageing management programme is moreextensive: it includes maintenanceprogrammes; it includes the ageing of codesand standards; it includes the ageing ofpersonnel, as has been said earlier. However,they are an important basis for ageingmanagement. I think a database can be usefulto help analyse both inspection programmesand ageing management programmes. Itgives early warning of trends and it can alsohelp assess if the individual programmes areappropriate. I think it is necessary to correlatedegradation mechanisms and inspectionprogrammes to include all systems and allcomponents and not limit it to piping. I wouldrecommend, to those that do not already knowabout it, the international cooperation on theOECD/NEA database, which covers piping.Maybe we should, in the future, extend that tocover more components. Thank you.

Dominique ARNAUD – Monsieur Figueras,spécialiste de la sûreté nucléaire et de lamécanique appliquée au domaine del’industrie, appartient à l’Autorité de sûreténucléaire espagnole CSN. Il va présenter ladémarche de renouvellement des autorisationsd’exploitation des réacteurs en Espagne. Cettedémarche sera illustrée par le cas du réacteurSanta Marìa de Garoña, en regard desproblèmes de vieillissement.

14

José Maria FIGUERAS, CSN Spain - Bonjour,Mesdames et Messieurs. Good morning, ladiesand gentlemen. I was asked to prepare apresentation in cooperation with some peoplefrom one of the Spanish plants: Santa Maríade Garoña nuclear power plant in northernSpain. The presentation mainly refers to the

specific case of how thisplant is conducting theirown analysis and studieson long-term operation.First, I will makereference to the Spanishregulatory framework forlong-term operationbeyond the 40 years’design life. Then I willgive a brief description ofwhat an ageing

management evaluation is, without describingit because I think all of you know the problemsand because it will save some time for thediscussion later on. I will go directly to the caseof Santa María de Garoña with the analysisspecifically of the reactor pressure vessel andsome examples of ageing managementprogrammes and time-limit analysis.

To briefly describe the framework of ageingmanagement in Spain, we can say it is dividedin four major phases or parts. The first one isthe management and evaluation of ageing withthe classical phases of scoping, screening,definition of the ageing management reviewand time-limit ageing analysis. The second partrelates to radiological impact. The third one isthe analysis of new regulations that could be inplace beyond the 40-year period – not only theactual duration but also new regulations in thefuture, that needs some analysis. Finally, asprevious speakers said, we also, in Spain,follow the European scheme of ten-yearperiodic safety reviews. All this informationmust be submitted to the CSN under theperiodic safety review package.Santa María de Garoña power plant started in1971 and is almost 33 years old today. It is aGeneral Electric Boiling Water Reactor, modelBWR 1, and containment type mark 3. It has aroughly 500 megawatt electrical output. It issimilar to some other plants in Europe and theUS, such as Mühleberg, Monticello, Dresdenand some others.

At that plant, the owner has prepared a projectcalled 2019 in order to prepare all the tasksneeded for a long-term operation and toprepare the documentation in order to submit itto the CSN, to the regulator. From 2002 to2005, they have prepared all those phases.The actual project status is that they are

preparing the other part: the periodical safetyreview documentation and integrating all theageing analysis in that periodic safety reviewapplication. They intend to submit to CSN thisinformation by next year, mid next year, and wehope that in two years or two and a half years,we will have the review performed in order togrant a renewal of the licence for ten additionalyears in 2009.

What does the ageing management reviewlook like? The components of the plant aredivided into sub-components like, in the reactorvessel, the reactor vessel bottom head. Then,are identified the intended function, thematerial, the environmental conditions to whichit is submitted, the degradation which isexpected, the ageing effects which should behoped for by management and the ageingmanagement programmes which are in placein the plant. Finally, – because we followalmost fully American regulation 10CFR54 forlicence renewal rule – we say in which chapterand which table of the GALL Report – the new1800-1 report – should be found information foranalysing this type of evaluation.

Examples of ageing management programmesare typically Section XI of ASME, for: in-serviceinspection; water chemistry programmes; thereactor head closure stud; for boiling waterreactors especially, the feed water nozzle andthe vessel internals; thermal ageingembrittlement; and so on. For time-limitedageing assessment analyses, the resolution isthe classic one, by extension of the actualanalysis to add an additional period of at leastten years and maybe 20 years. This means toreach the 60-year period. In some cases,generic information can be found in standardtechnical literature but in others, it should beanalysed specifically for the plant because it isa specific item.

Concerning the final values that can beobtained for the RTndt, for up to 60 years ofeffective full power years of operation, theincrease in reference temperature is less thanthe 200-degree limit. For the impulsive energyalso, the reduction is well over the limit. Thatmeans that embrittlement of the vessel wall bythe neutrons will not challenge the vessel in theSanta María de Garoña plant up to 60 years ofage.

Let me give some conclusions. The first is thatthe licensing requirement for long-termoperations in Spain has started. Santa Maríade Garoña is the first plant to apply for that andis now preparing the documentation. We willapply next year. The nuclear regulatory body

15

has prepared a document entitledRequirements for Long-Term Operation ofNuclear Power Plants in Spain, that containsthose aspects that I have reflected in the firstpart of the presentation. That means thatthose requirements put emphasis on ageingissues to ensure that key plant components willperform the intended function during theextended operation period in such a mannerthat licensing bases are maintained. The thirdis that, with this first case of the Garoñaapplication, it has demonstrated that there is arobust methodology available to the relevantaging effects of those key plant componentsand equipment. Finally, the preliminary resultsfor the Santa María de Garoña plant show that,in principle, there are no technical obstacles tothe extension to 60 years. Every 10 years, thelicence we will grant to Garoña will be to 2019,then they can apply again if they wish for afurther 10 years. Thank you very much.

Sophie MOURLON - Je souhaiterais d’abordremercier les trois intervenants pour cesprésentations très intéressantes. Ma premièrequestion va à Monsieur Figueras.In relation to Santa María de Garoña. I wouldlike to know if there are maintenanceoperations or replacements that will have to bemade for you to issue the licence renewals.

