ML23325A105
| ML23325A105 | |
| Person / Time | |
|---|---|
| Issue date: | 10/19/2023 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| NRC-2581 | |
| Download: ML23325A105 (1) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Reliability & PRA Subcommittee Docket Number:
(n/a)
Location:
teleconference Date:
Thursday, October 19, 2023 Work Order No.:
NRC-2581 Pages 1-162 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W.
Washington, D.C. 20009 (202) 234-4433
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com 1
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
(ACRS) 5
+ + + + +
6 RELIABILITY & PRA SUBCOMMITTEE 7
+ + + + +
8 THURSDAY 9
OCTOBER 19, 2023 10
+ + + + +
11 The Subcommittee met via Video 12 Teleconference, at 8:30 a.m. EDT, Vesna Dimitrijevic, 13 Chairman, presiding.
14 15 COMMITTEE MEMBERS:
16 VESNA DIMITRIJEVIC, Member 17 RONALD G. BALLINGER, Chair 18 VICKI BIER, Member 19 JOSE MARCH-LEUBA, Member 20 ROBERT MARTIN, Member 21 DAVID PETTI, Member 22 JOY L. REMPE, Member 23 THOMAS ROBERTS, Member 24 MATTHEW SUNSERI, Member 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 ACRS CONSULTANT:
1 DENNIS BLEY 2
STEVE SCHULTZ 3
4 DESIGNATED FEDERAL OFFICIAL:
7 ALSO PRESENT:
8 VICTORIA ANDERSON, Public Participant 9
KEITH COMPTON, RES 10 SUSAN COOPER, RES 11 JONATHAN EVANS, RES 12 ALAN KURITZKY, RES 13 EDWIN LYMAN, Public Participant 14 SCOTT MOORE, ACRS 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 CONTENTS 1
Page 2
I. ACRS Subcommittee Chairman Opening Remarks 3
Vesna Dimitrijevic 4
4 II. Introductory Remarks 5
6 III. Project Status Overview 7
Alan Kuritzky...............
6 8
IV. Overview of Reactor At-Power PRA Results 9
for Internal Fires, Seismic Events, High 10 Winds, and Other Hazards 11 Alan Kuritzky............... 54 12 BREAK...................... 93 13 V. Overview of Reactor At-Power PRA Results 14 for Internal Fires, Seismic Events, High 15 Winds, and Other Hazards (Continued) 16 Alan Kuritzky............... 93 17 VI. Public Comment 150 18 VII. Future Interactions 19 Alan Kuritzky..............
155 20 VIII. Meeting Adjourns 162 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 P-R-O-C-E-E-D-I-N-G-S 1
8:30 a.m.
2 CHAIR DIMITRIJEVIC: Good morning. It's 3
8:30, according to my time, so this meeting will now 4
come to order. This is the Reliability and PRA 5
Subcommittee Meeting in preparation for the Advisory 6
Committee on Reactor Safeguards Review of the NSC 7
Level 3 PRA project.
8 I'm Vesna Dimitrijevic, Chairman of 9
today's Subcommittee meeting. Members in attendance 10 are Bob Martin, Dave Petti, Joy Rempe, Matt Sunseri, 11 Ron Ballinger. I saw the message from Vicki Bier that 12 she was able to sign in. Vicki, are you there?
13 MEMBER BIER: Yes, I'm on, Vesna.
14 CHAIR DIMITRIJEVIC: Wonderful.
15 MEMBER BIER: Thank you.
16 CHAIR DIMITRIJEVIC: Wonderful. Okay.
17 Did I miss anybody?
18 MEMBER MARCH-LEUBA: Yes, Jose. Jose is 19 here too.
20 CHAIR DIMITRIJEVIC: Oh yes, Jose.
21 MEMBER ROBERTS: Yes, Tom Roberts is here 22 too.
23 CHAIR DIMITRIJEVIC: Okay. All right, 24 excellent. Tom Roberts and Jose March-Leuba are also 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 joining us. So some of our members are traveling so 1
will not be able to join us.
2 We hold this open meeting to gather 3
information to support our review of the NSC Level 3 4
PRA projects. The ACRS section of the U.S. NRC public 5
website provides our charters, bylaws, agendas, 6
reactor reports and full transcripts of all full and 7
subcommittee meetings, including slides presented 8
there. The meeting notice and agenda for this meeting 9
were also posted there.
10 The Subcommittee will gather information, 11 analyze relevant issues and facts and formulate 12 proposed position and action is appropriate for 13 eliminating by the Full Committee. A transcript of 14 the meeting is being kept and will be made available.
15 Today's meeting is being held virtual with 16 remote Microsoft Teams capability. There is also a 17 bridge line allowing participants over the phone.
18 When addressing the Subcommittee, 19 participants should first identify themselves and 20 speak with sufficient clarity and volume so that they 21 may be readily heard. When not speaking we require 22 the participants mute their computers, microphone or 23 phone.
24 Okay. So we will now proceed with the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 meeting. And I call up on Jonathan Evans, PRA Branch 1
Chief for NRC Office of Nuclear Regulatory Research to 2
begin today's presentation. Jonathan?
3 MR. EVANS: Hi, good morning. Thank you, 4
Vesna.
5 CHAIR DIMITRIJEVIC: Good morning.
6 MR. EVANS: So good morning to everybody.
7 My name is Jonathan Evans. I am the Branch Chief for 8
the Probabilistic Risk Assessment Branch in the Office 9
of Nuclear Regulatory Research. I just want to thank 10 the ACRS for the opportunity just to have us present 11 and have this opportunity to discuss the Level 3 PRA 12 project overview on the Volume 4 reports.
13 What I wanted to do is just a few moments 14 just to thank the Staff in PRAB, and also in the rest 15 of just the Agency who contributed to this project.
16 This has been a very herculean effort and just wanted 17 to say that we appreciate your efforts. And we look 18 forward to answering your questions today from the 19 ACRS. And with that, I'll turn it back over to Vesna 20 or to Alan.
21 CHAIR DIMITRIJEVIC: Okay, thank you.
22 Alan?
23 MR. KURITZKY: Thank you. Thank you very 24 much, Vesna. Dr. Dimitrijevic and Jonathan. I just 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 want to echo, you know, Jonathan's sediments. We very 1
much appreciate the opportunity to brief the 2
Subcommittee. I recognize that you guys have your 3
hands full with all kinds of exciting new and advance 4
reactor work, and so we welcome, or appreciate the 5
fact you're able to make time for us today.
6 We also had, oh. My name is Alan 7
Kuritzky. I'm from the division of risk analysis in 8
the office of research. I am the program lead for the 9
Level 3 PRA project.
10 And we briefed the Subcommittee last year, 11 I think in June, about Volume 3 of the Level 3 PRA 12 project. Today we're back to review on, essentially 13 an update on the project task, and also Volume 4 of 14 the project.
15 And let me just get here. So what we hope 16 to do today is to go over quickly the status of where 17 the project stands near to the end of the project.
18 Also, go over what public reports we have issued and 19 what other ones will be coming forward.
20 I want to discuss, to some level of 21 detail, but probably not excruciating detail, but the 22 information that's in the overview report for Volume 23 4 which is the report that addresses the reactor at-24 power, PRA models for internal fires, seismic events 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 and high winds. And also addresses other hazards 1
which we didn't model. And then the last thing we 2
want to discuss today will be future interactions 3
between the project and the Subcommittee and the Full 4
Committee.
5 With that, I do want to acknowledge that 6
a lot of work has gone into this project over the 7
years. We've had some really excellent support from 8
a number of organizations. Both within and without, 9
outside the NRC.
10 So inside the NRC, NSIR, NRR, NMSS, the 11 regions, TTC, everybody has had a hand in this and we 12 greatly appreciate their work. Some of these 13 organizations have had more input to the project than 14 others, but everyone, all of these organizations have 15 really done a great job in helping us out.
16 In terms of national labs, Idaho has been 17 our main contractor. Our National Lab for this work.
18 But also we've gotten a lot of support from Sandia 19 National Laboratories, Pacific Northwest National Lab, 20 and Brookhaven.
21 In terms of commercial contractors, NRG 22 Research Incorporated has been our major contractor, 23 but we also have gotten support from Applied Research 24 Associates. And also IESS, Innovation -- Innovative 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 Engineering and Safety Solutions I think.
1 The PWR Owner's Group also support us 2
later on supporting, organizing and running and 3
funding some, a PRA standard based peer reviews for 4
some of our early models. And Westinghouse and EPRI 5
have both been supportive of the project. In fact, 6
they have members on our technical advisory group for 7
the project.
8 And of course the ACRS. We've gotten a 9
lot of feedback. We've had, I've lost track now, this 10 is probably close to 20 minutes that we've had with 11 the ACRS since joining the project. Many of them were 12 early on pre-pandemic when we were doing a lot of 13 technical work. We had a lot of meetings to go over.
14 And closed meetings because of the proprietary 15 information. We had closed meetings to go into the 16 technical details of every aspect of the project. And 17 we greatly appreciate the time and the feedback we 18 received from ACRS members.
19 CHAIR DIMITRIJEVIC: Alan, sorry for 20 interrupting you --
21 MR. KURITZKY: Yes.
22 CHAIR DIMITRIJEVIC: -- but I noticed that 23 you have the slides acknowledgments. Have we uploaded 24 the slides because I don't see any slides?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 MR. KURITZKY: Oh, these are not being --
1 oh, wait a minute. I'm not sharing this with you?
2 CHAIR DIMITRIJEVIC: No. I don't see them 3
so --
4 MR. KURITZKY: Oh. No, you're right.
5 You're right, I have them on my screen but I didn't 6
share. Hold on one second. I apologize.
7 MR. BLEY: We are seeing you, Alan.
8 CHAIR DIMITRIJEVIC: We see you very 9
nicely.
10 MR. KURITZKY: Okay, sorry.
11 CHAIR DIMITRIJEVIC: Okay. So I heard 12 that Dennis is also here with us today.
13 MR. KURITZKY: Yes. Yes, I forgot to do 14 the share. My apologies. I'm glad you mentioned it 15 now and not ten slides into the presentation.
16 CHAIR DIMITRIJEVIC: Okay.
17 MR. KURITZKY: All right, thank you very 18 much. Okay, does everybody see the slides now?
19 CHAIR DIMITRIJEVIC: Yes.
20 MR. KURITZKY: Okay. Okay, fantastic.
21 Oh, also, that reminds me too. Because the slides are 22 on my laptop right in front of my face, that's where 23 the camera is so I'm facing that direction, so I do 24 not see all the other stuff like the Teams Meeting 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 window and stuff, people raise their hands, et cetera.
1 That's all to the side so I don't see that in my field 2
of view.
3 So, Jonathan, if I could again impose upon 4
you to let me know if anybody raises their hands or 5
makes a comment, and if you could just interrupt me 6
and pass it along. And everybody else, please feel 7
free to interrupt me, I don't mind interruptions. So 8
since I don't see the Teams screen, by all means feel 9
free to verbally jump in with any questions or 10 comments you may have.
11 MR. EVANS: All right, no problem.
12 CHAIR DIMITRIJEVIC: I think probably we 13 will help you with this.
14 MR. KURITZKY: Okay, thank you. Okay, so 15 now that we can see the slides, let me move on to, one 16 caution I want to bring out up front is that the Level 17 3 PRA project study is basically a state of a practice 18 study.
19 There were some areas where we had to push 20 the state of the art just because there was no real 21 state of practice. For instance post-core damage, 22 human reliability analysis.
23 However, because of limitations, either in 24 time, resources or the fact that we didn't have enough 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 information, we did have to make some simplifications 1
or assumptions in some areas and so therefore it's 2
important to recognize that even though we address 3
something a certain way in the study, that does not 4
mean that that's the way everybody else should do it.
5 And it's particularly if you're talking about a 6
regulatory application.
7 Just because we do something a certain way 8
in this study does not necessarily mean that it's okay 9
for regulatory purposes. So we just want to have that 10 caveat. And this caveat of course shows up in every 11 report of the project we have. A couple places in the 12 report.
13 MR. BLEY: Alan?
14 CHAIR DIMITRIJEVIC: Alan.
15 MR. KURITZKY: Yes?
16 CHAIR DIMITRIJEVIC: Okay, Dennis, go.
17 Go.
18 MR. BLEY: Yes, Dennis Bley. Two things.
19 And we have talked about this in the past with you 20 folks. With this published and on the street, 21 essentially disavowing it for regulatory purposes 22 seems a little strange. And I can't imagine that 23 people won't look to it to see what NRC is considering 24 as state of the practice approaches. I guess all 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 you're telling them is just look at the guidance, 1
don't look at this.
2 The second thing along this line, we got 3
to look at your slides a little ahead of time. I 4
don't see any that really talks about your first 5
bullet there. Summarizes what things you intended to 6
do and we were expecting to see that are not going to 7
be part of this study. And if you can address that 8
one I'd appreciate it.
9 MR. KURITZKY: Okay. And thank you, Dr.
10 Bley. And yes, you're right, this is something we've 11 wrestled with throughout the project as, you recognize 12 that this project will be looked at to people in the 13 technical community as to ways to address a whole, a 14 spectrum of PRA related issues. Particularly in those 15 areas that have not been routinely practiced before.
16 Obviously Level 1
17 Everybody and their brother have done them, and their 18 sister, had done these things for years and so they're 19 not necessarily looking to this project to tell them 20 how to do those. There are standards out and 21 everything else.
22 But some of the more novel areas that 23 we've included, like spent fuel pool and multi and 24 risk. Obviously people will be looking to that to see 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 what the NRC has done. And we do want them to look to 1
this report to see what we've done.
2 It is a state of practice study but it 3
does roll up all that state of practice in one big 4
study that looks at all of different scope elements.
5 And as I said, in some areas we have advanced the 6
state of the practice which is good for people to be 7
aware of.
8 But we do have to caveat the fact that, as 9
I just said, there are some areas, hopefully not a 10 lot, but some areas where we had to make assumptions 11 or do something in a more simplified manner and so we 12 don't want to have a blanket, we want to make sure 13 that there is no blanket expectation that whatever it 14 is in this study is okay for people who want to, you 15 know, for instance, submit a risk-informed license 16 amendment. You know, it's not necessarily that what 17 we did in this approach is applicable or appropriate 18 for that application.
19 In many cases it should be, and hopefully 20 will be. But as you just mentioned, regulatory 21 guidance supersedes. There is nothing in this study, 22 this research study, is not intended to supplant 23 regulatory guidance.
24 There may be a time when the NRC looks at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 the results of the study and decides to amend the 1
regulatory guidance, in which case then that will 2
filter out to people wanting to use the approaches 3
that are in this study, but unless and until that 4
occurs, existing regulatory guidance is much to be 5
followed.
6 So again, much of what's in this study, we 7
want people to look to the study. Much of what's in 8
the study should be useful and educational and 9
beneficial to people doing PRA in many of these areas, 10 but we just don't want there to be the expectation 11 that just something is done a certain way in this 12 study that it's necessarily appropriate for a 13 regulatory purpose. That's the response --
14 MR. BLEY: Thank you.
15 MR. KURITZKY: -- to the one comment. The 16 idea of what things that are covered and what are not.
17 This category is really more at a lower 18 level to how we treated specific items. So there is 19 no rule up to that, except in the fact that there are 20 many areas, each of the detailed topical reports has 21 sections on the areas where there is modeling 22 uncertainties, which is really where these things 23 would fall under. Or candidates for future research 24 or additional investigation, which is the other place 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 where you would see these things and those two lists 1
overlap significantly.
2 So that's where you'd see these things.
3 But there is many of them, each aspect of the study.
4 Not all of them are simply locations or limitations 5
but they are areas that we've identified and have been 6
either accepted as uncertain elements of PRA to date 7
or new ones that we've ran into as we did this work.
8 CHAIR DIMITRIJEVIC: Alan --
9 (Simultaneously speaking.)
10 CHAIR DIMITRIJEVIC: -- has raised their 11 hand. Joy?
12 MEMBER REMPE: Oh yes, thank you. When I 13 was looking through the summary report in Section 4 I 14 got the impression that you were going to get a lot of 15 great insights that could be used to maybe simplify 16 future PRAs when people looked at it and said, well, 17 this wasn't found to be important. And I, maybe it's 18 coming later in another report, but I didn't see any 19 specific list of items that said, okay, you know, we 20 spent a lot of time modeling something or other and it 21 wasn't that important and it doesn't need to be model 22 of that detail. Did I just get the wrong impression 23 or is it going to be something that comes later?
24 MR. KURITZKY: So yes. Partly yes and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 yes. It is something that's going to come later. The 1
summary NUREG volume is where we're going to take a 2
look at the various insights from the study.
3 Right now, any insights that we've gleaned 4
are more, for lack of a better term, stovepipe to the 5
specific models that we're looking at. And 6
particularly for those areas where many PRAs, I just 7
recently mentioned have been done already, like Level 8
1 internal event LERF, internal event type PRAs, there 9
are not a lot of earth shattering insights that are 10 coming from this work, as this work, this is well trod 11 ground.
