A TECDOC onSeismic Soil Structure Interaction – IAEA draft TECDOC (Chapter 1-7 of the R0 Version))
COMMENTS BY REVIEWERReviewer: J.Ducau JP. Tardivel
Country/Organization: FRANCE / IRSNDate:February 2017 / RESOLUTION
Alain Pecker (AP)
JJJohnson (JJJ)
Comment No. / Page/Section / Comment / Accepted / Accepted, but modified as follows / Rejected / Reason for modification/rejection
1 / 15/ 1.3 / Design includes…design etc to an existing facility for DBE and BDBE.
JJJ-comment: I do not know exactly to what is referred here – need clarification from other team members; the issue of using the term DBE (and in the old days OBE/SSE – OBE/SSE is still used in various documents used internationally, e.g., in OBE exceedance definition and criteria) is somewhat universal – I recommend we state that DBE in this TECDOC refers to SL-1 and/or SL-2 as appropriate in IAEA terminology – SL-1
We suggested to remove in a separate bullet “Assessments for the forensic analysis of facilities that experience significant earthquake ground motions.” / X / X / We can remove the bullet on “assessments for the forensic analysis….” Here, but later in the TECDOC we need to add a short paragraph summarizing that sucha analyses are to be performed by the most realistic techniques to predict responses in the plant due to the earthquake ground motin at the site.
2 / 15 / 1.3.1 / Design requires a certain amount of conservatism Design exhibits a certain amount of conservatism. / X (JJJ agree with AP) / Yes design exhibits a certain amount of conservatism but it shall deliberately incorporate conservatism to cover future (unforeseen) evolution of the facility (modification of the design actions, openings in bearing walls for pipelines, etc…)
3 / 15/ 1.3.1 / Performance goals may be defined -> SSI is part of the seismic behaviour demonstration, related to safety objectives and performance goals coming from the safety analysis.
JJJ-comment: Boris added/modified the text as follows: “Safety objectives and” performance goals are may be defined:… / To be discussed by team
Boris’s change/
addition is good. / “behaviour”, as recommended here, has a very similar meaning to “performance”, e.g.., the performance (behaviour) of a structural member under the DBE loading conditions is to remain elastic.
4 / 15/1.3.1 / In France, the performance goals are defined deterministically and at the SSC design, qualification, and evaluation level.
In France, seismic safety requirements and associated behaviour requirements are defined according to the safety analysis. As a general goal, earthquake should not lead to an accidental situation (unacceptable releases).
JJJ-comment: this addition should be linked to Comment No. 6, which is very good!! / To be discussed by the team – which earth-quake is referenced here? The DBE? Or BDBE? Or DEC? or all EQs? / “behaviour”, as recommended here, has a very similar meaning to “performance”, e.g.., the performance (behaviour) of a structural member under the DBE loading conditions is to remain elastic; the performance (behaviour) of a structural system is to limit inelastic behaviour defined by story drift to less than 0.01.
5 / 16/ 1.3.1 /
- Establish performance goals, or develop a procedure to establish performance goals, for seismic design and beyond design basis earthquake assessments for SSCs.
- Partition achievement of the performance goal into elements, including SSI.
- Develop guidance for SSI modelling and analysis to achieve the performance goal.
- Establish behaviour requirements, or develop a procedure to establish behaviour requirements, for seismic design and beyond design basis earthquake assessments for SSCs.
- Partition achievement of the behaviour requirements into elements, including SSI.
- Develop guidance for SSI modelling and analysis to achieve the behaviour requirements.
6 / 17 / 1.3.3 / French practice:
New version was discussed and nearly agreed in a new version of paragraph 1.3.3 during december meeting.
We suggest some additional clearance as follows:
In the past and to now, the generalfrench practice for seismic design and periodic (about every other 10 years) safety assessment differs from the US practice in that the focus is based on a deterministic definition ofSMS (safe shutdown earthquake), designground motion and on adeterministic seismic analysisof safety SSCs :
i.The site specific design earthquake ground motion is defined from a mainly deterministic seismic hazard analysis (RFS2001-01);
ii.The seismic demand and capacity checks are required to be performed by deterministic approaches (ASN Guide2/01).
On the other hand, french practice aims to use probabilistic safety analysis in specific domains, when needed,as a complementary approach. Few specific probabilistic seismic analyses have been performed by operators and/or the regulator TSO in the last years; some others have started recently and are still in progress.
Forthe post Fukushima checks regarding the hard-core components in case of Beyond Design Basis Earthquake so called “SeismeNoyauDur” (SND), the Regulator has specifiedgeneral conditions as follows:
- the definition the seismic ground motion as the envelope of 150% the site specific Design Basis Earthquake motion and of a probabilistic motion with a return period of 20,000 years (annual frequency of exceedance of 5 x 10-5);
- the use of deterministic demand and capacity checks, includingspecific criteria in a step by step approach from design to more realistic and less conservative practices, consistent with the hard-core components functionality.
7 / 25 / 2.7 / 2.7 UNCERTAINTIES
It seems that this chapter describes mostly probabilistic approach
We suggest a specific chapter or sub-chapter describing uncertainties (parametric and enveloping variability of parameters) in SSI deterministic approach / X / See Chapter 2, 7, 8
8 / 51 / 3.7 / 3.7 UNCERTAINTIES
If additional testing is performed the range could be narrowed to values obtained by multiplying or dividing the mean value by (1+COV) where COV is the coefficient of variation, to be taken larger smaller than 0.5. / X (JJJ agree with AP) / The cross hole tests realized within the framework of the InterPacific project, and described in the TECDOC, show that under the best conditions the COV is equal to 0.3. Therefore a minimum value of 0.5 is reasonable; in addition this is the minimum value recommended in IAEA NS-SG-3.6
We can rephrase the sentence as
“If additional testing is performed the range could be narrowed to values obtained by multiplying or dividing the mean value by (1+COV) where COV is the coefficient of variation; under no conditions COV should however be taken smaller than 0.5.
9 / / chaper 8 / As discussed during december meeting, additional Chapter 8: “Seismic Response Aspects for Design and Assessment” is expected to be added to the SSI TECDOC (Chapter 9: Available software and Chapter 10: Sustainable capacity building were still in discussion).
We have suggested that chapter 1.3 and chapter 8 could be linked, may be even merged. / To be discussed by the team.
Page 1 of 5