José Maria FIGUERAS - Yes, I think so. Ithink there will be changes on maintenance.Personally, they are moving to a more centredmaintenance than the classical prescriptivemaintenance. If you want to know some moredetails I prefer that the owner answers thequestion.

José TORRALBO, NUCLENOR, Santa Maríade Garoña - Last year we changed ourmaintenance programme in accordance withASME, for a more reliability centredmaintenance. Our maintenance programmehas been changed recently, in the last six toeight years, according to this new approach,LCM and maintenance rule.

We are again discovering new changes for thepassive components. We have a list ofimprovements to incorporate in our programmeto adapt our actual programme to these newissues that we are discovering. We have thisframework from now to 2009 to incorporate inour programme, and we have decided toincorporate them now and not to wait for a newpermit.

Laurent FOUCHER, ASN France - Maybe Icould ask a complementary question on thispoint. Since the replacement is an element in

the safety demonstration, sometimes youcannot replace parts and you have to justify byresearch and development methods. Do youhave such programmes to complete thereplacement or to complement the safetydemonstration?

José Maria FIGUERAS - Yes, there are. HereI will refer only to the Garoña scheme, but thisscheme is also relevant to all the plants inSpain. With the exception of the Solita Plantthat is going to be closed next year, there is astandard programme for the replacement ofcomponents. The most renowned is, ofcourse, the steam generator replacement, butalso the vessel head. The big headache thatthe vessel head has created all over the world!But there are also other systematicreplacements, for instance, turbines lowpressure and high pressure bodies in theturbine. Also the balance of the plant is moreor less systematically changed. And specific orparticular problematic parts, for instance, in theGaroña plant they have replaced the clean-upsystem that is exactly the same system as thepressurised water reactor and the chemicalsystem. All the stainless steel parts have beenreplaced.

Laurent FOUCHER - My question is: at timesyou are at the limits of the classical justificationmethods and you have to improve thejustification methods by research anddevelopment actions. For instance, you cannotreplace the parts or you want to improve yourknowledge of the margins which are effectivelyavailable. To complete the safetydemonstration files, this is a point which I didnot find in the presentation. All theprogrammes are very structured. We mightthink that the problem is under control but animportant part of the demonstration is thatsometimes you are at the limit of the classicalmethod, so what do you do?

Karen GOTT - In Sweden we have arequirement that any replacement must beeither identical, or a proven technology ortested for that specific application. For

16

example, in 2003, a thermal sleeve wasreplaced to a T-junction. According to thesupplier, it was proven technology, and it hadbeen used in other industries. It very quicklysuffered from thermal fatigue and wearproblems and caused problems with the flowinto the core. They had to shut down. It is veryimportant that you use proven technology forthe nuclear industry or that you test thecomponents which you are replacing.

Sophie MOURLON - You said yourself thatyou think that problems are client specific.

Karen GOTT - Yes.

Sophie MOURLON - So, of course, theproblem with qualification of design andmanufacturing, having proven technologies is agood thing, but I think that probably it is notenough. So what is good for one plant, mayjust be very bad in another plant.

Karen GOTT - You can never guarantee, butyou can always monitor if there areuncertainties.

Sophie MOURLON - I also have a questionabout the database. You told us how carefulone has to be with such a database because ofthe data that is in it. The French approach isthat the utility should have the database, notthe safety authority. So we expect the utility tohave this database and to give us the analysisof what is in the database. What do you thinkof that and what approach is taken in othercountries?

Karen GOTT - We certainly would not rely onlyon a utility database. Over the years, we havehad four different utilities. That is oneadvantage you have, you have one utility, soyou have one French database. To get anational database in Sweden we decided thatwe needed to actually do the work. The utilitieshave different reasons for having databases.In Sweden they have them in a proactivemanner so they know that if they find cracks inone weld, they know all the other welds that

have similar material. So they can expandtheir sample to cover all those welds. Theyalso know whether the supplier has suppliedthe same material to another plant, so it maybe that the sample has to be expanded toother units. I think that you need a nationaldatabase rather than a utility database.

Sophie MOURLON - That is true. What aboutSwitzerland?

Susanne SCHULZ - In Switzerland we havedifferent companies with nuclear power plants,so they are in competition with each other.They can take advantage of the database ofWestinghouse or the GE nuclear power plantsfeedback. And of course, can get someinformation through the basis of the Siemensworld of pressurised water reactors. Theworking group of the Swiss Nuclear PowerPlant Operators, the GSKL is not anorganisation; it is really just a working groupwhere specialists come together from theirrespective areas and do some work together.There is no institute or anything like that behindit. So sometimes it is not easy to get powerplants to collaborate. Each one wants to makeits own way in the cheapest manner possible,which can be problematic. We try to encouragethem to collaborate because we are sure thatall parties will profit. We have no database,the nuclear power plant operators do not havea common database. We have somecollections but I would not call it a database.

Katsuji MAEDA, NISA Japan - I think thatdatabases should be shared betweenlicensees and regulators. The regulatory sideshould confirm the adequacy of the ageingmanagement programme. They should usethe same database for the same calibrations.Not only should the database be shared byutilities and licensees of domestic regions, butalso I think ageing management databasesshould be established internationally. Manypeople are gathered to discuss how toestablish or how to make integrity of ageingmanagement programme.

Sophie MOURLON - Maybe for internationalcooperation databases, we will have somemore information at the beginning of theafternoon, with the role of internationalorganisations.

Matthieu SCHULER, ASN France - I am theDeputy Head of École des Mines engineeringschool in Nantes. My question is again on thedatabase. It is always interesting to look atnumbers and figures. I must admit that I wasastonished by one number that came up.

17

When you add the percentage of degradation,discovered by very simple methods, includingvisual control, penetrate testing and workdown, you have nearly 33% of the detection.This means that in the surveillanceprogramme, we have to pay attention to thisvery simple method. As Sophie Mourlonoutlined in her introduction, when we aretalking about risk inform programmes, theapplication of sample checking of very simplemethods is, to me, very useful to detectdegradations which were not expected before.In this database, I did not see anything thatwould be safety relevance of the degradationwhich has been seen. Do you have ananalysis of what could have been the safetyrelevance of the degradation if it had not beendetected?