12 But there are going to be some novel areas 13 where we hope to have learned things that we can share 14 and would be insights that would be useful for the 15 technical community at large. And that, and 16 particularly we start to look at the medical level 17 and, for instance, how things for reactor at power 18 compare to reactor at shutdown or how things for the 19 reactor compared to the spent fuel pool, the fuel pool 20 to dry cask storage, et cetera.
21 And those types of items will not come 22 until we do this more project-wide analysis at the 23 end. And will be documented in that summary NUREG 24 volume which I'll talk about in a minute. But Volume 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 1 will be the summary NUREG. And that will address 1
more of these higher level items.
2 In terms of specifically insights that 3
would lead us to, say hey, here is something that you 4
did a lot of work on, it didn't really make much 5
difference, it's not worth paying attention to in the 6
future. I don't know if we call out things 7
specifically in that way.
8 I think if people look through the list of 9
results and insights they may see where we mentioned 10 certain things did not tend to be important. It was 11 something that we were uncertain about, we evaluated, 12 it did not end up showing to be important, 13 particularly we have sensitivity analyses to kind of 14 demonstrate the impact --
15 (Off microphone comments.)
16 MR. EVANS: Hey, I'm sorry, it looks like 17 not everyone is on mute. Can we make sure that we're 18 all on mute for this? Thank you.
19 MR. KURITZKY: So, in any case, so if you 20 look at some of the sensitivity studies or some of the 21 discussion on areas of uncertainty, that's where you 22 might be able to find something and say, hey, we don't 23 really need to focus too much and this isn't really 24 showing to be important. But it's not like we're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 going through an existing, using existing PRA and 1
saying, okay, here is Item 1, 2, 3, 4, 5, all these 2
things, take it out of the PRA because they're just 3
not important.
4 That really wasn't the goal though. That 5
might fall out in just looking at some of the results.
6 It might seem that certain things aren't as important.
7 And it wouldn't be that you would necessarily rule 8
them out of the PRA but you might not put much, you 9
weren't worry about doing a lot of rigorous analysis 10 to refine the numbers for some things just because 11 they don't make that much difference.
12 But again, there is a difference between 13 what is important for this PRA, which is based on a 14 specific reference site and plant, and what might 15 apply to other plants. So things that we discover in 16 this study certainly apply to the reference plant.
17 They may in fact apply to many, the 18 reference plant is a PWR Westinghouse four loop plant 19 large dry containment. So other plants of that design 20 might also think things may apply to them. Not 21 necessarily but might.
22 Less likely, but still possible, it might 23 apply to PWRs as a whole or all reactors as a whole.
24 So again, you just have to be a little careful that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 the insights that we get from this study, some of them 1
we think will be applicable for the industry at large 2
or other sub-populations of plants or sites. But 3
they're not necessarily going to be universal, many of 4
them might be site specific or plant specific.
5 MEMBER REMPE: So, Alan, I was just trying 6
to cite an example, but I guess where I'm going with 7
my comment is that yes, I realize there may be some 8
caveats but there is such a large amount of material.
9 I would encourage you to think, waiting to the final 10 summary report might make it difficult and may miss 11 some things.
12 If there had been some more summary 13 highlights in this volume, that's the summary for the 14 Section 4 of external events and things, I think it 15 would have helped readers and others to say that, you 16 know, some, to have an interim summary insight that, 17 you know, with some caveats that it may not apply to 18 everything, but it's very hard to get, to extract that 19 is where I was trying to go with my comment, okay?
20 MR. KURITZKY: Yes. And I appreciate 21 that, Dr. Rempe. And I agree. I mean, that's 22 something, again, that we've wrestled with because we 23 don't necessarily have anything just to wait till the 24 very end, but the problem is that, as I said, many of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 the insights are going to be more project-wide, which 1
do have to wait till kind of the end. Even the ones 2
that would apply to some of these areas.
3 The first reports that we had done, again, 4
are areas that are well trod and there is not a lot of 5
earthshattering new insights. Also, as I went to look 6
at some of the material to try and extract some 7
elements it really, you know, there is just so much, 8
as you mentioned, there is so much material that is 9
not going to be some like quick list of like five, 10 here is the five. So here is the big five and boom, 11 boom, and so everybody can see that.
12 There is just lists and lists and every, 13 not just in every report but, and often times multiple 14 sections of report because there is so many different 15 areas covered. And every area could have a fairly 16 extensive list of areas of either modeling uncertainty 17 or key assumptions that in some cases we try to give 18 relevant importance to them, high, medium, low. Other 19 cases it's more of just a list.
20 And to try and extract those and compare 21 it to each other, like this item for the Level 2 PRA, 22 how important is that compared to this item we have 23 here for the Level 1 intro fire PRA. You know, it's 24 just an exercise that we just didn't have time for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 right now. And we're hoping that when we get to the 1
focus on the summary NUREG we can sit back and now 2
look at a more holistic way and try and identify some 3
of those things.
4 But again, we're almost handicapped by the 5
sheer volume of information we have. As you mentioned 6
it's difficult for the reader of course to try to suss 7
out that information. It's honestly difficult for us 8
too.
9 And to echo your concern about waiting 10 till the end, not only does that make the stakeholders 11 or the readers of the reports wait a longer time to 12 get that information, but unfortunately many of the 13 technical leads for this work, the ones who really 14 know it, have since moved on. Either they've left the 15 project, they've left the agency, so there is, yes, 16 we're losing some of the information with the exit of 17 those people too.
18 So I'm very sensitive to your concern. I 19 just, we just haven't really had the ability, as we're 20 trying to crank these things out, to step back and do 21 that at this point because it would not be an easy 22 exercise. But I am very sensitive, and I agree with 23 your concern.
24 MR. BLEY: Alan?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 MR. KURITZKY: Yes.
1 MR. BLEY: Dennis Bley. I've got three or 2
four things, kind of high level. But I want to follow 3
up Joy's questions.
4 And, you know, one thing is that after 50 5
years doing PRAs of one kind or another, we've seen 6
that risk is very much plant specific, even with 7
similar kind of facilities. I think being able to say 8
X, Y and Z are not important, don't look at them, is 9
probably not the answer.
10 But one thing you might have insight into, 11 and might include in your summary report, are if 12 you've actually done this. If you looked at some of 13 these issues to decide if you would include them or 14 not and came up with some approximate approaches that 15 were helpful in eliminating some things or de-16 emphasizing some things from the scope, I think that 17 could be very helpful to most practitioners.
18 The real question I wanted to get to, well 19 the three, were, you mentioned the opportunities for 20 future research that you point out. Are any of those 21 active at the current time or are these things that 22 will be a follow-on at some point in the future?
23 MR. KURITZKY: Well, the intention for us 24 is just that these are things that would be done in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 the future. So out of, in other words, they're really 1
just out, we're specifying that they're out of the 2
scope of this project. Now whether some of them are 3
already being worked on, some are.
4 And there are different some areas that 5
we've acknowledged. And this is an area that current 6
research is ongoing and so it's acknowledged that 7
there is current research in the area. Others are, 8
here is something that would benefit from additional 9
research. Most of those may not have started yet, 10 though so many have. It depends.
11 If it was just purely associated with this 12 project it's not like that it's been started already, 13 but if it already was something that was being thought 14 about in other areas of the Agency and they already 15 may have started work on that, in that area, then it 16 would be ongoing.
17 MR. BLEY: Okay.
18 MR. KURITZKY: But we, you know, we have 19 not initiated follow-on projects based on the results 20 of this project yet.
21 MR. BLEY: Okay.
22 MR. KURITZKY: That's something that 23 would, the Agency as a whole, other decision makers 24 would weigh in on probably after the project is 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 completed.
1 MR. BLEY: You mentioned something that 2
many of the members might not be familiar with because 3
they weren't here in the early says of this work. The 4
reference plant. And we were involved in discussions 5
with you about how you would select that plant.
6 I think some caveats for the Committee 7
would be helpful. I know we can't identify the 8
reference plant at this point any longer.
9 You and I had had quite a bit of 10 discussion with them. I'm wondering if you and your 11 reference plant PRA people ever came to a meeting of 12 the minds. I know there were some areas where there 13 was real dispute about things you were including that 14 they didn't include. If you can talk about that a 15 little bit I think that would be useful to all of us.
16 And the last thing is, one of your goals 17 early on was training the Staff at NRC on performing 18 PRAs so I'm a little, unhappiness is probably not the 19 right word, when you say many of your key people are 20 leaving or have left and the knowledge is gone, I sure 21 hope they did enough training that maybe the knowledge 22 isn't gone. You have others who can follow on with 23 that. So if you can talk about those two I'll get out 24 of your hair for a while.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 MR. KURITZKY: Oh no. and I thank you, 1
Dr. Bley. Please stay nearby and I always welcome 2
your comments.
3 So I'll hit one and then the other and 4
hopefully I'll remember the second one while I'm 5
talking about the first one. So let me jump first to 6
the knowledge transfer one. And that is a, that is a 7
very good point.
8 One of the main focus of this work has 9
been to bring up the state of practice. The 10 capabilities of staff in PRA, particularly as we 11 become a more risk informed regulator we want to have 12 ample PRA capabilities on staff to process and 13 evaluate risk-informed applications and other risk-14 informed issues. Or policies.
15 We have a new project that lasts a long 16 time. I mean, people are constantly cycling. That's 17 the nature of the beast. Whether they're getting at 18 the end of their career and retiring, whether they 19 find better opportunities in other agencies or outside 20 the government or just move on in order to get, in 21 reality, in order to get promotions in much of the 22 agency or any other agency you have to move to a 23 different organization within the agency in order to 24 get a promotion so there is going to be constant flux 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 in turn. That's the reality of it.
1 But yes, we have tried to focus on making 2
sure that when people go that there are other people 3
to fill in. That's, in the macro, that was the whole 4
idea. Was as more of the senior PRA people retired 5
over the years that there would be this new crop of 6
people to take over. And that's really happening in 7
the climate with microcause and with people leaving 8
the project over the last few years.
9 So we have had people that have come in.
10 In some cases the change has been fairly abrupt and 11 we've had, lucky to people that are still in the 12 agency, in different organizations. We've been very 13 fortunate that those organizations and those people 14 have agreed to continue to support the project, that's 15 helped us a lot.
16 And then therefore that knowledge does 17 stay here in the agency. But there are other cases 18 where we have had to bring in new people and bring 19 them up to speed. Or at least get them as familiar as 20 possible with this stuff.
21 My point was, the person who actually did 22 the study, did the work, did the PRA model with all 23 the thought thinking and documenting assumptions and 24 everything else, they're going to know the stuff a lot 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 better than someone who comes along and is told about 1
or taught about or plays around with it a little bit.
2 So there is just natural leakage of knowledge when you 3
go from someone who did the work to someone who now is 4
familiar with the work or has been trained in that 5
area.
6 Also, we've unfortunately, no secret, the 7
agency has struggled to stay properly staff. We in 8
PRA-B are very understaffed. And so because of that 9
it's just a body count limitation in terms of people 10 to be able to pick up and take over certain tasks.
11 But that's certainly a priority.
12 We are trying to bring in people to fill 13 in the rolls. We have brought in people just recently 14 that are taking over more and more of the rolls and 15 picking up information from more experienced people in 16 the project and in the field of PRA.
17 And so, that's something we're always 18 working at. And just because some of the people that 19 did the actual initial studies, or models, have left, 20 that doesn't leave us totally empty handed. We have, 21 you know, success in many cases. We've had people 22 that have just moved in seamlessly and taken over when 23 one person has left.
24 And so I don't want to sound like 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 everybody has jumped ship and now we don't know 1
anything that's happening on the project, I'm just 2
saying for some of the details, even for people that 3
are still here, I mean, they did the work five, six 4
years ago, they're not going to remember every little 5
detail, but there are some people that have left the 6
agency and so it's just, it's a little more difficult 7
to create every single thought process once people 8
have moved on and new people have inherited the work.
9 But the other item? I knew I'd forget 10 that.
11 MR. BLEY: Reference plants.
12 MR. KURITZKY: Oh yes. Thank you very 13 much. So the reference plant. Yes, so, you know, I 14 think I know what you're referring back to. We did 15 have, early on, some disagreements and some modeling.
16 This goes back to the Level 1 internal event PRA. It 17 really was the only case where we ran into technical 18 disagreements with how things are modeled in the 19 reference plant PRA versus our own.
20 And those cases we just, we work with them 21 to see if they could justify to our satisfaction what 22 they did. We were never quit comfortable and so we 23 relied on our own modeling and approach. And that's 24 what the results show. So if the reference plant PRA 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 was compared to our PRA results you would see that our 1
results differ, to some extent, because of the 2
difference in how we treat those issues.
3 Primarily, I don't want to go into the 4
technical details, but it primarily dealt with how we 5
were modeling station blackout sequences. How we were 6
modeling recovery of AC power. How we were modeling 7
failures of certain emergency AC power components. As 8
well as certain, quantification of certain human 9
reliability actions. Not human, operator actions.
10 So that led to a difference in station 11 blackout related core damage frequency, which is the 12 main driver at this plant. Particularly in our model.
13 And so you see some differences there.
14 But that was really, I think, the only 15 area that we had any significant technical 16 disagreement. Is that what you were referring to, Dr.
17 Bley?
18 MR. BLEY: Yes, it was. And thanks for 19 that answer. The other thing was, since many of the 20 members weren't around at that time I guess the only 21 thing is we cannot talk about who the reference plant 22 is, so it's just a note to everybody.
23 MR. KURITZKY: Yes. And I appreciate that 24 caution, yes, for those who don't realize that yes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 It's, we don't mention that now. Because in reality 1
this model doesn't actually reflect an exact plant 2
anymore anyway.
3 I'll get into this a little bit later, but 4
this was based on the design operation of the 5
reference plant in 2012, which is far different than 6
in 2023. And in addition, we made other modeling 7
decisions and other used various modeling technics 8
from our SPAR models, et cetera, that would lead us to 9
have different results that what the reference plant 10 does.
11 So it doesn't really reflect any existing 12 nuclear power plant, it's based heavily on a reference 13 plant and site. But it doesn't really reflect any 14 existing plant so it's legitimate to just refer to it 15 as a reference plant.
16 Okay. Other questions?
17 CHAIR DIMITRIJEVIC: Now that I think of 18 it, this has taken enough time now. I just want to 19 mention that this is a Subcommittee meeting and all 20 which we discuss here with you is our personal 21 opinions and, you know, insights as we see them now.
22 And our official suggestions will be provided in the 23 latter, so, you know, all these discussion is based on 24 our personal views of the things.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 So, all right. I mean, you hear there a 1
lot of comments. And, I mean, I had the, I mean, 2
Dennis and Joy expressed a lot of what I wanted to say 3
about these things, but in this moment we are not 4
going to, at least in my opinion, we are not here to 5
comment on the technical aspects of your work because 6
this work is done and documented in so many volumes, 7
so not any changes reasonable or should be discussed.
8 But we are sort of looking on how will all 9
of these things be presented and documented. And, you 10 know, and this is where most of our suggestions are.
11 And when I was reading all of these 12 volumes I felt like, you know, I was in the, like a 13 diamond mine. I would occasionally run into something 14 which made me think, oh wow, this is really 15 interesting, you know. But the thing is, there is so 16 much material and those things are spread and sprinkle 17 through the report, you know. So in the, what would 18 you subtract to the summary report, which I guess you 19 will now talk about structure, will be very important 20 from the, you know, point of the project point of 21 view. Okay, please continue.
22 MR. KURITZKY: Yes, I was thinking. Oh, 23 okay. And thank you all for that. I appreciate that 24 because that's exactly what we're looking for now.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 You're right, the technical work has long 1
since been done for the most part. We'll mention a 2
few things shortly that are still being worked on.
3 Much of this work has already been completed. So the 4
work here on the Volume 4 models is done and so, yes, 5
the messaging and the presentation information is 6
really what we're looking for feedback on, so thank 7
you.
8 Okay, so moving forward. I want to now 9
just go over the project status. Oh, I'm sorry, one 10 thing that is my mind, Doctor, because I got 11 distracted when we were talking about the 12 acknowledgments section.
13 So one thing that I did want to say and 14 mention is, again, I am the program manager, I'm 15 presenting this information today, but I'm not the one 16 who did this work. As you just mentioned, there was 17 a ton of work. And there's a ton of great technical 18 work on the, work we're doing today as well as the 19 project as a whole.
20 I've been very fortunate that we've had a 21 tremendous number of really topnotch technical experts 22 working on this, both within the NRC and also with 23 National Labs and commercial contractors. And I've 24 been very fortunate to be able to work with these 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 people and get their support.
1 And there are way too many names for me to 2
individually list. I do suggest that if people get a 3
chance, go look at those individual technical volumes, 4
look at the title pages for the authors, look into the 5
acknowledgment pages for other people that have 6
supported the work and you can see all the people in 7
there that have really just done some tremendous work.
8 And some of them have left the agency, but 9
many of them are still here. And that gives me a lot 10 of confidence that we will have a lot of PRA 11 capability going forward.