Karen GOTT - Visual inspection also includesthe use of TV and video. For example,internals are almost exclusively inspectedusing visual techniques, so they are notnecessarily the simple walk down type of,‘seeing it with your eyes’. Visual techniquesused for inspection of internals have to becalibrated, as well as ultrasonic techniquesbecause of the lighting problems, the oxidesand the different colorations. So visualtechniques can be at least as complex asultrasonic techniques. I do not think you canjust call them ‘simple’ techniques. The safetyrelevance of the degradation, yes, we doregular analyses and study trends. I am due todo another complete analysis of the databaseover the next 18 months or so. To date thesehave been produced in Swedish, but I feel I amunder international pressure to write it inEnglish next time.

Claude FAIDY, EDF France - Concerningdatabases in our country, we have a first

exchange with thesafety authoritybecause we havesome databases,we probably can notanswer all thequestions they have,but we are ready todiscuss that withthem. I have a moregeneral question tothe three authors.

You mentioned risk-informed use at differentlevels. This does not apply to France and wewould be interested in knowing to what level doyou apply it? And what are your requests froma safety aspect for local crack or fracture orlow consequences situation. It is the most

sensitive aspects which are connected to theunknown aspects.

Karen GOTT - We have applied this qualitativerisk-based approach for more than 20 years.You do not know what you do not know and

you do notknow whereit is going tohappen. Wehave asamplingsystemwherebyyou have to

choose components, materials, combinations,environments, so that you do get a goodsample of the plant. And this samplepopulation should be studied in a 10-yearcycle. We found PWSCC in safe ends beforeit was through a wall, using this methodology.It may not be the answer to everything, but itseems to have worked fairly well to date.

Susanne SCHULZ - For Switzerland we alsohave a simple risk-informed procedure in ourISI regulation for safety Class 2 components.It is only a qualitative method. We must look atthe risk relevance of components from PSAstudies that cover a whole power plant to seewhere are the risk-relevant components,independent of the classification system. Thisway we can get inside the components, whichmight be important for the safety of the powerplant but are not included in the considerationsof ageing up until now. It is a means tocomplete the whole picture of safety.

Laurent FOUCHER - Did you want tocomment?

José Maria FIGUERAS - Also in Spain we areperforming that kind of risk inspection activity,mainly for Class 1 piping and the primaryloops. For instance, you know that on theASME code XI, you do not need to make aninspection for each piece of piping. When youperform this risk-informed application, you getmore risk on the lower than four-inch pipingthan on the bigger ones. There are someimportant facts that can be obtained by usingthis method. On the other side, if you areperforming that, you will normally reduce thescope of inspection by a drastic number.If you take into account that you are going to, in10 or 20 years, you are reducing the quantity ofinspection, you have to balance it. There arebenefits and the non-benefits to theapplication.

18

Susanne SCHULZ - We also had pilot studieson applying risk-informed procedures, one forthe Westinghouse plant and one for GE. Theresults of these procedures are completelydifferent. At the moment, we are not able tomake an interpretation of these findings so wedecided to go slowly in that direction.

José Maria FIGUERAS - I agree. It’s a caseby case application.

Karen GOTT - We have never had the ASMEcodes as the inspection procedures. If, andwhen, our utilities go over to a full risk-basedinspection procedure they are not going to wina lot of inspection. They are going to direct theinspection to areas where maybe the currentsystem is not capturing the sampling numbers.

Ann MacLACHLAN, NUCLEONICS WEEK,France - You mentioned a requirement for acost-benefit analysis to be integrated into thelong-term operating licence review. You didnot detail that point. Could you tell us what isthe importance of that or how is it done? Thesecond question is you told us that you arefollowing quite closely the NRC regulations, yetyou are preparing to issue a licence for another10 years. The question is: why not another20?

José Maria FIGUERAS - The first part on thecost-benefit analysis is something which hasbeen added by the utility. In essence, I guessit will not deal with the safety analysis of theapplication for a long-term period. There aresome aspects which could be studied. The aresome possible new regulations to be applied inthe next 10 or 12 years. It might be worthhaving some cost-benefit analysis; it is reallyimportant to add a new regulation on howmuch it implies in the balance of cost and thebalance of benefits to the safety.Related to the second question, we followAmerican law-making but not fully. Onereason we are not fully implementing that is, forinstance, we are using the European processof instruction for the safety review so we wouldlike to be more careful granting permits. Whenyou perform TLAA analysis the opportunitynormally will perform that for 20, or even morethan 20, years. For the regulatory parts, weprefer to grant permits every 10 years and notto extend, as our American colleagues say, to20.

Dominique ARNAUD, ASN France - Avantd’appartenir à l’Autorité de Sûreté Japonaise,NISA, où il travaille depuis 2003, MonsieurMaeda a dirigé le département de métallurgiechez Toshiba. Auparavant, sa carrière avait été

consacrée aux questions relatives à la chimiedans les réacteurs à eau bouillante. MonsieurMaeda vous présentera l’approche de l’Autoritéde Sûreté japonaise et la problématique duvieillissement et de la durée d’exploitation desréacteurs.

Katsuji MAEDA - I will show the situation ofthe ageing management in Japan.First, the background of ageing management.In Japan, the operating period is permitted onlyabout one year. We will also take account ofthe current legal obligation of implementationof ageing management. Now, we considereffective cooperation of industry and theregulatory side, and the academy. TheExamination Committee was implemented lastDecember to make effective ageingmanagement. We prepared an interim reportthis spring and the final report will bedistributed in August.

What is the background of ageingmanagement in Japan? 53 nuclear power

plants are operating inJapan, 30 nuclear powerplants, NPP, will have beenoperating over 30 years in2010. Some BWR plantshave been operating forover 40 years. So this kindof situation indicates theimportance of creating anadequate ageingmanagement programme.

We have nuclear power plants operating inJapan for over 35 to 40 years.

In Japan, the operating period is permittedunder this kind of law and regulation. This nextpiece of background information is veryimportant. Under Section 91 there is limitedoperation. Beyond 13 months, no nuclearpower plant can operate without passing alegal inspection. This is a very differentsituation compared to international laws.Again, I must say, in Japan, every nuclearpower plant can operate only 13 months. Thismeans that if the 13 month inspection can bepassed, any nuclear power plant can operatefor a very long time period, over 40, 50 or60 years.

There is a scheme of regulation related toageing management. Regulatory inspection isrequired every 13 months to continueoperation. PSR is a Periodic Safety Review.PSR is evaluated over the validity of pastmaintenance activity for plant safety andreliability. One of the parts of PSR – ageingmanagement – will be implemented when the

19

plant is operated just before 30 years.Validation is required every 10 years over 30years’ operation. PSR reviews maintenancemanagement adequacy of the past, ageingmanagement programmes are focused on thefuture.

Basic policy and concepts of ageingmanagement. The office of natural resourcesand energy prepared ‘The Basic Policy on theAgeing Management of NPPs’ in April 1996 butnow, in Japan, many BWR plants have longoperation of over 30 years. Basic policy,required that these kinds of items, implementageing management review before 30 yearsand established that 10 years’ long-termoperation and maintenance programme basedon the above technical evaluations. And thiskind of technical evaluation, the long-termmaintenance, should be evaluated every10 years after past ageing managementreviews.

What is the concept of ageing degradation andmaintenance? The ageing degradationphenomena copes with routine maintenanceactivities. Because, when ageing managementreviews are conducted by utilities, at that time,one of the major purposes is to review theeffectiveness of current database maintenancemanagement programme for the ageingdegradation. On this evaluation, there is somekind of additional requirement for long-termoperation. Key ageing phenomena, forexample some kind of degradation, will occurrapidly or will occur maybe in the future inmany parts or components. Such kind ofdegradation should be considered to be veryimportant.

Then we have the concept of ageingmanagement analysis: extraction andextraction systems structure, components andthe ageing phenomena. Ageing phenomena isextracted, not only domestic nuclear powerplant electric production, but also overseas andother industries and we can compare this kindof phenomena and systems. One of the majorviewpoints is the review of currentmaintenance programme adequacy.

And after the evaluation, PRS ageingmanagement for long-term operation isprepared. There are three items, one isconfirmation of current maintenancemanagement programme to be continuedadequately. The second is the extraction ofadditional maintenance programme. The thirdis extraction of R&D items. Ageingmanagement is covered in three categories:one is predict evaluation of ageing effect and

the second is surveillance inspection andmonitoring. And the third is repair andreplacement.

Utility implemented ageing managementreviewed in nine nuclear power plants inJapan. The government has evaluated theadequacy of these licences and reportsgovernment review for the nation. As regardsthe implementation of the ageing managementreview, nine NPPs have been reviewed forageing management. This shows the actualnames of the plants.

Ageing management has been legally specifiedas an obligation since October 2003. BeforeOctober 2003, ageing management was notconducted on a mandatory basis. But sinceOctober 2003, it has become legally specifiedbecause the nuclear power plants will operateover 30 years in the future.

Ageing management reviews should beevaluated every 10 years after the first ageingmanagement review. The ageingmanagement review is the one that is basedon quality management system. It is veryimportant to conduct it as part of a qualitymanagement system.

There has been improvement andconsolidation of ageing management. 53nuclear power plants were operating inJune 2005, at that time. And 20 nuclear powerplants will have been operating 30 years andsome of them will have operated 40 years in2010. So ageing degradation will be frequentlyactuated in ageing nuclear power plants,therefore more careful maintenance should berequired. Ageing management is a greatchallenge to ensure safety and integrity. Andanother viewpoint, as you know, where theMihama accident, a secondary pipe killed fourpeople. It was because of the flow acceleratedcorrosion in the pipe seam. There is no goodmanagement of ageing. This background hasdetermined that it is necessary to verify ifageing management has an appropriateresponse to the ageing effect and there-examination that the government shouldenforce to the ageing management. We alsohave actual improvement and consolidation ofageing management.

System strengthen of government. AgeingManagement Office was implemented in thegovernment last December. The mission ofthe Ageing Management Office was theConstitution of Guidance Documents asfollows: the constitution and the verification ofutilities, ageing management implementation

20

and so on. And another strength is theimprovement and conservation of ageingmanagement is system strengths of JNES.JNES means Japan Nuclear Energy SocietyOrganisation. And JNES’s mission is […]provides technical support to AgeingManagement Office of government. It is veryimportant to evaluate technical views, toevaluate licences, ageing management review.So AEO, Ageing Evaluation Office in JNES hasresponsibility for reviewing that.

There is a relationship between ageingmanagement and the guidance documents.Publication of the ageing management reviewreport is provided by licensees and thereflection to routine maintenance program. Onthe other hand, some kind of ageingmanagement document, control documentshould be prepared. A prescription guide ofbasic policy considerations for ageingmanagement is prepared by government, NISAand detail relative to ageing managementdocument. One is a standard examinationguidelines. It is similar to as a standard reviewby NRC. And another is the Generic AgeingTechnical Database. This kind of document isused and combined with consumerspecification. And some kind of consumerspecification, codes, guides or standardswould be prepared by the requirement of thebest policy consideration for ageingmanagement.

This document is applied for the technicalreview and ageing management programmeby licensees. After that, a publication aboutageing management review report bylicensees, NISA and the government shouldevaluate the adequacy of the licensees reportby using this kind of document.One of the more important and interestingitems is the Examination Committee for AgeingManagement. This Committee wasimplemented last December. The Committeehas been reviewed and discussed best policyon important matters such as clarification ofageing management, constitution of guidancedocuments and technical base for ageingmanagement. The Committee has met fourtimes and submitted a final report in August2005. These are the activities of theCommittee.

Future deliberations in the Committee. ThisCommittee will provide clarification of SSCs inthe scope of ageing management. The role ofPSR, direction and promotion of safety relatedR&Ds, active and effective collaborations,among industry and government and NISA.The combination of the three parties on the

industry side and the academia and thegovernment. This should make an umbrellanetwork. That network should be coordinatedby a responsible coordination function. Andthis consideration should produce informationfor the nation and other industries and makeinformation exchange for overseas.