12 So with that, let me just mention for a 13 few minutes where we stand in terms of the project 14 status. I want to focus, figure up, it's a figure 15 that probably many of you have seen before. It's in 16 most of my presentations. But I just want to, just 17 to, I'm not going to go in detail here, but just to 18 kind of refresh people's ideas in terms of the work 19 being done.
20 All the models and internal reports are 21 done in two phases. Phase 1 is where the bulk of the 22 work is done. And that's the initial model and the 23 initial report, internal report. And then Phase 2 is 24 where we finalize the model and update the reports 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 based on some review feedback from external to the 1
project reviews.
2 And so that's, there is Phase 1 and there 3
is Phase 2. Phase 1 being the much bigger phase. And 4
I just want you to have that in mind as we go to the 5
next slide.
6 CHAIR DIMITRIJEVIC: So, Alan, before you 7
leave --
8 MR. KURITZKY: Yes.
9 CHAIR DIMITRIJEVIC: -- there is a color 10 code here, right? Is there some color code of what is 11 green, what is orange and, or is that for artistic 12 purposes?
13 MR. KURITZKY: Yes. The only color thing 14 here is green, is actually the work on the models.
15 And really models, well, documentation is kind of like 16 a download, but that's models and documentation. And 17 the orange is the reviews. So the green is really the 18 work that the project team is doing, and the orange is 19 the various reviews.
20 CHAIR DIMITRIJEVIC: So remind me again, 21 where you are with this PWR peer review? Is that the 22 finished --
23 MR. KURITZKY: The PWR Owner's Group led 24 peer review?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 CHAIR DIMITRIJEVIC: Yes.
1 MR. KURITZKY: Okay. So yes, that was 2
something that early in the project we were very 3
fortunate that PWR Owner's Group was willing to 4
support us on doing some standard based peer reviews, 5
just like they do for licensee PRAs.
6 So they did that for us for the Level 1 7
PRA for internal events, and the Level 1 PRA for 8
internal floods. And then also for the Level 2 PRAs, 9
Level 2 and Level 3 PRAs, provincial events and 10 floods.
11 And they are, I think they also, they 12 supported a, it wasn't a peer review, but they came up 13 with criteria for us to evaluate the dry cask storage 14 PRA because there was no standard for dry cask 15 storage, they, we had kind of like an expert meeting 16 that the PWR Owner's Group was in charge of where we 17 came up with criteria for which to evaluate the dry 18 cask storage PRA.
19 And I think we may also have had some 20 support for them and some of the other hazards work.
21 There was, or -- I can't remember. For some reason 22 there was another, I think there was another thing 23 they supported us on but I can't remember offhand.
24 But anyway, it was primarily the initial 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 PRAs that they did. Unfortunately budget constraints 1
and other issues led to that, eventually stopping that 2
work. But the initial models we were very fortunate 3
to have. Those owners who led peer reviews.
4 Both because we got a lot of good feedback 5
from them on those reviews. Also, they allowed NRC 6
Staff to both be on the peer review panel as well as 7
observe, which normally people don't do.
8 But they allowed us to observe some of the 9
peer review proceedings and so we both got a very good 10 understanding of what that process is, which is 11 beneficial. As well getting the input on our specific 12 PRA models, which sort of helped us in two different 13 ways.
14 CHAIR DIMITRIJEVIC: And --
15 MR. KURITZKY: And it was also beneficial 16 for them because some of those were trial used 17 standards and so they got a lot of use from us as we 18 allowed them to pile up those standards. Like for 19 Level 2 and Level 3. Sorry, go ahead.
20 CHAIR DIMITRIJEVIC: Are you planning to 21 document some of those results because you didn't in 22 Volumes 2 and the, you know, the 4th. I mean 3 and 4.
23 You didn't document, I didn't see any, the commutation 24 on results of those peer reviews.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 MR. KURITZKY: Yes, that's all internal 1
information. All those, the studies they were 2
reviewing were official use only. They had a lot of 3
proprietary information. And the reviews themselves 4
from the PWR Owner's Group are identified as, what do 5
they call it, confident, I'm drawing a blank on the 6
word. Not, you know, proprietary. They were 7
proprietary.
8 CHAIR DIMITRIJEVIC: Oh, proprietary.
9 Okay.
10 MR. KURITZKY: So we have them internally 11 but we don't, they have not been --
12 CHAIR DIMITRIJEVIC: You can't share the 13 results. Okay.
14 MR. KURITZKY: Right.
15 CHAIR DIMITRIJEVIC: Okay.
16 MR. KURITZKY: Right. Okay. So that's 17 Phase 1 and Phase 2, just to keep in mind as we go to 18 the project status dashboard. This is how I kind of 19 keep track of where things stand.
20 On this dashboard, which you guys have 21 probably see at some point before, but along the top 22 you see the various PRA levels, one, two and three, as 23 well as the last column, 2020, as we work to support 24 the overview reports. Which because the base model, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 the model is based on the plant, the reference plant 1
as designed and operated back at the project cutoff 2
date, which was in August of 2012.
3 So there's a lot of features that are 4
currently in the plant that were not reflected so we 5
went and did an updated. Essentially a sensitivity 6
study where we incorporate some of these more recent 7
features.
8 And those are documented in the overview 9
report, and so therefore there was a lot of internal 10 calculation notes, Level 1, 2 and 3 PRA calc files to 11 support that. So that was something that was added to 12 the scope of the project afterwards, and that's in 13 that last column. And we'll discuss more about that 14 in a few slides.
15 So those are the four columns on top. The 16 rows going down are the different rheological sources 17 on the site, as well as different plant operational 18 states and the different hazard groups.
19 So if you kind of step back and look at 20 the whole picture, we pretty much completed most of 21 the technical work. The only areas that really have 22 technical work left, you know, the phase one work is 23 just in, there is yellow boxes.
24 So for the low power shutdown there is the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 2020 FLEX sensitivity case which we've completed for 1
the Level 1 shutdown PRA, but we still have to 2
complete for the Level 2 and Level 3 shutdown PRAs.
3 Also, for the spent fuel pool, the Level 3 work for 4
the spent fuel pool still has to be completed. And we 5
were doing active work on the integrated site risk 6
task.
7 So outside of those three there is not 8
that much more that has to be done. You'll see a few 9
of these boxes in blue. Those, for the shutdown PRA.
10 Spent fuel and dry cask storage. There is some 11 additional tie-up documentation items and reviews that 12 have to be completed, but the technical work is 13 essentially done.
14 MR. BLEY: Alan?
15 MR. KURITZKY: Yes.
16 MR. BLEY: It's Dennis again. So as I 17 understand you, and you'll talk about this more at the 18 end I think, the three that are Phase 1 will 19 eventually become Phase 2 when they complete?
20 MR. KURITZKY: Yes.
21 MR. BLEY: Okay.
22 MR. KURITZKY: God willing, yes.
23 (Laughter.)
24 MR. BLEY: And you'll talk about the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 schedule for that before you're done today?
1 MR. KURITZKY: Yes. I'm going to talk 2
about the schedule because, so just to be clear, this 3
is all, the technical work and internal reports have 4
to be done.
5 And then, the one thing that remains, and 6
thank you for reminding me, the only thing that 7
remains after all this is converting them into public 8
reports which really is just scrubbing out proprietary 9
information. That's really the bulk of it. And 10 formatting for, editing and formatting for NUREG 11 publication. So that's the piece that will come out.
12 And that's what happens before they come 13 out to the public and before we come back to brief the 14 ACRS on those. And so I'll go over the schedule for 15 those being released publicly.
16 MR. BLEY: Well I'm pleased to hear that.
17 I heard rumblings that you kind of given up on anybody 18 that cited risk, and we were very interested in that.
19 So we look forward to when you wrap that up.
20 MR. KURITZKY: Yes. And just to be clear, 21 because I'm not discussing integrated at-risk at this 22 meeting, but just for your information, I don't know 23 where the grumblings came from, but we have recognized 24 that we are not going to a full quantification 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 integrated site risk for this study. What we'll end 1
up doing is, more of a proof of concept.
2 We have some Level 1 multi-unit core 3
damage frequency results for this referenced site.
4 But for Level 2 and 3 multi-unit core results as well 5
as integrated site risk. We bring in the spent fuel 6
and dry cask storage. Those will mostly be just some 7
pilot applications or some discussions because we 8
recognize that the level of effort required to do that 9
work completely unfortunately we just don't have the 10 time or resources left to do it.
11 Just to kind of give you the concept, when 12 you go look at multi-unit core damage frequency in the 13 Level 1 space, if core damage results for one unit, 14 you can have core damage results for the second unit 15 and you can combine them together for a core damage 16 result that involve both units. But again, you have 17 one metric, it's core damage.
18 When you go to the Level 2 space, and we 19 have 16 release categories. So when you have a 20 release category, and it's not even the fact that if 21 you have at least Category 1 for Unit 1 that you'll 22 have released Category 1 for Unit 2, and so you have 23 16 now instead of one metric, you actually could have 24 cross combinations. So you could have really 16 times 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 16, minus some duplications, but you end up having, I 1
can't
- remember, but a
hundred plus different 2
combinations. So it really wasn't practical for us to 3
do all that here. But we do it for --
4 The good thing is having the Level 3 5
results for the single unit we know which are the more 6
important release categories, and so we are doing a, 7
kind of a pilot application, looking at some of the 8
more important release category combinations. Which 9
should actually give us a fair idea of what type of 10 Level 2, and then possibly Level 3 risk, multi-risk is 11 involved. But we will not be doing the complete soup 12 to nuts.
13 So that may be what they're grumbling.
14 We're not doing a full qualification but we are going 15 to have a lot to say about integrated service.
16 MR. BLEY: Okay. And we look forward to 17 seeing that. I think it's going to be important for 18 other plants in the future. And anything that we've 19 learned that can be passed on will be very helpful.
20 MR. KURITZKY: Yes, thank you. Okay, I 21 think that's it's for project status. Let me now move 22 on, as I was mentioning, the public reports because 23 that's, the public facing information.
24 And so, what we intend to do is produce 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 the work in eight different volumes, including, or 1
consisting of, probably more than 20 individual 2
reports, as you can see on this diagram. This summary 3
report. Let me see if my mouse will work. Summary 4
report here Volume 1.
5 And that's the one I was mentioning 6
before, is going to capture the overall results and 7
the insights. It's going to look into things like 8
perspectives, comparisons, maybe other studies.
9 Recommendations for future work. Identifying what are 10 the areas that, you know, drive the risk or areas that 11 we're uncertain about. So that all will show up in 12 the summary volume, which is I believe the final 13 volume that we produce.
14 Volume 2 is the background volume that --
15 CHAIR DIMITRIJEVIC: Alan, sorry to 16 interrupt you.
17 MR. KURITZKY: Yes.
18 CHAIR DIMITRIJEVIC: You know, you always 19 talk about the four main goals on this project, you 20 know. And the last one there is to, you know, to the 21 realistic cost of developing levels to be PRA. Will 22 that also be part of the summary Volume 1?
23 You know, because, you know, this is your 24 Number 4 goal. First is, you know, develop these new 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 methods, get new insights, train the staff, you know.
1 And then the fourth one is sort of summarize the 2
extent of this effort. So, are you planning to 3
include that in the Volume 1?
4 MR. KURITZKY: Yes. Volume 1 will address 5
all the four main objectives of the project. But just 6
to be clear, that last objective will be the one that 7
probably gets the least amount of space in the report 8
because it was something that initially, I think there 9
was some thinking amongst the Commission at the about 10 whether or not we have Level 3 PRAs performed by other 11 plants and so they wanted to understand what the level 12 of effort and cost would be.
13 I don't think there is really that much of 14 a, you know, with the current Commission I don't think 15 there is that much of interest in that. Pursuing 16 that.
17 And also, the one thing that we realize 18 too is that the cost of doing this is very project 19 specific because it depends on a lot of key factors.
20 It depends on how many of the, what's the extent of 21 the PRA models that the plant already has? Do they 22 already have a very well developed Level 1 PRA, do 23 they have a Level 2 PRA? Have they looked at, you 24 know, what type of hazards are we currently, are we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 looking just at internal events and internal floods, 1
are we also looking at fire and seismic wind? Are we 2
looking at shutdown, are we looking at just plant, you 3
know, at full power? And so there are a lot of 4
factors involved.
5 Are the people that did those earlier 6
models, are they available to the organization doing 7
the study or do they have to have other people try and 8
understand what's done before them and come up to 9
speed and further that work? How much access do they 10 have to the plant and to get plant information? There 11 is just a lot of factors.
12 Is the team that's going to be working on 13 it, are they dedicated to just getting that done? Are 14 they experienced people? There is a lot of factors 15 that go into it.
16 And the way we approached it here with NRC 17 is very different than the way, versus a private 18 organization would go about it. One of the things 19 that many of you have heard me discuss it over the 20 years is the fact that we don't have a dedicated team 21 for this project and so we get people's time when it's 22 available. There is other higher priority work we'd 23 have to work on and so they are constantly cycled in 24 and out of this project.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 Because the project is one that a lot of 1
interconnected steps where things build on each other 2
so that if someone, at some point, is pulled off and 3
doesn't get their stuff done then the person who 4
needed their input can no longer do their work. And 5
that person comes back, they may finish their work 6
after some runup speed again, but now the other person 7
is no longer available to do it.
8 And so, we also, again, as you mentioned 9
before, one of the things we wanted to bring up and 10 train lots of people, we used a lot of junior and mid-11 career staff. We didn't have a team of just experts 12 working on this.
13 And whereas a private organization might 14 higher a consulting company or higher their own staff 15 that are just focused, better experienced PRA people.
16 So there are so many variables that our experience, 17 particularly with this project, wouldn't really 18 extrapolate to other organizations.
19 So we'll probably talk quantitatively 20 about some of the experiences we had with this 21 project, but there is not going to be a lot of our 22 number type information that we can say, hey, this is 23 what it takes to do this type of work and if you want 24 to do it expect this is what you would have to put 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 forward to doing it. So yes, it's something that will 1
get addressed but probably not in --
2 CHAIR DIMITRIJEVIC: Not for the --
3 (Simultaneously speaking.)
4 MR. KURITZKY: -- detail.
5 CHAIR DIMITRIJEVIC: All right, thanks.
6 MR. KURITZKY: Yes. Okay, so again, 7
Volume 2. The background volume. So that kind of 8
just describes the background of the project. It also 9
describes the reference plant insight. And includes 10 at a high level the technical approach for the various 11 aspects of the study. The overall study.
12 And volume, then the rest of the volumes 13 hit more of the detailed technical information.
14 Volume 3 is the reactor at power for internal events 15 and internal floods. And we have separate Level 1 16 reports for internal events and internal floods for 17 Level 2 and 3, they're combined together. And then we 18 have also the overview report that came later.
19 Same idea for internal fires and external 20 events for Volume 4. We have separate Level 1 reports 21 for fire, seismic and a combined high wind and other 22 hazards report.
23 And then for Level 2 and 3 it's all 24 combined into a single report. And again, an overview 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 report, low power shutdown, separate Level 1, 2 and 3 1
reports, we just look at internal events.
2 By the way, that goes back to a question 3
that I, I can't remember if it was Dr. Bley, but, you 4
know, what stuff did we, what do we no longer look at.
5 And individual items, too difficult for me to say 6
offhand, there is a million things that we did or 7
didn't do. We either didn't look at or did a separate 8
occasion for all the different parts of the study.
9 But one thing that was major on a scope 10 level is we decided not to, the reason we're going to 11 do shutdown for all hazards also. And we realize, 12 Don, that we just didn't have the time or resource to 13 do everything in the original scope so we had to 14 decide what to cut back on.
15 And it was decided that we would not 16 pursue the low power shutdown for the other hazard 17 just because that was one that we didn't really, that 18 was one of the areas that we really didn't have a lot 19 of experience with. There was a lot of open issues 20 that would have to be addressed in order to move 21 forward with that work, so if we did it, it wouldn't 22 have been that complete of a job and so it was felt 23 that the ACRS, actually the Subcommittee I think 24 agreed with us that if we couldn't really do it, a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 really good job, I thought that was probably something 1
we best to leave off.
2 So now the low power shutdown just looks 3
at internal events. So that's what will be in Volume 4
5.
5 Volume 6 is the spent fuel pool. We have 6
a combined Level 1, Level 2 PRA, which will be in one 7
report. And then the Level 3 in another report.
8 We may in fact not actually have an 9
overview report for spent fuel pool because we're 10 probably not going to have a FLEX sensitivity case for 11 that. But our dry cask storage will be in Volume 7.
12 Everything together. Level 1, 2 and 3 PRA and all 13 hazards, and then integrates that risk that's in 14 Volume 8. So that's all the reports we'll be 15 producing publicly.
16 So far, the Level 3 reports, they went out 17 for public comment back in April of '22. We've since 18 gotten the public comments back, updated those 19 reports. And they have been submitted to the Office 20 of Administration for final publication as NUREGs.
21 Volume 2 was also something that we set 22 out for public comment in April of '22, but we have 23 not submitted that one back for final publication 24 because we realize that it references all the other 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 reports and so we can't really put the bow on that one 1
until we know the reference information from the other 2
reports. So those are actually going to get published 3
at the end of the project when all the other reports 4
are going out too.
5 And then Volume 4, which is the subject of 6
today's presentation, those reports went out for 7
public comment back on August 18th. I think the 8
public comment period ends tomorrow, so I'm interested 9
to see what comments we receive on those reports.
10 And then the remainder of the reports I 11 have on this list right here. You can see if you look 12 down, for the third bullet and further down, these are 13 the ones that have yet to be completed. And they have 14 various dates in calendar date year 2024. So they'll 15 be released at various times in 2024.
16 Those dates are in gray to indicate that 17 there is great uncertainty with the dates for anything 18 that's more than a few months away, particularly 19 because, as I mentioned before, we don't have a 20 dedicated team so it really depends on how much time 21 the various technical leads can spend on this work 22 will determine how fast those reports can get out.
23 But again, we're targeting to have them all done, for 24 our calendar year, 2024.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 Okay, so that's the overview of where we 1
stand on the technical work and the public reports.
2 I now want to spend the bulk of the remainder of the 3
presentation, let me just do a time check, 9:30. So 4
we're going to discuss Volume 4. So particularly the 5
overview report for Volume 4, which is the reactor at-6 power results for fire, seismic and wind.
7 MR. EVANS: Yes, Alan?
8 MR. KURITZKY: Yes.
9 MR. EVANS: Before you get started it 10 looks like we have a question from --
11 MR. KURITZKY: Okay, sure.
12 MR. EVANS: -- Dr. Rempe.
13 MEMBER REMPE: It's Joy.
14 MR. KURITZKY: Yes.
15 MEMBER REMPE: And I'm a little late in 16 raising my hand, I apologize. But could you talk a 17 little bit about the public comments you did receive 18 on Volume 3, just at a high level?
19 Were they in the weeds about you, you 20 didn't have a component accurately modeled or were 21 they high level concerns?
22 MR. KURITZKY: So thank you, Dr. Rempe, 23 that was something I was thinking of mentioning and I 24 appreciate you bringing that up.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 So yes, the comments that we received on 1
Volume 3 were more high level, they really weren't, we 2
had a few technical questions from the PWR Owner's 3
Group, but mostly they were higher level. They were 4
essentially like, well, first let me mention that we 5
got comments only from three stakeholders.
6 Which would be an individual stakeholder 7
that was more concerned about the consequences and the 8
focus, our focus on just radiological consequences 9
when Fukushima and other accidents show that the 10 biggest health concerns are associated with the 11 evacuation, more than the radiological exposure. And 12 so we were able to respond to that one because our 13 Level 3 reports, not only do we calculate risk 14 measures or consequence measures for health effects, 15 but we also look at things like affected population.
16 Which is kind of a surrogate for the impact on people 17 having to evacuate. And we also lack contamination 18 and economic costs, et cetera. So that was one 19 comment.
20 But the other two organizations that 21 submitted comments were NEI and the PWR Owner's Group.
22 NEI comments were more of the line of, hey, we agree 23 with the results you came out with, your Level 3 24 results for the internal event and floods.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 And basically they felt that, hey, you've 1
done enough, there is no real need to keep doing work, 2
we don't agree, we're completing the project. I think 3
that was more of, in a nutshell, in a higher level.
4 The PWR Owner's Group also had some higher 5
level comments and some, as I mentioned, some more 6
technical comments. But their basic thing was, hey, 7
we see the margins that you're showing in the Level 3 8
space to the QHOs, has there been some thought about, 9
you know, how that would roll back into regulatory 10 guidance? And that's something that is not something 11 we're addressing as part of this project.
12 So when we, when the reports finally do 13 get published, when Volume 3 gets published as final 14 NUREGs, like I said, they're with our publications 15 department in the Office of the Administration right 16 now, when they come out there is also going to be a 17 file in the package that has a table with all the 18 comments and our proposed responses to them. So 19 you'll see some of our responses there.
20 But basically, any impact that the results 21 of this study has on regulatory applications, it will 22 be done, not by the project team, but will be done by 23 the cognizant organizations and the agency, so that's 24 basically what, in a nutshell, what the comments were.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 MEMBER REMPE: Great, thank you.
1 MR. KURITZKY: Sure. Okay. So in terms 2
of the Volume 4 results, I do want to mention up-front 3
that, well, as I mentioned earlier, the base case 4
model for the study, which we refer to as the Circa-5 2012 case, is based on the design and operation of the 6
reference plant back in, at the cutoff PRA study 7
cutoff date of August 2012. And so there is a lot of 8
more recent changes to the plant and other aspects 9
that have not included in this study.
10 So we decided to do a, essentially a 11 glorified sensitivity case. We call it the 2020 FLEX 12 case, in where we incorporate certain things that have 13 changed since that time.
14 And specific are the items you see right 15 here on this slide. We now incorporate the passive 16 shutdown seals for the reactor coolant pumps. These 17 are Westinghouse reactor coolant pumps. And we also 18 include the FLEX strategies. Particularly for dealing 19 with an extended loss of AC power or relap.
20 And then if FLEX were to not be 21 successful, we also credit the continued operation of 22 the turbine-drive aux feed for secondary side cooling 23 given that you lose all installed AC and DC power 24 essentially referred to as blind feeding of the steam 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 generators. So those are the things that we include 1
in this 2020 FLEX case.
2 CHAIR DIMITRIJEVIC: Alan --
3 MR. EVANS: Hey, Alan, before you move on 4
it looks like we have a, yes.
5 CHAIR DIMITRIJEVIC: Yes. From Tom 6
Roberts. Go ahead please.
7 MEMBER ROBERTS: Yes. Alan, if you go 8
back to the slide you just had up? The last bullet 9
gives me the chance to ask the question I wanted to 10 ask about operator actions. The credit for a 11 continued turbine-drive aux feed pump operation seems 12 to credit the ability of the operators to continue 13 operating that system after the releases are modeled 14 in Level 2 and Level 3, is that right?
15 MR. KURITZKY: So thank you for that 16 question. You for me to go into a little bit of 17 technical detail here. And also expose some laundry, 18 dirty laundry. I don't know.
19 So here mentioned, in the Level 1 PRA we 20 did not, in the base case Level 1 PRA we did not 21 create continued turbine-drive aux feed in the absence 22 of installed AC and DC power. And that was for 23 several reasons. We felt that it was just too 24 uncertain about success.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 When you have those conditions there is a 1
possibility to over or under feed the steam 2
generators. You can end up tripping off the turbine 3
generator from getting water in the steam lines. And 4
given the conditions we felt that it was not 5
appropriate to give credit for that and so we didn't 6
in the Level 1 base case model.
7 However, in the Level 2 base case model we 8
did credit that. The Level 2 team did create that for 9
its effect on acts and timing. They did give a very 10 high failure probability. I think the failure 11 probability was something around.6, so it wouldn't 12 have made that much difference on the results whether 13 they credited you or not, but they do have that in 14 there.
15 So in fact, when we go in and credit here 16 for this 2020 FLEX case, we had to strip out its use 17 in the original Level 2 model and replace it with this 18 new FLEX model, which actually credits it for Level 1 19 and then therefore propagates the Level 2 and Level 3.
20 So I don't know that I directly addressed 21 your question. Do you want to just repeat this to 22 make sure I --
23 MEMBER ROBERTS: Yes, right. I have a 24 more general question that this just offers a chance 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 to ask which is the role of operator action in the 1
Level 2 and 3 progression after the release becomes 2
significant to the environment.
3 And we had a meeting yesterday on the high 4
burn-up fuel rulemaking where there was some 5
discussion on control room dose. And there was a 6
footnote in that report, I'm not sure if you're 7
familiar with the footnote that says that, basically 8
risk analyses don't care about the operator dose, that 9
they don't model the change in errors of a commission 10 or omission caused by, you know, whatever the dose 11 rate environment might be.
12 Yes, I was wondering, a, if that's 13 accurate, and then b, just your general thoughts on 14 the role of operator action in the control room and in 15 the old site technical support center during the 16 release phases of a Level 2 and Level 3 PRA because it 17 would see like the presence of operators in the 18 facilities could be helpful in terms of accident 19 management. And if it's not modeled in the PRA I'm 20 just wondering how you account for that?
21 MR. KURITZKY: Okay. Thank you for that 22 question. So I'll probably give you a partial 23 response and maybe I can phone a friend for more 24 details.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 So this treatment was done pre-core 1
damage. It's a Level 1 treatment so it has impacts 2
for Level 2 and 3 but the actions occur pre-core 3
damage. So they're not impacted by any radiological 4
effects in the control room.
5 We do have, as I mentioned before, the 6
areas we kind of pushed to the state of the art was we 7
do have a post-core damages HRA that we performed and 8
we do credit certain mitigation actions in the Level 9
2 space. You know, in post-core damage. Up to a 10 certain point.
11 But the, and I don't know specific, I 12 think we considered all
- aspects, including 13 habitability concerns which would directly include not 14 just temperature and steam but also possibly 15 radiological concerns. And for Level 3 I don't think 16 we considered specific operator actions in the Level 17 3 space, we considered things more broadly about just 18 assuming certain evacuation things.
19 So most of the stuff with the TSC or other 20 things could be in the Level 2 space as part of that 21 post-core damage HRA. I think we consider things like 22 that, but I don't specifics.
23 I don't know if Susan Cooper is on the 24 line and is able to answer any more on that. I don't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 like to put people on the spot because, again, Susan 1
was just one of three people that worked on that and 2
they did it many, many years ago so I don't know --
3 MEMBER PETTI: She put it in the chat, 4
Alan. So you can --
5 MR. KURITZKY: Oh, okay.
6 MS. COOPER: Yes, I can --
7 MEMBER PETTI: That doesn't help, say it 8
on the record, Susan.
9 MS. COOPER: Okay. Susan Cooper, Office 10 of Research. Yes, the Level 2 HRA approach addressed 11 many aspects with respect to environmental hazards, 12 and habitability was one of those. So that was 13 addressed.
14 We had a lot of information from the 15 larger Level 2 PRA about different areas of the plant 16 that would impact operators and equipment and its 17 availability to perform. So that was definitely part 18 of it.
19 I will say that we found out somewhat late 20 that there were some areas of the plant that we didn't 21 have radiation information. We did chase that a bit 22 at a site, plant site visit. And worst case we 23 decided that probably maybe they would do a survey 24 first so it would slow things down.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 But one of the things that we did in the 1
Level 2 HRA was say, hey, don't give us some made up 2
time we'll estimate how long it's going to take these 3
actions to be performed and then figure out where that 4
levels you so far as containment end states. So it 5
might have stretched the time some, but the general 6
answer is yes, we considered habitability, including 7
radiation, among other things.
8 MEMBER ROBERTS: Thank you very much, 9
Susan. I assume that's a report that was issued that 10 we could probably get access to?
11 MR. KURITZKY: Susan, I'll jump in for 12 that one. So there are two reports that document our 13 work on the post-core damage HRA. There was one that 14 addresses the approach and there is one that addresses 15 the implementation.
16 The approach report I think will be made 17 publicly available, but I can't remember where that 18 stands. But the implementation one will not. So 19 possibly we're re-discuss the specifics and what we 20 consider for this study will probably not be publicly 21 available because there is a lot of proprietary plant 22 specific information in it.
23 MEMBER ROBERTS: You'll get that one.
24 (Simultaneously speaking.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 CHAIR DIMITRIJEVIC: That one may include 1
action discussion. You say that those, you know, 2
actions you use the simplified method so that was the 3
one, it was one of the products you're not endorsing, 4
right, when we were discussing introduction.
5 One of the things you mentioned analyzing 6
this accident management action was something that you 7
didn't have too much guidance and the simplified 8
approach was developed. Is that a true statement?
9 You said that in the introduction when we 10 were discussing what approach, you know, what 11 approaches you don't want to endorse.
12 MR. KURITZKY: Yes. Actually, I apologize 13 because I don't think I was very clear about 14 discussing that. That was, at the beginning of that 15 bullet when I was talking that was an example of an 16 area we advanced the state of the art. And so that is 17 an area where we would like people to look.
18 We don't expect that our approach for 19 post-core damage HRA is going to be the official final 20 approach that all post-core HRAs have done for the 21 rest of PRA eternity, but we think we put a pretty 22 good beachhead down for how to do this. And hopefully 23 people as they, like other areas in PRA and everything 24 else, people will work with it, build on, improve it, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 you know, whatever and hopefully it will find, you 1
know, widespread acceptance.
2 But it something that we do want people to 3
look at and work with. So that was -- and that's the 4
example of something we did push to the state of the 5
art and we do want people to use as opposed to the 6
simplifications that we don't necessarily want people 7
to use. So no, that wasn't an example of the 8
simplification, that was an example of an area that we 9
did push the state of the art and we're very happy to 10 work with and critic it and improve it.
11 CHAIR DIMITRIJEVIC: Okay. Because in 12 this Volume 4, the actions which were discussed, you 13 know, connected in preventing corrosion events or 14 controlling containment pressure or flooding cavities, 15 they all come with the number one or a 01. And also 16 there is mention that those HRA analyses didn't 17 include equipment. And Susan just said that that 18 actually, the state of the severability of equipment 19 was also analyzed.
20 So I guess we will have to wait for your 21 report on this HRA, right?
22 MR. KURITZKY: Well, so again, I apologize 23 because maybe that wasn't as clear as it should be.
24 So the zero and the one for, like for instance, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 containment venting and preventing base mat melt 1
through, and there was another one, I think 2
controlling hydrogen, et cetera, that was probably a 3
sensitivity study that I did.
4 I'm going to get to that when we talk 5
about the Level 2 results, but just since you bring it 6
up now, so we credit, in the post-core damage HRA we 7
credit actions in the short time frame. We credit 8
actions up around the time or shortly after vessel 9
breach. Okay? And generally at most two actions per 10 sequence.
11 But we carry on this severe accident 12 analysis generally for up to seven days. And so what 13 happens, we don't credit other mitigation action in 14 that longer time frame. So what we wanted to do is 15 just see if we did credit additional actions how much 16 impact could that have on the result.
17 So that table you're referring to that has 18 the ones and the zeros, I think that just shows up in 19 Volume 3 on internal event, internal flood --
20 CHAIR DIMITRIJEVIC: Right. Right.
21 MR. KURITZKY: -- report. So that is 22 just, it's just a sensitivity case to see what 23 potential impact it would have.
24 And we show that there was some of those 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 things could actually be very beneficial. I think 1
particularly the one where we control containment 2
pressure. It was very influential. And then as it 3
was combined with the control of hydrogen combustion 4
even more so. So that was just a sensitivity study.
5 In the actual evaluation for the base case 6
model we do HRA analysis and consider the specifics of 7
the SAMGs and the severe accident management 8
guidelines and the extensive damage mitigation 9
guidelines at the reference plant to come up with 10 fairer probabilities. So that was more of a --
11 CHAIR DIMITRIJEVIC: We have a bunch of 12 questions. I think Joy was first, then Dave, then 13 Dennis, all right?
14 MR. KURITZKY: Okay.
15 MEMBER REMPE: Oh, sure. Thank you. Back 16 when the industry decided to implement FLEX there was 17 always an interesting characteristic that the 18 building, as I recall, the building or house that the 19 equipment is housed in, is not required to withstand 20 a higher seismic load or wind loads, flooding.
21 I'm not sure about what the situation was 22 at the reference plant, but how did you address this 23 in the study and did you consider, as you've talked 24 about, that you've tried to broaden it beyond to the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 reference plant. Did you do sensitivities to say, 1
okay, yes, there is the FLEX equipment but the 2
building may not be able to withstand the wind loads 3
or the seismic loads and did you look at that issue?
4 MR. KURITZKY: Thank you, Dr. Rempe. Yes.
5 So as I'll discuss, we're going to get in just a few 6
minutes here to the FLEX case, and we did not do a 7
rigorous detailed analysis of the FLEX fire 8
probability we did a parametric study. But in doing 9
so we considered the impacts of, for instance the 10 hazards, and also considered whether or not if the bio 11 was higher or lower what would be the impact on the 12 results. So we did address that topic. And I'll go 13 into more specifics in a couple slides.
14 MEMBER REMPE: Great, thanks.
15 MR. KURITZKY: Yes.
16 MR. EVANS: Next?
17 CHAIR DIMITRIJEVIC: Dave?
18 MEMBER PETTI: I think you're going to get 19 there. I was just interested in some of the FLEX 20 results and the human reliability aspect of it and 21 whether or not you accounted for if they failed the 22 first time that they try again and again, because 23 that's what operators will do, if you just assumed 24 they failed and that, you know, you only get one 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 chance to, you know, to implement it. So when you get 1
there, I'd like to sort of understand that.
2 Because I think what struck me, again, not 3
being an expert, is the letter on its face, FLEX 4
doesn't seem effective. That's what I took away from 5
reading the summary report. Now I'm sure that's not 6
what we want to say, but again, I could have misread 7
it, but that's the impression I have, so.