Establishment of a technical information base.The concept of the technical database andageing management is an issue for the safetyand reliability of NPPs. This could beperformed by establishment of technicalinformation base. And that will be for researchand development. The results or experience ofresearch and development indicates thedirection of investigation and R&D result. Andafter that, synthetic technical information will beimplemented.

Development and consolidation ofmaintenance management and safety ensuringactivity for ageing management. So reflectingon actual operating experience is veryimportant. The development and review ofmaintenance activities should be spiralled,umbrella view, at the same time beingconsiderate of the time axis. Experience andtechnology will change and be accumulatedover time.

We also have to consider emphasis items.One is safety ensuring activity and optimumand rational maintenance. We must considerthe maintenance and the improvement of thetechnical capability by recruiting anddeveloping capable people.

In conclusion, constitution of guidelinesdocuments for ageing management. Thesethree guidelines will be prepared; and now weare discussing what kind of content should beimplemented or included in these guidelines.And that committee discussed the adequacy oftiming and a period of ageing management.Some kind of ageing management review isrequired, especially before 30 years operation.But some kind of ageing phenomena will occurbefore 30 years. So now, we have requestedto review the ageing management just before30 years. The Committee has discussed thesekinds of points.

Effective and rational safety regulations.Measures against non-physical ageing such assafety culture, technical transfer, humanresources, administration management,corporate culture, and organisational climate.It is one of the most important discussion itemsin the committees.

21

Worldwide effective collaboration. What kindof collaboration should be considered? ThisCommittee will prepare some kind of answerthis August. The final report will be publishedin August 2005. And these kinds of aboveitems will make some kind of concrete results.Thank you.

Dominique ARNAUD - Monsieur Michel quiappartient à l’appui technique de l’Autorité desûreté allemande va maintenant nousprésenter la problématique et les aspectstechniques du vieillissement des équipementssous pression en Allemagne.

Frank MICHEL, GRS Germany - Ladies andgentlemen, GRS acts as a technical support

organisation, in particularon behalf of the Germanfederal ministry for theenvironment, natureconservation and fornuclear safety, the BMU. Iwill spend first a fewminutes on the service lifeof German nuclear powerplants and the approach toageing management. Then

I will speak about the GRS knowledge base onpressurised nuclear power plant components.I will present the overall results of theevaluation of operating experience. I willinform you about the development of tools forquick access to access to information onageing degradation mechanisms by GRS.Last, but not least, a few words on recentregulatory activities on ageing management inGermany.

The nuclear power plants, presently beingoperated in Germany, were mostly constructedat a time when sufficient knowledge had beenobtained to avoid the detrimental aspects ofageing from the very beginning. Wedistinguish between four PWR, designgeneration and two PWR, construction lines.The operating time for the eleven PWRs ranksfrom 16 to 30 years, and for the BWR, from 20to 29 years. Since the year 2000, the lifetimein Germany is restricted by a so-called‘agreement’ between the federal governmentand the utility companies. Accordingly, themaximum electricity volume, which each plantis allowed to generate, is specified in principlesof volume, corresponds to a standardoperating life of 32 years. Moreover, theutilities can transfer their electricity volumesfrom one plant to another. However, thisrestriction led to the shut down of theObersheim Nuclear Power Plant last month,after 37 years of operation.

The German approach to address ageingissues is, in general, characterised by :

- continuous evaluation of operatingexperience to identify changes in thereliability of systems, structures andcomponents,

- extended plant monitoring to enhance theunderstanding of system behaviour of loadconditions.

- Evaluation of safety margins for lowerbound conditions,

- generic studies to identify areas of limitedknowledge and potential future problems.

- Early replacement of components,potentially sensitive to degradation andenforcing technical requirements in codes

- standards to avoid repetition ofnon-optimised technical solutions.

The GRS knowledge base on pressurisednuclear power plant components consists ofseveral modules containing comprehensiveinformation on codes and standards, designand material, operating experience, analysisand qualification methods. To use thisinformation more effectively, GRS has beendeveloping qualified databases and networkscalled KomPass with regard to the evaluationof the ageing behaviour of pressurisedcomponents.

Databases, KomPass and ALMA MATER playan important role. The database, KomPass,contains detailed information on the operatingexperience with pressurised components inGerman light water reactors. It is based onreportable events. And the current databasecontains about 800 safety related events,occurred in a time span of 30 years. Wedistinguish between events that occurred atdifferent components, such as pipes, vesselsand housings and also between differentsystems affected, such as the main heatgeneration system, or the auxiliary systems.On the basis of this database, GRS hasperformed detailed analysis of the ageingbehaviour of pressurised components. In thebeginning, this was done on generation

22

specific level, later, more specific evaluationswere performed.

What are the end results of the overalloperating experience with pressurisedcomponents in Germany? We distinguishbetween different design generations orconstruction lines and different operatingperiods. The overall number of safety relatedevents in Germany, at pressurisedcomponents, was low. Moreover, nosignificant increase of events with operatingtime has been recognised so far. A moredetailed evaluation shows the frequency of theevents involving piping in German nuclearpower plants due to fatigue. The differentareas indicate the significance of therespective annual frequency. In addition, it hasbeen investigated whether events due to aspecific mechanism are accumulated in anyplants and whether there are any indications ofsafety related shortcomings from thechronology of these events. Correspondinginvestigations were performed for all types ofrelevant damage mechanisms, as well ascomponent specific. The overall results can besummarised as follows. In the past, thepressurised components used in Germannuclear power plants have yielded experiencewith different ageing-related degradationmechanisms such as mechanical and thermalfatigue and several types of corrosion such asintergranular and transgranular stresscorrosion cracking, strain induced corrosioncracking and flow accelerated corrosion. Theoverall number of events, due to ageing relateddegradation, is low. Up to now, no significantincrease of ageing-related events withoperating time has been recognised.

A few words on the development of thedatabase system, ALMA MATER. Worldwideoperating experience and research has yieldeda large variety of information on ageing-relateddegradation mechanisms. However, ourpractical experience has shown that the quickaccess to this information often causesdifficulties. And for this reason, the databasesystem ALMA MATER is being developed byGRS, starting with a survey of relevantdegradation mechanisms, the browser-basednavigation give access to relevant informationon the individual damage mechanisms.Following a brief initial statement in which therespective mechanism is characterised withregard to its boundary conditions and damagesymptoms. The user is guided to moredetailed information, these lead to the 4categories : operating experience, state ofknowledge, codes and standards andyellow pages. In the survey results, we

differentiated between embrittlement,corrosion, fatigue mechanisms, as well assynergisms such as corrosion fatigue andirradiation assisted stress corrosion cracking.Moreover, materials susceptible andcomponents affected are compiled.

The lead into the category operatingexperience is via a so-called time bar, wherethe international and national operatingexperience with the corresponding mechanismis summarised for a time period of severaldecades. This was done for flow-acceleratedcorrosion in PWRs. The information about theinternational operating experience, which Ihave outlined, was taken from the OPDEdatabase which is available at GRS. TheGerman experience indicates that localdamage in various parts of the secondarysystems occurred during the ‘70s in Germannuclear power plants. It becomes clear thatwater chemistry plays the key role for a givendesign. And to avoid further flow acceleratedcorrosion, the utilities changed their turbinecondenser tubes, made from copper alloys, tostainless steel or titanium, creating suitableconditions for the evaluation and for theapplication of high or volatile treatment. And inconsequence, no safety-related damageoccurred anymore in these systems.

A few words on recent regulatory activities inGermany. In July, last year, the GermanReactor Safety Commission, issued arecommendation on the management ofageing processes at nuclear power plants. Itwas prepared on behalf of the federal ministry,BMU. It describes principles on theprocedures of managing ageing processes atnuclear power plants. It considers, in detail, allsafety relevant, not only physical ageingprocesses. And it contains requirements of anageing management system to be appliedduring the lifetime of German nuclear powerplants.

A few concluding remarks. The results of theinvestigation performed by GRS provide atechnical basis for the evaluation of ageingbehaviour of pressure-retaining components inGerman nuclear power plants that can be usedin licensing and supervisory procedure. So far,the limited number of ageing-related incidentsand the corresponding trends confirm theconservativeness of the safety concept chosenby the design as well as the sufficiency and theremedial actions and the ageing managementsystem applied. However, the currentknowledge of damage mechanism and thepredictive capabilities are limited. That’s whyfurther plants and component-specific

23

investigations are required, as well asprocedures to maintain a sufficient level ofawareness.

In future, German licensees need to addressageing management of nuclear power plantson a more comprehensive and detailed leveland have to submit periodical plant-specificreports on it, following correspondingrecommendations of the German ReactorSafety Commission. Thank you for yourattention.

Dominique ARNAUD - Monsieur EdmundSullivan va faire une présentation sur leprocessus de renouvellement d’autorisationd’exploiter à la NRC en regard desphénomènes de vieillissement. MonsieurSullivan travaille à la NRC et sa carrière a étéconsacrée à l’analyse de la sûreté industrielleet à l’analyse de la sûreté des systèmesindustriels, en particulier nucléaires.

Edmund SULLIVAN, NRC USA - Goodafternoon. I am going to talk about the nuclearpower plant licence renewal process in theUnited States.As was mentioned earlier this morning byR William Borchardt, there are 103 licensed

plants in theUnited States. Initially,they were licensed for a40-year term. And infact, a couple of thoseplants, by the year 2009,will have reached thepoint where the initiallicensing term will beexpiring. As with manycountries, NRC hasdeveloped a licencerenewal process with

requirements for extending the licence. In thiscase, as I think we discussed earlier thismorning, it is for an additional 20 years. Myunderstanding of this process is that it is notthe end. A plant can come back for a second,or possibly more, extensions.

Applicants for licence renewal mustdemonstrate that there are programmes inplace to manage ageing effects applicable topassive long-lived structural components haveadequate programmes in place. We basicallyrefer to a number of terms that will keepcoming up. The AMR, the ageingmanagement review and the other is ageingmanagement programme, another istime-limited ageing analysis.

The principal focus that I want to devote is tothe GALL Report. The GALL Report was

developed over a number of years but the firstversion was published in the year 2000. Ourheadquarters are working on a revision to thisGALL, which will be issued in September. Thisreport was issued to assist utilities indeveloping their licence renewal applications,LRA, and to assist staff in performing reviews.GALL includes a set of ageing managementreviews, with typical components, as illustratedon the slide. Ageing management reviews arepresented in a tabular format.

We review the applications for consistency withGALL and, insofar as these applications areconsistent with GALL, we give them credit.And that basically is the level of review that isdone. The GALL also contains a set ofprogrammes. The review consists of lookingat the components, the materials, theenvironments, the ageing effects and theprogrammes. In addition, there are detaileddescriptions of these programmes, whereageing management programmes, AMPs, andwhat the applicants need to do is identify thoseexceptions that they are taking to the criteria inthis report. Whether it is with respect to theway the ageing management review is done orwith respect to a particular programme.

An interesting feature you will see underageing management programme, ‘aplant-specific ageing management programmeis to be evaluated’. And what that basicallymeans is that in the GALL there have beencertain place-holders put in place where thereis no particular ageing managementprogramme which has been described in thatreport. In these cases, it is up to the licenseeto identify what programme they believe isappropriate for their station, so it is not allcompletely cook-booked into the GALL Report.Then there are other ageing managementprogrammes that you might be familiar with,such as the boric acid control programme,water chemistry programmes, stainless steeland so forth.

As I mentioned earlier, we are in the process ofupdating this report. I would like to talk briefly

24

about some of the reasons we are pursuingthis update. We found it necessary to increasethe understanding of the use of this report toavoid inconsistencies and to avoid the numberof exceptions that licensees are taking to thisreport by improving definitions. There is afairly lengthy section on definitions thatdescribe what we mean by different materials,terms and another section describing theenvironments. We have incorporated certainlessons into GALL. For example, there havebeen a number of industry standards andthings like chemistry guidelines that have beenimproved since GALL was first put together.These are folded into the new report. Therewere some inconsistencies that people werenoticing across systems and these have beenstraightened out. There has been a certainamount of consolidation.