8 MR. KURITZKY: Yes, I think we are going 9
to get to the results in the next slides and so --
10 MEMBER PETTI: Right.
11 MR. KURITZKY: -- you'll see that, no, 12 FLEX does have, the reference plant does have 13 significant impacts.
14 But just not to jump the gun, but again, 15 we did not do a detail analysis for FLEX there 16 probably so we don't have extensive detailed HRA on 17 the actions it's more of a, like I said, a parametric 18 study. But I'll go into that --
19 MEMBER PETTI: Okay.
20 MR. KURITZKY: -- probably in the next 21 slide or two.
22 MR. BLEY: Alan, it's Dennis.
23 MR. KURITZKY: Yes.
24 MR. BLEY: Your discussion with Susan 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 irked my interest. You went through the eight volumes 1
of your study earlier, but you didn't mention side 2
technical reports. And it sounds like there is at 3
least one for HRA, or at least level 2 HRA. And I'm 4
wondering if there are others? Maybe things you did 5
with success criteria or is that all included in the 6
main eight reports?
7 MR. KURITZKY: So success criteria is 8
included in the reports themselves. Particularly, for 9
instance, for the Level 1 for intro management. There 10 is a Chapter 4 which goes into detail about all the 11 MELCOR runs and work that we did. And the reference 12 plants max, not max, map runs --
13 MR. BLEY: Yes.
14 MR. KURITZKY: -- and stuff. So that's 15 all integral to the reports.
16 But the, just making a note about, so the 17 other reports, and particularly the one is the HRA 18 reports. You know, that is one that we had, we had 19 these two reports. They were internal reports.
20 The one is, doesn't really have 21 information that would preclude it from being publicly 22 released, so we had someone work to pull together some 23 of the HRA from all the different parts of the study 24 including that, into, well that was a Level 2 report, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 but even the Level 1 HRA stuff into some kind of 1
document which, unfortunately that person has since 2
left the agency, but we have it in some kind of form.
3 And it just hasn't, like folks on the other reports we 4
haven't done much with it.
5 But that's one of the decisions we have to 6
make when we get near the end of the project is, which 7
of these other reports, besides the exact volumes you 8
saw, what, if any other reports, should we put out?
9 Mostly we would like to attach these things as 10 appendices to the existing report so that it's --
11 MR. BLEY: Sure.
12 MR. KURITZKY: -- easier for people to 13 find them and then they don't have to, like hey, how 14 do I find this, I'll just get a report, you know, it's 15 in ADAMS dockets, how do I know that. But the, so we 16 have to figure out exactly what we're going to do 17 there.
18 I'm trying to think now. Yes, the Level 19 2 report, we added a number of appendices to it from 20 other supporting reports that we had internally, but 21 we did not have the HRA report as one of them. So 22 that's a good point. We may have to, that's one that 23 we might just have to put out as a separate report, 24 but that's a decision we're going to make a little bit 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 further down the road.
1 MR. EVANS: We've got a comment --
2 (Simultaneously speaking.)
3 CHAIR DIMITRIJEVIC: Susan has --
4 MR. EVANS: -- by Susan.
5 CHAIR DIMITRIJEVIC: Susan? Susan, we see 6
your hand up.
7 MS. COOPER: Okay, thank you. Thank you, 8
Vesna.
9 So, not for FLEX because was a different 10 case. And Alan is going to talk about that in a 11 minute. But since we're talking about the Level 2 12 HRA, yes, I think there is an awful lot about the 13 Level 2 HRA that would be good to get out and a lot, 14 unfortunately the specifics are always going to be 15 proprietary, but there is some good information in 16 there.
17 And we did address things, Dave said a 18 moment ago about operators trying multiple times to 19 get equipment going. I mean, that was in fact the 20 rationale, the reason behind us saying, we're going to 21 estimate how long the operators might need to get the 22 equipment, you know, transported from the warehouse to 23 near the water tanks and connect up the hoses and pull 24 the hoses where they need to go and get pumps started 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 and all that kind of stuff.
1 We recognize that that was not going to be 2
just like pushing a button in the control room because 3
we walked it down. We talked with the operators and 4
so forth. So that is part of the underlying HRA, 5
Level 2 HRA method for the Level 2 PRA. But that's 6
not what was done for FLEX, as Alan is going to 7
explain.
8 I did do a paper with my coauthors for one 9
of the PSA conferences way, way back when. I don't 10 know if it was, I don't know what year it was, but, I 11 mean, it does explain something about the method and 12 something about what we learned from the plant site 13 visits that shaped how we developed the method. But 14 until then I think that's the only thing that's out 15 there. Thanks.
16 MR. KURITZKY: And thank you for that 17 comment too because, yes, I've highlighted that needs 18 to arise further on my radar. The idea of how we're 19 going to get that post-core damage HRA report into the 20 public domain. So thank you for that feedback.
21 MR. EVANS: Alan, we have another hand up.
22 Dr. Roberts.
23 MR. KURITZKY: Okay.
24 MEMBER ROBERTS: Yes. Just to close out 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 my question. That was very helpful, Susan, in terms 1
of explaining the Level 2 HRA. I wonder if there is 2
any thought on Level 3? Whether there is some 3
influence in the ability to make better decisions on 4
protective actions based on having people available in 5
the control room with the technical support center or 6
whether that's stuff we didn't factor in at all?
7 MR. KURITZKY: Well, my, the quick answer 8
is, I don't think that is something that we 9
specifically focused on. But if Keith Compton is here 10 he's our Level 3 expert and so he can speak to it more 11 intelligently. But I don't think we got to that level 12 of detail. Keith, are you on the line?
13 MR. COMPTON: Yes, I'm on the line. This 14 is Keith Compton from the Office of Research. Can you 15 hear me?
16 MR. KURITZKY: Yes, we can hear you.
17 MR. COMPTON: Okay. I just wanted to make 18 sure. Yes. No, that's an interesting question. I 19 have to be honest, I hadn't really thought that much 20 about it. So that implies that the answer is no, we 21 didn't include it. But I'm intrigued by the thought 22 so I'll take that thought back.
23 MEMBER ROBERTS: Okay. Yes, thank you, 24 Keith. And this relates a little bit to a discussion 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 we had yesterday on the high burn-up fuel rulemaking.
1 Again, they're looking at changing the allowable 2
control room dose for the design basis cases.
3 And a question I had asked, whether that 4
has any implication on some of the work in the Level 5
2, 3 space because that's not a direct requirement 6
anywhere that, at least that I'm aware of. So again, 7
it probably is worth some thinking about so thank you.
8 MR. COMPTON: Sure.
9 MR. KURITZKY: Okay, thank you for the 10 question. And, again, Keith and Susan, thank you very 11 much for your responses. Okay, so are we clear to 12 move forward? I'm assuming no more hands up. Okay.
13 So just one thing I do want to mention 14 about the FLEX case. For those who are as familiar 15 with FLEX so, it involves a three-phrase approach.
16 Phase 1 is where the plant will initially cope to, 17 with ELAP by relying on installed plant equipment and 18 resources.
19 There may be some different strategies.
20 For instance, shedding, load shedding for the safety 21 batteries, but you're using the existing normal 22 equipment. And Phase 2 you start to rely on the FLEX 23 equipment. Which would be your backup pumps and 24 diesel generators, et cetera. And batteries.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 And then Phase 3 is when you start to, in 1
the longer term, you need to bring in additional 2
resources and equipment from offsite. For instance, 3
one of the SAFER Centers that have been established in 4
the country.
5 This analysis, our 2020 FLEX case, only 6
considers those first two phases. We don't go into 7
the long-term phase in bringing stuff in from offsite.
8 CHAIR DIMITRIJEVIC: Alan?
9 MR. KURITZKY: Yes.
10 CHAIR DIMITRIJEVIC: I just want to do 11 some clarification. Since you have this separate 12 slide on the FLEX.
13 You're 2020 FLEX case is only partially 14 connected with FLEX. To other important parts of the, 15 you know, the design change on RCP seals, and also 16 crediting these auxiliary feedwater extended 17 operation.
18 So, the changes which we see in the tables 19 when you consider the FLEX case, you know, 2012 case, 20 are combination of these three, and we don't really 21 know how much of that change can be contributed to 22 FLEX and how many do the very important change in RCP 23 seals or crediting the extended, to the auxiliary 24 feedwater operation?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 MR. KURITZKY: That's correct. And we 1
don't have specific sensitivity studies to break that 2
out, that was, because we had it limited to exactly 3
what we were going to evaluate. We thought that one 4
wasn't as essential to breakout, but you're right, it 5
includes all those three things.
6 The FLEX and the extended turbine-drive 7
aux feed are really two sides of the same coin in the 8
sense that they're doing the same function. You just 9
need one or the other. And so that's why, as I'm 10 about to describe on the next, maybe the next slide, 11 or sometime in the next couple of slides, this whole 12 idea of this Parameter P that combines the two 13 together.
14 And the actual, the other item, which was 15 the new passive shutdown seals, that one we did 16 evaluate as a sensitivity in the original study for 17 the Level 1 internal events. And we showed that it 18 reduced core damage frequency by around ten percent.
19 And what it really addressed was the 20 scenarios involving loss of nuclear service coolant 21 water. Because those, you lose all the cooling for 22 the seals. As well as the makeup systems that you 23 would need in the case of an RCP seal LOCAL.
24 So those loss of service water scenarios 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 contributed around more or less 14 percent to core 1
damage frequency for the internal events. Level 1 2
internal events.
3 And so, the use of the new RCP seals 4
reduce core damage frequency around ten percent or so.
5 Maybe a little more. So that's kind of like 6
ballparking it back then, if you want to kind of carry 7
that ten percent thing along.
8 And you'll see when we show some of the 9
results, as we go forward, you can see that the, in 10 general the reductions are more than that in the 2020 11 FLEX case because of the FLEX pieces to this. The 12 FLEX in the extended turbine-drive aux feed as opposed 13 to the seals, but they both contribute.
14 And when we get into, it's hard to 15 partially get to Level 2 and 3 results because now you 16 know and see the nuclear service clean water versus 17 the other scenarios. But much of the reduction is 18 occurring because of station blackout scenarios which 19 more is attributable to the FLEX and extended turbine-20 drive aux feed. But again, even in those scenarios 21 you do result in RCP seal LOCAs many times but in 22 those cases --
23 CHAIR DIMITRIJEVIC: Right.
24 MR. KURITZKY: -- even if you didn't have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 an RCP seal LOCA, if you have, essentially an 1
recoverable station blackout, you're going to end up 2
having core damage anyway. You may have a slightly 3
different flavor of core damage and different timing 4
of core damage but you're getting there anyway.
5 So I think the bulk of the reduction in 6
the Level 2 and 3 arena really comes from the 7
combination of the FLEX strategies and extended 8
turbine-drive aux feed. Though there is contribution 9
from the new shutdown seals too.
10 CHAIR DIMITRIJEVIC: Okay. I just want to 11 point out that it's not really clear, you know, to 12 say, okay, with introduction of the FLEX strategies 13 that's what happened to model because it's a little 14 more complex mix of the changes.
15 MR. KURITZKY: Yes. Yes, thank you. And 16 just going back to the previous slide, so that's why 17 I tried to, I tried to refer to it as the 2020 FLEX 18 case in the discussion just because of human nature to 19 shortcut things sometimes I might talk about FLEX.
20 Usually if I'm talking about FLEX I mean 21 the 2020 FLEX case so it's more than FLEX, it's these 22 other items too. But I'll try and say the 2020 FLEX 23 case to be clear. But yes, even the fact that we call 24 it the 2020 FLEX case is a little bit misleading but 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 we need some shorthand way of referring to it and so 1
that's just what we came up with.
2 Okay. So moving on to the result for the 3
2020 FLEX case. First, looking at Level 1 PRA. So if 4
you look at the table here, down at the bottom the 5
total for all hazards. For the Circa-2012, the base 6
case, the total hazard CDF is around one and a half 7
ten to the minus four. In the 2020 FLEX case it drops 8
almost 40 percent down to a little over nine ten to 9
the minus five.
10 And so again, the bulk of that reduction 11 is because of the FLEX and turbine-driven aux feed.
12 Changes go, part of it is also due to the new RCP 13 shutdown seals.
14 If you look at the different hazards 15 individually you can see that the internal events and 16 floods and the high winds have reductions of around 17 60, in the ballpark of 60 percent, while the fires, 18 internal fires and seismic events, is much lower. And 19 the reason for that is that the internal events and 20 the high winds have a large contribution from a 21 station blackout sequences. And those are the 22 sequences, the types of scenarios that FLEX has been 23 designed for so they're fairly effective there.
24 For the fire and seismic you also, there 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 are a lot of station blackout type sequences for those 1
hazards too, but we gave less credit for the FLEX 2
strategies and the turbine-drive aux feed, et cetera, 3
in those cases because of the direct impact of those 4
hazards on both the equipment itself, as well as the 5
operators who have to take, whether it's accidents in 6
the main control room or local FLEX actions.
7 So, and that goes back to, I think it was 8
Dr. Rempe's question, do you consider the fact that, 9
hey, in this case the reference plant, EDMG pump, 10 there is two, but the one that's most readily 11 available is in a warehouse which is not seismically 12 qualified. So we intentionally gave less credit for 13 the FLEX in those cases because of the more 14 significance of the hazard itself. And I'll talk a 15 little bit more about that in a couple of slides.
16 Looking at these pie charts, just to kind 17 of get a breakdown of what hazards are contributing, 18 the base case, Circa-2012 on the left, FLEX on the 19 right. So you can see, if you squint, you can see 20 that internal events and internal fires are the 21 dominate contributors for the base case. Which one 22 contributes around 40 percent.
23 And if you go over to the FLEX case and 24 you see that now internal fires has really jumped 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 ahead, and that's because, as we just mentioned, we 1
gave more credit to FLEX for the internal events so 2
you see a bigger reduction there. You don't see as 3
much of a reduction on, for the fire, so now the fire 4
has become essentially twice as important as the 5
internal events.
6 You also see something similar with the 7
high winds and seismic events. In the base case on 8
the left you can see the high winds is around nine 9
percent of total CDF and seismic is seven, but when 10 you go to the FLEX case it flips and you have seismic 11 at nine and wind around five. And again, it's because 12 the FLEX was much effective. In our assumptions it 13 was much more effective for the high wind than it was 14 for seismic.
15 One thing you don't see on these charts, 16 again, you see these four hazard categories, you don't 17 see all the other hazards. So again, we did, and I 18 think someone asked about this earlier on, we did go 19 through and evaluate and screen out either through 20 qualitative or semi-qualitative analysis all the other 21 hazards based on the criteria in the PRA standard.
22 However, there were some external hazards 23 or other hazards that we did not, we did not screen 24 per say by those criteria, but we just were not able 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 to, we did not include qualification for it either 1
because there was ongoing active research in those 2
areas, going back to a question, a comment someone had 3
before about whether there was active research in some 4
of these, active research in some areas so we were 5
not, we did not address those items because the 6
information was changing.
7 And also, things that were just beyond the 8
state of the art. For instant, space weather. So 9
things like solar flares. It was something that 10 definitely could be of concern, and no question. In 11 fact, that's actually one of the public comments we 12 just got today on these reports. But it's just 13 something that's beyond the standard. We don't have 14 any way of evaluating that right now.
15 So most the other hazards were all 16 screened out by the criteria in the standard, but 17 there were a few things that we just had to leave off 18 the table because they are beyond our capability right 19 now. Or they were subject to ongoing research.
20 Okay, moving on to --
21 MEMBER REMPE: This is Joy.
22 MR. EVANS: Alan, we have a question.
23 MR. KURITZKY: Sure.
24 MR. EVANS: Go ahead, Joy.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
82 MEMBER REMPE: I just want to make sure I 1
understand you. I did see something in the report 2
saying what you've said here about that they did 3
realize in higher seismic events that the building 4
wasn't qualified.
5 What about
- wind, was the building 6
qualified for high winds at the reference plant?
7 MR. KURITZKY: I don't remember. You 8
know, honestly we ended up having a very top notch, 9
five research associates, a very top notch wind curate 10 outfit come in and do walk-downs and evaluations for 11 us.
12 I don't know if they actually walked down 13 that building because it wasn't part of the, in this 14 case PRA, so I don't think we ever got an actual 15 evaluation of that building.
16 But, again, as I am going to mention, and 17 we can go back to, or on the next slide, we don't go 18 into a very detailed analysis of FLEX, and so this P 19 value that we pick for the different hazards -- Well, 20 actually, let me just, because I was going to this 21 anyway, so if you look at this slide, again, we did 22 not do a detailed analysis.
23 We did a
parametric study using 24 engineering judgement. We used P as the parameter of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
83 merit and we defined P down below here as the 1
probability of the FLEX failing and the probability of 2
the extended turbine-driven aux feed failing, because 3
you just need one or the other.
4 And so you see that, you know, for P for 5
internal events we had a value of 0.09, but for these 6
other hazards the values are higher, 0.50 for fire and 7
seismic and 0.25 for wind.