In the current version, which is being updated,there were a number of locations where therewas essentially redundancy. A lot of materialcombinations could be put together orcomponent combinations could be put togetherto reduce the number of ageing managementreviews. I have given a couple of exampleswhere for some components, low alloy steelsand carbon steels grouped together. Or forexample, for some ageing effects – obviouslynot all ageing effects – we have been able tocombine precipitation-hardened steel and castaustenitic stainless steel (CASS). We havestandardised the terminology of environmentaleffects. We have established sometemperature thresholds for certainmechanisms. So for example, there is atemperature threshold for initiation of stresscorrosion cracking that we have put into thereport. And this has actually enabled somelicensees to not take exceptions because forexample, if a licensee had a stainless steel inthe application below the threshold, theywouldn’t really be able or willing to conform tothe existing GALL because it would direct themto a programme for management that wouldn’tbe necessary. So doing things like this hasreduced the number of exceptions that

licensees have had to take to GALL. Ageingmanagement of nickel alloys and reactorvessel upper-head penetrations. This is atopic that many people are familiar with andwas referred to extensively this morning. Thisupdate basically folded in the newrequirements of the order that NRC issued in2003.

We consider that the changes in the licencerenewal process and the update of theGALL Report will provide for a more efficientand approved basis for reviewing the licencerenewal applications.

A few words on logistics. It was suggested bysome of the organisers that we talk a little bitabout logistics. The NRC devotesapproximately 20,000 hours to each reviewthat we do. Who gets involved in doing thesereviews? I think first of all we might say thatthe application comes in to a group of projectmanagers. The project managers basicallyhandle the whole application through its reviewprocess. They also do a little bit of technicalwork in that they identify the exceptions thatthe licensee takes to the GALL or to theplant-specific line items that require a little bitmore detailed review. And they farm thosereviews out to the technical folks. There isalso a scoping review that is done to makesure that we agree with the way the applicationis constructed in terms of the components thatare covered. We have a group of systemsengineers that look at consistency with ourregulation in terms of scope. The engineeringgroup, of which I am a part, looks at theexceptions, the plant specific programmes andthe time limiting ageing analyses. There is awhole group that does audits. This is one ofthe efficiencies that we have evolved to inrecent years. Instead of the engineeringreview, doing reviews of these programmes,they are done on an audit basis by staff andcontractors who go to the licensee’s facilities.This process of doing audits basically cutsdown on exchange of questions and answersbecause these people who do the audits godirectly to the facilities and cut down on thepaperwork by talking directly with the peoplewho prepared the application. The last groupthat gets involved are regional inspectors.They have a programme of inspections forlooking at the effectiveness of theimplementation of the programme. In otherwords, they take the programmes to the nextlevel and make sure that they are beingimplemented in an effective way.

In terms of schedule, we set a date of22 months to complete the review. They are

25

very well prescribed, in a detailed prescribedfashion. So far, we have been able to adhereto that schedule without any problems. Thisschedule is the schedule if there is no publichearing. There is an ongoing process in theagency to continuously monitor ways toimprove the process; GALL revision 1 is thelatest outcome of that. But afterwards, we willcontinue to look for ways to improve thatprogramme and improve the document.In terms of programme status we havereviewed 32 applications for licence renewalsso far, and granted those applications. Wecurrently have 16 under review and we expectto continue to review about six applicationsper year until they are all completed, at least allthose interested in licence renewal. I think thatconcludes my talk.

Sophie MOURLON - Maybe we can starttalking about the last slides presented by TedSullivan. I think it is interesting, as we are hereas safety authorities, to talk about regulatoryaspects of ageing issues. One of the ageingissues for the safety authorities, I think, is howmany people, and how much time, can we,should we, must we, devote to these ageingissues. I am very interested in hearing whatGermany and Japan have to say about that.

Katsuji MAEDA - In Japan about two years forthe ageing management review. And themeeting, it will be conducted about 200 times.It is very hard work so, in Japan, we are nowplanning to establish standard procedure andstandard operation analysis method. Andprepare the same database with the utility andthe licences. They use a corrosion date, forexample, 15 MDD. But if that number iscorrect or not, though it is very difficult to judge.So now, we licensee, we needed to decide torecommend to establish the same databaseand the same analysis method.

Frank MICHEL - GRS has developedcomprehensive databases on the operatingexperience with regard to the ageing behaviourof components, first of all for mechanicalcomponents. Of course, we have databases

for other types of components too. This was awork which was done over a long period ofyears – I think we started 20 years ago with thefirst small database – and this has now beendeveloped. And now I think we are at thatpoint where we are able to get new data veryrapidly inside this and we have a structuredevaluation of this database so we are able tomake a new or updated evaluation without toomuch effort. Again, this is only operatingexperience. On this basis we are looking fortrends and things like this.

Sophie MOURLON - Are there any othercountries in the public which have studied orevaluated or decided on a way to handle thisissue?

Alain SCHMITT - I think that in France we aredealing with this subject in a framework ofperiodic safety reviews. The average time thatwe devote to the evaluation and the topic isfour or five years. After four or five years, wetake a position about ageing management.

Sophie MOURLON - And then we start overagain for the next periodic safety review.I have another question for Mr MICHEL. Thesituation in Germany is very specific. The kindof operation is limited, you are definitely limitedin time. You have not talked about eitherlicence renewal or periodic safety review after30 years because obviously the topic is not…Do you think it is a big difference?

Frank MICHEL - The lifetime is restricted forpolitical reasons to a standard time of about 32calendar years. As I said, it is a politicaldecision, I am not in a position to predict thepolitical situation next time. And so far, this isour current situation.

Sophie MOURLON - Do you mean that thesafety authority and the technical institutewould be ready to examine the possiblecontinuation of operation for German plants?