8 So, again, the fire and the seismic have 9
fairly high values because of what we were just 10 mentioning, that the impact, the potential impact of 11 these hazards that they'll be able to implement FLEX 12 or the turbine-driven aux feed leads us to think that 13 they, you know, would be much higher than just a sunny 14 day internal event type situation.
15 The high winds was not as extreme of 16 conditions that you would get for some seismic so we 17 picked a value that's kind of in between internal 18 event and seismic event.
19 So to go to your question, Dr. Rempe, we 20 didn't do a detailed analysis and we I don't think we 21 did an evaluation of the fragility of that maintenance 22 building to wind, but we have picked a value that 23 says, hey, there is a much better chance that you are 24 going to have problems under a wind condition than you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
84 would just for an internal event but maybe not as much 1
as if you had a seismic event.
2 As you are going to see in the next slide, 3
let me just go to the next slide, because of the fact 4
it was just a parametric uncertainty and we did not do 5
a rigorous analysis, so we wanted to see what would be 6
the impact of using other values.
7 Now in the internal event case in Volume 8
3 we actually had a more detailed look at different 9
values for P for internal events.
10 Here we looked at just a few cases, but in 11 this graph right here you can see that the three 12 points that are on it are if we give no credit at all 13 to FLEX and the turbine-driven aux feed, and then you 14 have the base case, which is really the 2020 FLEX 15 case, that means the base case of the FLEX case, of 16 the 2020 FLEX case, and then the last one is if FLEX, 17 well either FLEX or turbine-driven aux feed were 18 perfectly reliable, and so you could 100 percent rely 19 on them.
20 What you see here is that the slope of the 21 lines from the base case to the perfect case are 22 fairly flat, and so this shows you that doing a more 23 detailed rigorous analysis wouldn't really buy you a 24 lot more in terms of CDF reduction.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
85 By the same token, for the purposes of our 1
study, for our study, this really supports our 2
decision not to do a lot more work to get a more 3
refined number because it really wouldn't, it wouldn't 4
really change things for us.
5 Now that's not to say that for some other 6
applications, for our regulatory application, for 7
instance an event assessment type of application, that 8
you wouldn't want to have a much more detailed and 9
rigorous analysis.
10 In that case you might want to do 11 something like a detailed HRA and also incorporate 12 operational experience, you
- know, operational 13 equipment failure data for the FLEX equipment, to get 14 a much more accurate number.
15 So that's something that you probably 16 would want to do, but for the purposes of our study it 17 wouldn't make much difference in the insights. Again, 18 this is just a sensitivity case. This is not part of 19 our base study, so as a sensitivity analysis there was 20 no need for us to put that much more effort into it.
21 MEMBER REMPE: Okay.
22 MR. KURITZKY: But if you do -- I guess 23 more to your question, Dr. Rempe, it's the other side, 24 it's going from no credit to base case, so, yes, you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
86 do see if we did not -- If we assume that the failure 1
of probability for high wind was, you know, even 2
higher than that 0.25, you can see from that lower 3
gray line at the bottom that you do see an increase.
4 Again, this is -- No. Actually, no, this 5
is an arithmetic scale, it's not logarithmic, so, yes, 6
you see some increase but it's not substantial and you 7
wouldn't expect it to go to one in either case, so --
8 Because, again, this is a, another simplification is 9
in reality you would have a value of P, you know, a 10 failure of probability for FLEX and a failure of 11 probability for the extended turbine-driven aux feed 12 that was contingent on the exact magnitude of the 13 event.
14 You know, with seismic in our study we had 15 eight seismic bins and in wind we have multiple bins 16 for straight line winds and tornados, et cetera. So 17 you would expect for the lower bins the probability 18 would be close to what you have for internal events 19 and for the higher bins it would be getting closer to, 20 you know, the no credit line or failure of probability 21 of one.
22 But what we have done is actually picked 23 values as kind of like in the middle that is kind of 24 a hedge and then this parametric study to look at what 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
87 happens when you move it up or down.
1 If you are looking at seismic, the very 2
high seismic bins like seven and eight, which have 3
widespread destruction, we don't credit the FLEX at 4
all anyway so we wouldn't have to worry about being 5
non-conservative for those.
6 But, again, to the extent that we were 7
doing a sensitivity study just to see the general 8
impact, we were comfortable with what we have done 9
here.
10 So, again, 0.09 for internal events, 0.50 11 for fire and seismic, and 0.25 in the middle for high 12 winds. While we recognize that for everything except 13 probably the internal events these are probably 14 pessimistic values in reality.
15 We do that to be, okay, just because, as 16 I mentioned, there is a lot of uncertainty associated 17 with these types of hazards and both of their impacts 18 on equipment and the OP interactions both in the 19 controlment and locally, so we felt justified using, 20 you know, higher values for P for those hazards.
21 MEMBER PETTI: So, Alan, this is Dave.
22 MR. KURITZKY: Mm-hmm.
23 MEMBER PETTI: Again, this is where I got 24 really confused because if you would incorporate the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
88 uncertainties as you note and put some error bars on 1
these is there really that big of difference in FLEX?
2 I mean I look at the seismic case, which, 3
you know, you think about FLEX being born out of 4
Fukushima, this says it won't, it may not help in a 5
seismic event, a severe seismic event at least.
6 So that's where I am -- I just -- Yes, I 7
mean I look at these numbers and, again, I'm not a 8
risk person, but, you know, 30 percent doesn't sound 9
to me in light of the uncertainties as being 10 significant.
11 Am I wrong there? Is that really from a 12 risk perspective an important reduction? I can see 13 when you get, you know, 50, 60, 70 percent, but help 14 sort of calibrate me on what's the, you know, when is 15 something significant versus insignificant in light of 16 uncertainty.
17 MR. KURITZKY: That's a very good 18 question. We have -- So specifically for the 19 uncertainty analysis I think in the actual reports, I 20 don't know, it may not have made it to the, I don't 21 think we had it in the overview reports, they are in 22 the supporting internal calculation files we did, you 23 know, we propagate uncertainties.
24 We assigned uncertainty bound to these 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
89 values and propagated them through the model just like 1
we do for all the other basic events in the PRA model 2
to come up with a parametric uncertainty balance.
3 But, more specific to your question, yes, 4
when you're talking about 10, 20 percent, given the 5
uncertainties it is not significant, you are correct, 6
and it goes back again to what Dr. Rempe had 7
mentioned, that because the design specs or whatever, 8
requirements for the FLEX equipment, was that not that 9
it had to be in a, you know, seismically robust 10 structure.
11 So when you look at seismic events you 12 have to accept the fact that there's a good chance 13 it's not going to work. Yes, it was born out of 14 Fukushima, but as Dr. Rempe pointed out, they are not 15 required to being seismically qualified structures.
16 So, you know, there is only so much credit 17 you can give for not -- I don't know whether, if a 18 licensee were to come in with a detailed analysis, 19 again, remember, we're doing a very crude parametric 20 study, but if a licensee were to come in with a more 21 detailed analysis for some type of regulatory 22 application or some type of event assessment or 23 whatever, it would be interesting to see what kind of 24 credit they gave to it for seismic events.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
90 And it may be that, you know, some plants 1
might have it, even though it's not required they 2
might have it in a seismically qualified structure, 3
others may not, so, you know, it's going to be 4
somewhat case by case.
5 But in terms of just a general idea of 6
uncertainty and results, you are correct, a 10 to 20 7
percent difference is -- We report them because we do 8
the study, but are they significant in terms of the 9
overall uncertainties, no, probably not substantial, 10 I agree. A 60 percent change --
11 MR. BLEY: Alan, this is Dennis.
12 MR. KURITZKY: Okay. Yes?
13 MR. BLEY: You know, the assumption that 14 it sounds like is in there because the shed they keep 15 this stuff in isn't seismic that it gets wiped out and 16 you can't use the stuff, I don't know if you did it 17 but the Committee went out to visit the SAFER site out 18 in Phoenix some years ago and what we learned out 19 there was SAFER can deliver all of that equipment to 20 any site in the country in less than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, 21 substantially less in many cases, and that there are 22 agreements among the people who belong to SAFER and 23 FLEX that nearby or reasonably nearby plants would 24 share their equipment if a particular plant was 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
91 affected.
1 So I think that's a grossly simplified 2
assumption I think that if the shed fails you don't 3
have any of this FLEX equipment.
4 MR. KURITZKY: Okay. So, Dr. Bley, thank 5
you very much for that because, again, I want to -- It 6
got lost through all the discussion, so I want to 7
reiterate that our 2020 FLEX case only focuses on 8
Phase 1 and 2 of the FLEX response.
9 We specifically do not consider the 10 offsite resources, so, yes, the fact that we have 11 minimum credit for FLEX in a seismic event in this 12 study for Phase 1 and 2, for the onsite FLEX 13 equipment, is not to say that FLEX as a whole is not 14 more effective.
15 Yes, with the SAFER Centers and also 16 figuring that, you know, whatever hazard is impacting 17 the site is very unlikely to impact the SAFER Center, 18 you know, except for maybe if there happens to be a 19 plant in the near vicinity of it, but I think there is 20 only a couple of sites in the country.
21 MR. BLEY: Well and they are widely 22 separated.
23 MR. KURITZKY: Yes, and they are widely 24 separated, so there is always going to be something 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
92 available, except for the solar flare, but, anyway, I 1
am not going there.
2 So in any case, yes, thank you very much 3
for pointing that. We are talking just about the 4
impact of FLEX Phase 1 and 2 using onsite equipment, 5
not the overall benefit of FLEX when you consider 6
offsite resources also.
7 MR. BLEY: And I think you were clear 8
about that, but the discussion implies -- The reader 9
is not always aware of the significance of those 10 assumptions and limitations.
11 MR. KURITZKY: Right. Let me just make a 12 note on that to see if we can iterate that in other 13 points in the report.
14 MEMBER PETTI: Yes, thanks, Dennis. That 15 was my concern as sort of misinterpreting what you are 16 really trying to say.
17 CHAIR DIMITRIJEVIC: Okay. This could be 18
-- Because we are going to switch from Level 1 to 19 Level 2 in the next slide, this could be a good time 20 for us to take a break.
21 We are back on the schedule, so let's take 22 the 15 plus minutes break and let's get back together 23 at 10:35 and resume our meeting, all right. Thank 24 you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
93 MR. BLEY: Bless you, Vesna.
1 CHAIR DIMITRIJEVIC: Mm-hmm.
2 (Whereupon, the above-entitled matter went 3
off the record at 10:18 a.m. and resumed at 10:35 4
a.m.)
5 CHAIR DIMITRIJEVIC: Okay. It's now 10:35 6
so we will resume back our meeting. Scott Moore 7
expressed the desire to make some remarks. Scott, are 8
you there?
9 MR. MOORE: Yes, I am. Thank you, 10 Chairman Dimitrijevic. Just a brief comment to 11 everybody that's online, please do not use the chat 12 feature in Teams.
13 The chat does not get recorded in the 14 transcript and we're trying to keep, you know, a 15 running transcript for the meeting.
16 The one thing you can use chat for is if 17 you are having audio or visual or computer problems, 18 but if it's related to the content of what is being 19 presented or discussed please don't use chat. Thanks.
20 That's it, Chairman.
21 CHAIR DIMITRIJEVIC: Okay. Thank you, 22 Scott. All right, Alan, so now we can see you again.
23 We hope to see slides soon, too.
24 MR. KURITZKY: Oh, okay, I forgot. Thank 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
94 you for reminding me. Let's see.
1 CHAIR DIMITRIJEVIC: Excellent. We see 2
them now.
3 MR. KURITZKY: Okay, good. Thank you. I 4
don't know why they disappeared, but, anyway. Okay, 5
so I hope everybody had a good break. Thank you for 6
coming back. Now we will move on to the Level 2 and 7
Level 3 results.
8 First, for the Level 2 PRA you can see 9
here on this table there is a comparison between the 10 circa 2012 case and the 2020 FLEX case for three 11 different surrogate risk metrics.
12 LERF, L-E-R-F, which is Large Early 13 Release Frequency, which has been defined for this 14 study as the frequency of a large release that occurs 15 early enough that there is the expectation of the 16 possibility of early fatalities.
17 Then we have LRF, or Large Release 18 Frequency, which has been defined as the frequency of 19 any large release that occurs prior to the termination 20 of the severe accident analysis for the study.
21
- Then, lastly,
- CCFP, Conditional 22 Containment Failure Probability, which is just the 23 conditional probability if the containment fails given 24 core damage.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
95 So looking at those three results, for 1
LERF, first off, you can see that in the base case, 2
circa 2012, it's 1.9 e minus 6 per reactor critical 3
year for all hazards combined and if you recall the 4
core damage frequency for all hazards combined was 1.5 5
10 to the minus 4, so it's a little over 1 percent 6
that's LERF.
7 The reason why LERF is so low in this case 8
is the fact that the accidents that do not involve, 9
the severe accidents that do not involve containment 10 bypass occur slowly enough that evacuation would be 11 effective. So that leaves with you containment bypass 12 events which are primarily three types.
13 There is interfacing system LOCAs, loss of 14 coolant accidents, and then there is also pressure 15 induced steam generator tube ruptures which occur 16 primarily after an ATWS event and you end up having a 17 pressure induced steam generator tube rupture prior to 18 core damage and that rupture occurs at a level above 19 any water, so there is no scrubbing of the release.
20 Also, that scenario involves at least one 21 secondary side relief value, either intensity being 22 opened or it being in the stuck open position.
23 The third primary contributor to LERF is 24 temperature induced steam generator tube ruptures and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
96 these are post-core damage steam generator tube 1
ruptures.
2 All three of those cases, all three of 3
those scenarios, have very low frequencies and so 4
that's why you don't see a very large LERF 5
contribution.
6 If you look at the 2020 FLEX case you see 7
that it drops down to 1.3 minus 6, around a 30 percent 8
reduction, but the same types of scenarios are 9
contributing with the primary reduction being in the 10 category of those post-core damage thermally induced 11 steam generator tube ruptures which often result from 12 station blackout scenarios.
13 The other two categories, the FLEX and 14 turbine-driven aux feed, et cetera, had very little 15 impact on it.
16 Moving to the second, well actually the 17 second and third, the late release frequency and the 18 CCFP, both of those you see a fairly high value, 19 particularly in comparison to core damage frequency.
20 The reason there is because most, a large 21 contributor to those are station blackout sequences 22 and station blackout sequences you tend to not have 23 containment heat
- removal, so eventually those 24 sequences would tend to lead to over-pressurization of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
97 the containment.
1 So you have a fairly high conditional 2
probability of containment failure and, therefore, 3
large release for those sequences.
4 Again, looking at the FLEX case you see 5
that FLEX is effective in reducing them by nearly 40 6
percent. If we recall that's kind of the same 7
reduction you saw for CDF for all hazards and that's 8
because, again, it's the similar profile, risk 9
profile, it's primarily driven by station blackout 10 sequences.
11 CHAIR DIMITRIJEVIC: Alan?
12 MR. KURITZKY: Yes?
13 CHAIR DIMITRIJEVIC: Okay, so here is 14 where I was really, my first really big surprises came 15 looking in the results. This is an awfully high large 16 release frequency, right?
17 MR. KURITZKY: Mm-hmm.
18 CHAIR DIMITRIJEVIC: That technically is 19 different than it was reported for the new reactors, 20 right, because new reactors reported large release 21 frequencies, not the large early release frequencies, 22 and also CCFP and the requirement there was, you know, 23 to meet the safety goal the same as the LERF.
24 So here there is a lot of question 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
98 actually how we define large release frequency. I 1
notice that there is a lot of -- you have some 2
consideration, you know, for that the radiological 3
release is either seven days or two days timeframe.
4 So can you discuss that and what does that 5
actually mean? Is this actually new mission time for 6
Level 2? Actually, can you actually discuss really 7
why is this LERF so big, how does it differ from the 8
one which was reported for advanced reactors, what are 9
those different timeframes considered in this?
10 They make actually a pretty big difference 11 but still LERF is very large. All right.
12 MR. KURITZKY: Okay. So let me just go on 13 to the next slide because that's where I am going to 14 really get into that and the timeframe.
15 So as you mentioned the LRF is relatively 16 large here. It's, I don't know, 70 percent of core 17 damage frequency. Well, inconsistent with the CCFP.
18 So because of that -- So with -- In order to get a 19 better, a more complete understanding of long term 20 severe accident progression and radiological release 21 considerations, we took the severe accident analysis 22 out to a stable state with a 7-day backstop.
23 But as I mentioned earlier, we credit post 24 core damage and mitigative actions up to, around, or 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
99 slightly after vessel breach, but we don't consider 1
additional actions in the longer timeframe after that.
2 So because of that and because there is no 3
consensus mission time for severe accident analysis 4
for core damage frequency, there is a generally 5
understood mission time of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
6 You might take a little bit longer for 7
certain sequences, you know, especially if they are 8
coming up to a cliff edge effect, but, you know, there 9
is kind of a consensus around 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
10 Well there is no such thing for Level 2 11 space for severe accident, so given that we wanted to 12 look into what would be the impact on the results if 13 we used a shorter time, if we terminated the severe 14 accident analysis earlier.
15 So we looked at two different cases, both 16 of them peg to when you enter the Severe Accident 17 Management Guidelines, SAMG entry, which essentially 18 is when core damage occurs.
19 So we looked at one case where we stopped 20 the analysis 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after SAMG entry and in another 21 case it was 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> after SAMG entry. As you 22 mentioned in terms of the timeframe from event 23 initiation, so the base case was seven days after 24 event initiation, the time of core damage will change 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
100 depending on the sequence but it's generally going to 1
be somewhere between a few hours after event 2
initiation up to maybe around 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after event 3
initiation.
4 So if you just want to kind of get a 5
ballpark idea for comparison, SAMG plus 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> is 6
more or less two days after event initiation and SAMG 7
entry plus 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> is roughly three days after event 8
initiation, so you are looking at two, three, and 9
seven days after event initiation here.
10 If you look -- First let me just mention 11 just to get it off the table, large early release 12 frequency you see no change in the numbers, either for 13 the base case or the 2020 FLEX case, and that's 14 because by definition those releases are occurring 15 early on and are occurring before you even get to 36 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> after SAMG entry, so the shorter timeframes make 17 no difference on the results.
18 But that's certainly not the case for the 19 large release frequency. If you look at an LRF in the 20 top table for the circa 2012 case you can see that in 21 the SAMG plus 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> case, well, you know, 22 essentially both of the shorter timeframes, you have 23 a significantly reduced large release frequency.
24 The reason for that is the large release 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
101 frequency of course is composed of all the different 1
release categories in the study, but one of the major 2
drivers of large release frequency is a release 3
category that we call LCF, or Late Containment 4
Failure, and what it really involves is a containment 5
failure tens of hours after vessel breach. It's a 6
quantity static over-pressurization failure of the 7
containment.
8 The representative sequence for that 9
release category in MELCOR shows that the release is 10 occurring by SAMG, by 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> after entering the 11 SAMGs, but it hasn't, the cumulative release hasn't 12 risen to the threshold that we call large, which I 13 think is, I don't remember, it was something like 4 14 percent of cesium or -- It was some fraction of I 15 think it was cesium source term, but it hasn't gotten 16 to that level yet that we would call large.
17 In fact, it doesn't get to that level 18 until almost right before the seven days, in fact. So 19 that's why you see a substantial reduction in LRF for 20 those shorter timeframes.
21 But if you look down at the conditional 22 containment failure probability you see actually that 23 drop off doesn't occur, it only occurs for the 36 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after SAMG entry and that's because that same 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
102 representative sequence for the late containment 1
failure release category has, at 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after 2
entering the SAMGs it hasn't even failed containment 3
yet.
4 The containment fails sometime between 36 5
and 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> after SAMG entry and so that's why you 6
then have the jump up in conditional containment 7
failure probability and then the actual release 8
becomes what we call large near the end of the 7-day 9
period and that's why you see the bump up in LRF only 10 for the 7-day after event initiation.
11 So, again, it's tied to the fact that we 12 had that 7-day severe accident analysis termination 13 time that leads to such a large contribution.
14 Now, again, other plants might have it 15 different, and this is based on the profile, this is 16 based on the dominance of station blackout type 17 scenarios, which without containment heat removal it 18 will slowly build pressure to your failed containment 19 and then the release starts off small and slowly 20 builds up over time, so, of course, other plants might 21 have different results, but this is what we see with 22 this study.
23 But what this also leads us to is the fact 24 that because we don't credit any additional mitigative 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
103 actions after that shortly after vessel breach, so it 1
does tell you that if someone were to credit actions 2
or able to implement some type of actions to prevent 3
containment failure in say two days, within the two 4
days after the event initiation, by for instance 5
recovering containment heat removal or containment 6
venting, then they can prevent a large release because 7
they will prevent the containment from failing and 8
prevent the eventual large release.
9 Well, on the other hand, the flip-side of 10 that, the other big takeaway is that if a licensee or 11 some other applicant were to come in with a Level 2 12 analysis and they generally only carry out their 13 severe accident analysis for let's say 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> after 14 event initiation or 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> after event initiation, 15 they may very underestimate what the plant risk is 16 because they would not capture some of these longer 17 term issues or longer term failures. So that's the 18 two sides of it.
19 The fact that the shorter timeframes means 20 that you do have an opportunity to take additional 21 action to prevent the large release, on the other hand 22 if you can't prevent it and you don't model it you are 23 going to underestimate what the actual event risk is.
24 Did that address your --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
104 CHAIR DIMITRIJEVIC: Oh, yes. Well, this 1
is extremely, I mean, in my opinion, extremely 2
important. Because it may have a really regulatory 3
impact, you know, because the -- okay, so here you 4
consider 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> Level 1 mission time. And most of 5
the new advanced plants consider 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> mission time 6
for both Level 1 and Level 2.
7 But what you are telling us, that this 8
severely underestimates large release frequency. And 9
also, you know, the silent CCFP of 0.1, which is one 10 of the safety goals, if you extended this timeframe to 11 the, you know, seven days after that event, you may 12 not satisfy the surrogate measures for advanced 13 reactors. So in your opinion is this how you see 14 implication of this result?
15 MR. KURITZKY: Well, I'm not in a position 16 to say, because I don't know the -- obviously this 17 very design-specific, the new reactor designs. This 18 is an artifact of the large drive containment design, 19 and the station blackout sequences, and their impact.
20 And some should be going what mitigated activities did 21 you take.
22 But yes, for these types of plants, if you 23 don't mitigate the station blackout, and it just keeps 24 on going, you get containment over pressure, you will 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
105 eventually fail the containment and get a large 1
release.
2 But I don't know the design of the new 3
reactors or the advanced reactors, whether the same 4
types of sequences occur, whether the design features 5
of their containments, et cetera. You know, I can't 6
really extrapolate or speculate what the impact would 7
be for them.
8 (Simultaneous speaking.)
9 MR. KURITZKY: But the overall insight is 10 that, yes, make sure that they are looking at a 11 sufficiently long timeframe that they're not clipping 12 higher potential failure probabilities, containment 13 failure probabilities because they just arbitrarily 14 terminated their analysis at some point in time.
15 Sorry, go head --
16 (Simultaneous speaking.)
17 MEMBER PETTI: This is Dave. Just, you 18 know, many of the advanced reactors do not require 19 power from safety functions. And so the whole station 20 blackout picture looks very, very different, I think, 21 when we actually get, you know, an application.
22 But at least the stuff that I've seen, 23 that's when I was reading it, that's how I was kind of 24 trying to think about it, put it in that context. In 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
106 kind of many of these cases, there are, quote, "cliff 1
edge" effects in some of the designs. Here you're 2
just seeing all the cliff edge effects coming to the 3
fore, right. So, yes.
4 CHAIR DIMITRIJEVIC: All right. Well, I 5
think that this is, I mean, a lot of those plants have 6
a passive feature. I happen to work with one which is 7
very similar to this which is APR. It's very off-site 8
power dependent, in fact there are very similar 9
issues, like the four-loop with Westinghouse.
10 So there this is very applicable stuff.
11 But also the most important point is that maybe those 12 mission times cannot be prescribed in advance. You 13 know, they should be sort of related to what's 14 happening in the severe accidents in the plant 15 afterwards. So you cannot just say, okay, we're going 16 to use either 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />, because that's how 17 it is done.
18 So from my point of view, it's very 19 interesting that this brings up also how long you're 20 going to look in the containment response after the 21 initiation of the event. So I think that will 22 definitely -- I hope you will definitely, you know, 23 have a prominent place in your summary report on the, 24 you know, compare it with the coolant practice.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
107 Well, okay. So all right, well thanks for 1
the discussion.
2 MR. KURITZKY: Okay. And again, thank you 3
for the question. And let me also just mention just 4
two quick follow-up things. One is in this case it's 5
station blackout that's driving that result. But the 6
overall concept that you've got to be careful about 7
how long you run the analysis out is important.
8 It would take me too long to find the 9
actual MELCOR output graph. I'm not even sure it's 10 publicly available. But you see, in the graph you can 11 see, after the containment fails, you see measuring 12 the source term, and I think, again, I think it was 13 the cesium pressure, cesium release. And you see it 14 slowly going up.
15 And so it actually is right before, really 16 something like 6.9 days that you get to what we happen 17 to call, though again that's an arbitrary number too, 18 right, I mean whether you're a few, you know, periods 19 below, or above, or whatever, that's not important.
20 But nonetheless, it's a slow progression.
21 And so if we only did the analysis for two 22 or three days, we would see it never getting near the 23 threshold for large. But as we did do it to seven 24 days, you do see it, you could get that far.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
108 So any type of analysis, and recognizing 1
that I know the NRC is not going to see these detailed 2
analyses from applicants, but I think they have the 3
right to go look at them, you want to see what kind of 4
-- if they do have such analyses, and I guess there's 5
might be with MAP or something.
6 But if the source terms are trending 7
upward, and they just happen to cut the analysis at 8
that point, you know, a fair question is well, hey, 9
what it you extend this thing out? That would just 10 keep going up. Eventually you're going to get to a 11 problem.
12 So the concept about when to terminate the 13 severe accident analysis I think is important because 14 of what the design of the plant is.
15 MR. BLEY: Alan, this is Dennis.
16 MR. KURITZKY: Yes.
17 MR. BLEY: I'm going to emphasize that 18 even more. There ought to be a physical reason for 19 where you stop the analysis, something you can tie 20 back to what's happening in the plant and what might 21 be happening externally that you could use.
22 Just in an aside, 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> came about back 23 during WASH-1400 times. People sat around the room 24 and said, well, after about a day you could probably 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
109 get help from other places that bring you stuff, so 1
you probably don't need to go beyond that. There's no 2
reason for that to hang on as long as it has. There 3
ought to be a basis in physics and physical attributes 4
that determines when you stop that analysis.
5 MR. KURITZKY: And thank you, Dr. Bley.
6 (Simultaneous speaking.)
7 MR. KURITZKY: You're probably one of the 8
only people that were around back then to be able to 9
tell us that.
10 CHAIR DIMITRIJEVIC: And that is also, I 11 don't really -- I was going to go back to the 12 standards, PRA standards. I don't remember what do 13 they say about the mission time, and especially for 14 the Level 2.
15 MR. BLEY: I don't remember exactly, but 16 I know it had some element of, you know, it's not a 17 fixed number. You want to base it on what the 18 analysis is telling you.
19 CHAIR DIMITRIJEVIC: Okay. That would be 20 good then.
21 MR. KURITZKY: Yes, actually I could --
22 two points here. So yes, Dr. Bley, this is an example 23 of the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. It's one of those things where if 24 you say something and repeat it enough it just becomes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
110 fact of type of thing, right, even if there's no basis 1
for it. But I think the standard says you need to 2
reach a safe and stable state.
3 And so you'll see in our Level 1 PRA 4
provincial events, Line 3A, we have a whole discussion 5
on safe and stable. We went back to our event trees 6
and we added some additional nodes for -- and this 7
particularly became important with RCPC LOCAs, because 8
we had cases where we were not getting core damage at 9
24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
10 But you were not in a stable state. The 11 leak was increasing, and you were eventually, if you 12 couldn't take action to ameliorate it, you were going 13 to get to core damage. And so for those cases we 14 added these extra nodes to, you know, make up, 15 consider alternative ways to rod cooling or make up 16 charging to the primary system.
17 And so we added several nodes for that on 18 the basis of getting to a safe and stable state. So 19 I think really, the standard I think leads you to go 20 to a safer, stable state. But there's no definition 21 of what safe and stable means.
22 MR. BLEY: It's certainly not a case where 23 things are continuing to get worse.
24 MR. KURITZKY: Right, exactly.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
111 MR. BLEY: The pressure is still going up, 1
temperature is still going up, that sort of thing.
2 So I'm going to have to go back and look 3
at the Level 1, because I don't remember that when I 4
reviewed it. And I don't think I've gone back to take 5
a look to see what you added in. So that might be 6
something worth pointing out in your final summary 7
report, when it's reasonable to stop the timeframe of 8
an analysis.
9 MEMBER REMPE: But, Dennis, along the same 10 point, when you stopped it back at 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, not 11 including the offsite equipment, you're going for 12 seven days, is that not a bit misleading?
13 (Simultaneous speaking.)
14 MR. KURITZKY: I'm sorry, is that 15 question to me or Dr. Bley?
16 MEMBER REMPE: I'll give it to both of 17 you. But to me that seems a bit misleading.
18 MR. BLEY: Can you sat it again? I didn't 19 parse it --
20 MEMBER REMPE: Well, it didn't include the 21 offsite FLEX equipment, and why go for seven days if 22 you're not including something that could perhaps 23 mitigate what was going on, or at least say it's there 24 and say a 50 percent chance to get flown in, and get 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
112 it installed, and it does something. It sounds like 1
it was totally neglected. But you went for seven 2
days, right?
3 MR. BLEY: No. I agree with you. I mean, 4
they told us what they did. And what they did says 5
under this condition here's what the results are. My 6
worry, that I stated earlier, is because of those 7
assumptions it can be misinterpreted. And it's hard 8
to put warnings about results that are coming because 9
of assumptions. It's hard to put enough warnings 10 there that people really pay attention to them. So, 11 yes, it's a little misleading.
12 MR. KURITZKY: So here's my crack at it, 13 Dr. Rempe. So we recognize that the whole reason for 14 this thing you're seeing on the slide right now is 15 because we recognize that in doing it for seven days 16 we haven't credited other actions beyond essentially 17 the time of the vessel breach or shortly thereafter.
18 So we recognize that there are other 19 actions that could be taken. That was the thing that 20 Dr. Dimitrijevic was mentioning from the Volume 3 21 Report where we had the sensitivity of looking at 22 other potential mitigative actions in the longer 23 timeframe and see what the impact of those were. And 24 that's why we look at these shorter timeframes here.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
113 So we recognize that we're not considering 1
all the things you could do in that longer timeframe.
2 But as it specifically refers to FLEX, and the fact 3
that we only consider those first two phases, and we 4
don't consider the long term phase, is the fact that 5
FLEX is primarily focused on preventing core damage.
6 So the strategies from FLEX aren't really 7
intended to deal with a post-core damage environment 8
and mitigate releases. They're really set up to 9
prevent core damage.
10 (Simultaneous speaking.)
11 MR. BLEY: I've got to jump in, if you 12 don't mind?
13 MR. KURITZKY: Sure.
14 MR. BLEY: Originally that was true. And 15 when representatives of industry came to the ACRS and 16 described these, they were very adamant about that.
17 And as more and more individual plants began to look, 18 they began to see that this would be very useful in 19 other ways.
20 And eventually people backed off of that 21 only consider it pre-core melt. And we've got some --
22 well, I'll leave it there. That's just based on what 23 people brought to ACRS and talked about.
24 I agree with you that that's the way it 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
114 started. I think owners of these plants saw that they 1
could get a lot more out of FLEX than not. And, you 2
know, some of the things that were walked through in 3
meetings with the ACRS and with the industry, and we'd 4
only done that for BWRs, really showed that value. In 5
any case, I think that's not a good argument anymore.
6 MR. KURITZKY: Okay. Also appreciate 7
that, but again, if they want credit for that in the 8
PRA, and certainly we were going to credit it in the 9
PRA, something would have to be showing up in their 10 FLEX implementation plan or procedures to take credit 11 for that, not just that, hey, we would do something.
12 And the second thing is we do, and I think 13 it's in the older report, I know it's in the 14 supporting calculus, but I think it's in the older 15 report too for where they talk about with the Level 2 16 and Level 3 cases.
17 We mentioned that the FLEX, as we're 18 applying it into this sensitivity case, really only 19 impacts the core damage frequency. And so we're not 20 really -- it could be potential things that they could 21 do in the later timeframe in Level 2 or 3 space, but 22 we're not actually crediting them in this particular 23 sensitivity study.
24 And I don't know -- in the internal report 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
115 I know we identify some of the potential things that 1
could be done in Level 2 space. I don't know if 2
that's in the overview report or not, but it would be 3
in the chapter on Level 2 modeling for the sensitivity 4
case.
5 But anyway, the points are all valid.
6 There are definitely things that can be done beyond 7
what we've credited here up until the point, 8
especially you talk about these things that don't 9
become large until seven days.
10 But really, your best shot of avoiding 11 something is preventing the core damage, excuse me, 12 preventing the containment failure. Once the 13 containment has failed, if you're just marking time 14 until the release gets big enough, there's not a heck 15 of a lot you could do.
16 I mean, you can spray on the outside of 17 the wall, right, if you know where the leak is, et 18 cetera. But the bulk of what you can do, the most 19 stuff you can do is before core damage. The next most 20 things you can do is before the containment fails.
21 After that, your tool box is getting thin. And if you 22 wanted to create something at that point, you would 23 have to have procedures for it.