Frank MICHEL - On the technical basis we arevery well prepared. We have done a lot ofwork under the headline of ageingmanagement, not under licence renewal, andso far there is a lot on technical phrases toevaluate ageing behaviour and its influence onthe safety of the components.

Ann MacLACHLAN - You spoke of four tofive years for a periodic safety review for aspecific unit. What do you mean, for eachplant?

26

Alain SCHMITT - With a generic aspect inFrance because we have standardisedreactors.

Ann MacLACHLAN - Because others havespoken of two years or even less. My questionis: does standardisation extend the time periodneeded for the periodic safety review. It shouldshorten it for each unit logically.

Alain SCHMITT - I do not think you could saythat standardisation extends the time for aperiodic safety review. The depth of theanalysis of the periodic safety review may differfrom one country to another. Maybe it is themain explanation for the duration of theprocess. The idea also is that maybe we coulddo it faster but we take our time. We like toprepare things far ahead because then wehave many reactors undergoing periodic safetyreview. We have a schedule that starts someyears before the actual 10-yearly inspections ofthe reactors. I would not relate that to thenotion of standardisation. And then, of course,as was said by some of the speakers, theageing issues have a strong plant-specificaspect. We have to look at each plantspecifically.

Sophie MOURLON - Another question aboutthe databases. In France we only have onetype of reactor but, as we know, there arecountries with many types of reactors. Do youcross-reference the information? Do you shareit? Are there totally different issues from onetype of reactor to another?

Frank MICHEL - I think it is possible to make adistinction between PWR and BWR conditions.If you think about for example, intergranularstress corrosion cracking, it’s an issue forBWRs and not for PWRs. It is important tomake such distinctions and also to distinguishbetween different generations. It makes sensealways to ask if it is a generic issue or is it aplant-specific or a generation specific issue.

Ray NICHOLSON, HSE United Kingdom, UKRegulator - In terms of PSR timescales asapplied to UK PSRs. Essentially what happensin the UK that there is a very clearly-defineddate when the PSR has to be submitted to usand we have accepted it. If we have notcompleted our assessment at that date,because either the PSR is unsatisfactory or wehave raised additional issues, we will not givethe plant permission to continue to operatebeyond that date. Going back from that date Ithink it is about two or three years, we havediscussions with the licensees about the scopethat is going to go into the PSR, so they

produce a fairly detailed summary of what isgoing to go into the PSR from theirperspective. We add what additional items weexpect them to address and then theycomplete the PSR as such and then it comesto us for assessment. And it is over their twoto three year period, leading up to the decisiondate when the detailed assessment takesplace.

So I think in the UK, the timescale is perhapsconsistent with elsewhere, but we do have thisearly stage of discussions with licensees toensure that we know what they are planning onputting in the PSR and we can raise anyadditional items that we expect to seeaddressed. So it does perhaps reduce someof the iterations when the timescale is gettingrather short.

Dominique ARNAUD - Monsieur AlainSchmitt, directeur général adjoint de l’Autoritéde sûreté française, va conclure cette matinée.

CONCLUSION

Alain SCHMITT, ASN France - I will try toconclude these first two sessions briefly with afew ideas that I think came out of thediscussions. We had a whole range ofpresentations by regulators and by technicalsupport organisations about their experienceand their practices in the management ofageing.

The first idea that I would like to stress is that,overall, the objective of nuclear safetyauthorities and their technical supportorganisations, seem quite similar. And thisobjective is to ensure the safe operation ofnuclear power plants during their total lifetime.

The second idea is that regulatory processesseem quite different. They show quite a varietyof tools and procedures. For instance, somecountries use licence renewal processes todeal with ageing management. Others dealwith it during periodic safety reviews. We alsohave different uses of PSAs and the

27

risk-informed approach. Also, differentapproaches to in-service inspectionprogrammes. I think there are commonfeatures among all these approaches. Thefirst one is that ageing management isidentified as a specific issue by all regulators.The second one is that ageing managementprogrammes are required from the licensee toallow further operation. We have, in allcountries, it seems to me, something that lookslike ageing management programmes. Theseageing management programmes are basedon lists of components, lists of ageingmechanisms, and sometimes lists ofenvironmental conditions. We havedifferences between ourselves but alsocommon features. And I think we have to learnfrom each other about regulatory practices inageing management.

The third idea is about generic versusplant-specific aspects of ageing management.One very important theme is that ageingphenomena may very strongly differ betweenfacilities. Of course, between differenttechnologies, for instance, BWRs, PWRs,heavy water reactors, gas cooled reactors, notto speak of other nuclear facilities such as fuelfabrication plants. They also can differ verysignificantly between reactors of the sametype. And this is, I think, one important thingand this leads to the need for a plant-specificapproach to ageing management. This is onething that I think was present in allpresentations.

Also, we should recognise that ageingmanagement has a strong generic dimension.As one of the speakers said, nature hasimagination. The challenge for the licenseeand also for the regulator is to anticipate theoutcomes of this imagination. One of the waysto anticipate is to share operating experienceon a very extended basis. It is fundamental tokeep a questioning attitude and to think oftransposing the conclusions of thephenomena, which have been seen on onetype of facility to others. For me, one veryimportant and very historically fundamentalexample is stress corrosion cracking ofnickel-based alloys on reactor vessel heads. Italso shows the importance of maintainingdatabases, and not only maintainingdatabases, but also to use them properly,keeping a qualitative way of assessing thingsand not only relying on indicators and ondatabases.

One last aspect relating to this idea is the useof regulatory research. It is a topic that wasn’ttouched on very often this morning. I think

there is a specific workshop this afternoonabout this aspect. What should we do inrelation to regulated research? What shouldregulators do, what should regulators have thelicence to do?

The last idea is that the recognition of humanfactor-related issues in ageing management.The loss of skills and know-how in the nuclearindustry, and sometimes in the regulator,sometimes supplies may disappear,sometimes components may becomeobsolete. It is also an important aspect ofageing and certainly, the regulators should beinterested in this aspect and extend theirsupervision to these kinds of problems. Thiswill conclude my summary of this morning’sdiscussions. Thank you for your participation inthese discussions.