24 MR. BLEY: I'd almost say the most 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
116 satisfying things you could do, rather than the most 1
things you can do. Okay, ha, ha.
2 MR. KURITZKY: All right. I'll go with 3
that. Okay, let's see, I think we've finished this 4
slide, right? Yes, exactly, we were --
5 CHAIR DIMITRIJEVIC: I think we did, yes.
6 MR. KURITZKY: Yes, I think we --
7 CHAIR DIMITRIJEVIC: I mean, you point to 8
this interesting thing there like the CCFP was always, 9
you know, measures the core damage frequency over the, 10 you know, lot of, or a lot of, because nobody was ever 11 sure about the things. But here you have a very good 12 point, and that's a separate measurement.
13 MR. KURITZKY: Right.
14 CHAIR DIMITRIJEVIC: You know, so you 15 point that in the regs. So I think we are done with 16 the slide.
17 MR. KURITZKY: Okay, good, thank you.
18 Moving on to Level 3, it's a Level 3 19 period study, we've got to have Level 3 results. So 20 in this presentation, in the overview report we 21 focused on two specific risk metrics, the two that 22 are associated with the quantitative health 23 objectives, the Commission's safety goal policy 24 statement. And so that's individual early fatality 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
117 risk and individual latent cancer fatality risk.
1 That's not to say that we haven't perused 2
a lot of other measures. Actually in the Level 3 3
report which is Volume 4(e), I think, in the current 4
set, it was 3(d) for internal events, we report 5
results for over, I think, I dozen different risk 6
metrics. And so there's a lot more information there 7
than I'm going to talk about here. But we only did 8
the FLEX evaluations for these two risk metrics.
9 So looking for individual early fatality 10 risk, if you look at this chart you can see that the 11 quantitative health objective is up at the top, the 12 dashed line, 5(e) minus 7.
13 And then down at the bottom you have all 14 the contributors for the different hazards. The blue 15 bars are for the base case, circa 2012. The green 16 bars are for the 2020 FLEX case, and each pair of bars 17 is associated with a different hazard. You have 18 internal events and floods in the beginning and then 19 fire, seismic, wind, and then all hazards combined.
20 And so if you look at these bars, you can 21 see that the biggest contributors to individual early 22 fatality risk come from the internal events and the 23 seismic events. And that's because internal events 24 you have --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
118 CHAIR DIMITRIJEVIC: Alan?
1 MR. KURITZKY: Yes?
2 CHAIR DIMITRIJEVIC: I always said 3
incredible, this is totally personal, but I have an 4
incredible problem with presenting such low numbers.
5 You know, we are talking here in something that is 6
three trillion years, you know, one event in three 7
trillion years. And the age of universe or big bang 8
is, like, three billion years.
9 So I mean, you know, introducing such 10 small numbers in many period study, I'd really, you 11 know, couldn't we just say negligible or something.
12 Do we really gain anything? And then we compared 13 those ten to minus 30 numbers to each other.
14 That's really something I have a really --
15 it's totally, you know, it's totally, it doesn't make 16 sense, in my opinion. So I don't know how to avoid 17 that. I can see you are showing that you did the 18 calculation, and there is a difference.
19 All right, Tom has raised his hand. Tom?
20 MEMBER ROBERTS: If I could, just for a 21 minute --
22 MR. KURITZKY: Yes.
23 MEMBER ROBERTS: Yes, I'd like to add to 24 that question, right, before you answer it.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
119 MR. KURITZKY: Okay.
1 MEMBER ROBERTS: It wasn't clear in my 2
reading the report how those numbers came about.
3 Usually there's about a six order of magnitude plus 4
reduction between the larger release frequency and 5
these early fatality risks.
6 So in pouring through the reports, the 7
only thing I found was very deep in the Volume 4 (e) 8 was a discussion about meteorology and wind. And it's 9
kind of hard to imagine that you get six orders of 10 magnitude reduction from meteorology and wind.
11 So I just wondered if you'd, not to get 12 beyond Vesna's point about showing such low numbers, 13 how do you calculate them, and how have you concluded 14 they're reasonable?
15 MR. KURITZKY: Okay. So a good question, 16 and that's one where I'm going phone a friend and.
17 And I'm sure the friend knows who is going to get 18 tapped in a second. But before I go there, you know, 19 I do echo your concern, Dr. Dimitrijevic, about the 20 low numbers. But as you said, you know, what's the 21 alternative?
22 I do know that in the actual overview 23 report, in the summary table we have up in the key 24 messages, you know, the Section 2 both for the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
120 internal event and the fire, seismic, wind one, we put 1
in the table. We put in approximately zero as opposed 2
to Action Number 4 in that exact reason because, you 3
know, such low numbers look odd.
4 Then we have a footnote that I think 5
actually provides the actual numbers down in the fine 6
print. But in the table itself, we just put 7
approximately zero, because the numbers are so low.
8 But nonetheless, I mean, that's what the calculation 9
shows.
10 If you want to know exactly, going to the 11 question about why they're so low, Dr. Compton, are 12 you ready to jump in? There's more factors than just 13 wind that are involved. But, Dr. Compton, are you 14 online?
15 MR. COMPTON: Sure. I will jump in. I 16 don't know if I can fully answer the question, because 17 I can speak about this for hours when I want to. But 18 I will highlight a few things that maybe will hit some 19 things.
20 And one of the things, I do understand 21 these are really low numbers, and interpreting them in 22 kind of, I don't know, a realistic or actual point of 23 view, it's hard to understand what they mean.
24 They're useful, the actual numbers are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
121 useful to me as an analyst, because it helps me to 1
discern, hopefully a reader who spends a lot of time, 2
to discern why the numbers are that low. And it's 3
hard to do that when you're giving kind of 4
inequalities, you know, less than.
5 So anyway, so there's -- and I'm glad you 6
found it, it was buried in the, as you mentioned, in 7
the Level 3 report. There's basically four reasons 8
why the early fatality risk is so low. Actually, it's 9
not really just the early fatality risk, it's these 10 are the reasons why high doses, whether they lead to 11 early fatalities or not, these are the reasons why 12 high doses are low.
13 So the first reason is just that, as Alan 14 said, they only arise under bypass scenarios. You 15 only get those high doses under bypass scenarios which 16 are, you know, very large and very fast releases. And 17 there are some tables and figures in the report, I 18 won't go through them right now unless you ask, but 19 they can kind of illustrate that.
20 The second is the fact that there is a 21 pretty low, a very low frequency of what I call 22 adverse meteorology, even for those cases that lead to 23 early health effects. Only about one percent of the 24 weather, you know, trials, the actual weather 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
122 conditions, led to early fatalities in those, I think, 1
three or four cases.
2 So another factor is the fact that the 3
range at which you can get those early fatalities is 4
very short. And at least at this site the population 5
is very sparse. There is only, I think within one 6
mile of the site boundary, there is only kind of one 7
out of the 64 directions has a populated sector, one 8
or two. I don't remember the exact number. But the 9
point is that you have to be downwind. And if there's 10 not a lot of people downwind, that will drive the risk 11 lower.
12 And then the final factor is there's a 13 very low likelihood of delayed or slow evacuation of 14 the populations close to the site. We looked at, in 15 this study, we looked at a pretty fine grain detail 16 about, you know, the early evacuees, kind of the 17 medium evacuees, the late evacuees, the non-evacuees.
18 And so again, it doesn't, even if you met 19 the other conditions, you had a bypass, and you had 20 the right weather condition, even only the very late 21 evacuees, or those who don't evacuate, show any non-22 zero or early fatality risk.
23 So I think if you start adding all those 24 pieces together, and you look at the actual numbers, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
123 and you can see that they kind of hang together when 1
you slow the evacuation down, those low numbers get a 2
little bit larger.
3 And then if you start, you know, kind of 4
adding in things, you can see what's going to make 5
those numbers go up. And to me that's what's 6
important, is not just the low number but kind of 7
understanding that the story hangs together and what 8
could defeat that story.
9 That's all I have. Hopefully that is 10 helpful in some way.
11 MR. KURITZKY: It looks like we have a 12 question. Dr. Bley?
13 MR. BLEY: Yes. Not so much a question as 14 a comment. This same issue, of course, came up with 15 WASH-1400. And the primary authors of that report 16 testified before Congress, and they did something 17 similar to what Keith did. They wrote it out as a 18 product and showed these various things.
19 I think if you're going to keep something 20 like this in the report, even if it's in a subtle 21 place, having that litany of three things that all 22 have to occur to get fatalities makes it a lot easier 23 to see why there's, you know, another ten to the minus 24 six added onto the numbers we saw earlier.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
124 And Keith did it really well. And I think 1
it's just a sentence or a note to do that. But I 2
think it really helps a lot.
3 MEMBER ROBERTS: Yes, Dennis, I agree with 4
that. I think it's important to understand the plant-5 specific nature of a very large part of that delta.
6 You
- know, one concern I'd have is somebody 7
interpreting this report as refuting the NUREG 1860 8
discussion of why LERF is a reasonable surrogate 9
metric for the early fatality risk. Because you can 10 interpret this plot that that number should be a lot 11 higher and it maybe not even be a concern.
12 Because for any reasonable value of LRF 13 you're still going to meet the QHO for early fatality.
14 So I think it's important to have that perspective, 15 that there's a site-specific nature to this. And if 16 you have a slightly more adverse situation at the 17 site, or a different meteorology, these numbers change 18 by orders of magnitude. So again, just to, you know, 19 just a thought.
20 CHAIR DIMITRIJEVIC: Okay. If you just 21 point out exactly what I was, and I'm actually 22 questioning how the surrogate things makes -- because 23 it was based on the old study data. And you cannot 24 really avoid to question this, because we are not 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
125 talking one order of magnitude. You know, we are 1
taking about like five orders of magnitude.
2 So I don't know what kind of the wind, the 3
population will change that. But basically, based on 4
this, maybe some evacuation probability. So this is 5
why -- I talked directly with Dennis. You should 6
really present -- you know, you claim a small 7
probability, which I can also question why is the 8
LOCA, you know, re-sequenced as such a small 9
probability?
10 But he said that we're not going to go in 11 technical details. And that could be part of 12 uncertainty analysis. But this has all started with 13 probably something, you know, the one E to the minus 14 six. And then we are adding now ten to the minus five 15 factors based on evacuation, wind and population.
16 I would like to see how those factors add 17 up. Or otherwise I definitely will plan to question, 18 and I already do this, connection of QHOs to the 19 surrogate measures.
20 MR. KURITZKY: Okay. Well, thank you all 21 for those comments.
22 MR. EVANS: It looks like Keith has his 23 hand up. I'm sorry, Alan.
24 MR. KURITZKY: Oh, sure.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
126 MR. COMPTON: Sure. I would just, I 1
appreciate all the comments. And I have struggled for 2
a long time to try to really understand why we get 3
such different results in this study than we get in 4
NUREG 1150. I think is very much related to these 5
three things.
6 I have looked at the, you know, looked at 7
the specific source terms in NUREG 1150. And I've 8
been trying, and I'm still working on it, I haven't 9
come to really a clear exposition but, you know, to 10 try -- it's hard to compare them and do this 11 percentage of the reduction comes from this factor.
12 But the modeling was very different, you 13 know, the number of population sectors were different.
14 The meteorology was different, the approach to 15 modeling evacuation was different. And all that is 16 simply saying is that it is an important thing. It 17 may take some time to do that comparison and that 18 evaluation properly. But I do think it's important to 19 do.
20 CHAIR DIMITRIJEVIC: Well, and therefore 21 you still, if this plant was going to submit under 22 the, you know, the QHO requirement, it could make a 23 case that don't really -- that can have a, you know, 24 as Jon point out in some of his comments we exchanged, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
127 we can have a little fun. This is for this specific 1
plan, because all of those submittals will be based on 2
the plant, you know, specific locations and cases.
3 All right. Okay. I mean, I think this is 4
very interesting and really different for what was in 5
the original NUREG. So definitely it deserves a lot 6
of discussions.
7 MR. KURITZKY: I think it also, just to 8
mention too, I don't think he -- I think he's still 9
out of the country, but Sunil Weerakkody, he has been 10 also looking into, you know, as I'll get to in the 11 final slide in this part of the presentation, we'll 12 talk about the difference in margin to the QHOs versus 13 the surrogate risk metrics.
14 And he is fond of pointing out that the 15 safety goal policy statement doesn't just talk about 16 the quantitative health objectives. You know, it 17 talks about qualitative objectives too.
18 And it talks specifically about the need, 19 you know, for the nuclear industry to -- if any type 20 of severe accident, so it says that the staff, I 21 guess, the administrator, I don't remember the exact 22 word, should be striving to have no severe accidents.
23 So it's not just a question of meeting one QHO number 24 or another QHO number. The striving should be to have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
128 as low a CDF as possible, to have no core damage.
1 So, you know, it's one of these cases 2
where I think the QHOs have kind of taken on -- I 3
mean, they're certainly important. There's no 4
question about it. But they're not the end all and be 5
all and the single thing by which all things should be 6
measured. So we just have to, you know, keep that in 7
mind too.
8 But again, as Keith said, we will try to 9
dig up some more information that we can describe how 10 you get from, for instance, a lower frequency or a 11 core damage frequency to the individual early fatality 12 risk numbers and kind of show all the multipliers that 13 bring the value down, to the extent that we can.
14 CHAIR DIMITRIJEVIC: Well, you know, I 15 have this discussion with my colleagues in ACRS about 16 the safety goals ongoing. And I have not been really, 17 I have to say, I have not been very successful in 18 explaining well my concerns. So to be honest with 19 you, I was counting with your staff showing them you 20 know, and also when comes to uncertainties associated 21 with these quantifications. So I am sort of, like, 22 very interested in this results.
23 MR. KURITZKY: Okay. Well, thank you very 24 much. Hopefully this will provide some information 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
129 for you.
1 Going back now, if I could just wrap up 2
this slide, so as I was mentioning, the drivers here 3
are internal events and seismic provincial events, as 4
we were discussing before, the interfacing system 5
LOCAs, particularly those where you have the auxiliary 6
building failed, so you don't have any retention 7
really of the source, the radionuclides.
8 And then for seismic, it's primarily two 9
types of steam generator tube rupture. I think we may 10 have already mentioned them. It was the pressure 11 induced steam generator tube ruptures that occur prior 12 to core damage that generally come about because of an 13 ATWS event. And then there are the -- and in this 14 case the seismic leads to a higher frequency of ATWS, 15 that you get for internal events.
16 And then there's also the post core 17 damage, thermally induced steam generator tube 18 ruptures, which also come about for a number of the 19 sequences under the seismic initiator. So that's why 20 the internal event and the seismic are the drivers 21 here.
22 You certainly can't tell by the 23 logarithmic curve, you know, plaque, but if you look 24 at the numbers you see there's not very much 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
130 reduction. The FLEX case doesn't really reduce 1
individual early fatality risk much. And that's, 2
again, because these types of sequences that I just 3
mentioned, they're not the ones that the FLEX guides 4
used in the turbine driven aux feeder are targeted 5
towards.
6 The one area where you do see that 7
reduction is for the wind. If you look at the wind 8
you see that it drops from roughly 90 minus 14 to 40, 9
minus 14. And the reason the wind drops so much is 10 because the early fatality risk for wind is almost 11 entirely made up of station blackout sequences where 12 the wind causes a loss of offsite power.
13 And then you have a combination of wind 14 induced or random failures of emergency AC components 15 leading to station blackout sequences that eventually 16 propagate to these post core damage thermally induced 17 steam generator tube ruptures.
18 And so those are things that the FLEX and 19 turbine driven aux feed, extended turbine driven aux 20 feed, can ameliorate. And so you see that reduction 21 for the wind. Overall, you just see a 12 percent 22 reduction for all hazards combined. But again, as we 23 were just mentioning, there's a huge margin to the --
24 for the reasons that Keith specified a few minutes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
131 ago.
1 Okay. If we've beaten the early fatality 2
risk horse enough, we'll move on to latent cancer 3
fatality risk, the other QHO. And here we see that 4
the QHO is up at the top at 210 to the minus 6. You 5
see that you don't really have the same type of 6
margins you did for early fatality risk, but they're 7
not insignificant.
8 The drivers here, again a little difficult 9
to see because of the logarithmic scale, but the 10 drivers here are internal events and internal fires.
11 And they actually combine. You can't really tell from 12 here, but I know the numbers. And they combine to 13 about 80 percent of individual latent fatality risk.
14 And if you recall, that's essentially what 15 they were in terms of a contribution to core damage 16 frequency. And then that's for the base, the circa 17 2012 case. And the reason that you have that time is 18 because both of these, the latent fatality risk from 19 these two types of initiators are based on two primary 20 release categories.
21 And those are the ones that are driven by 22 station blackout sequences, so you get that same type 23 of profile in there for the same type of impact from 24 FLEX and the turbine driven aux feed as you do for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
132 core damage. And so that's why they're very analogous 1
in terms of the impact of FLEX.
2 If you look at the overall drop, you see 3
it goes from 6.5(e) minus 8 to 4(e) minus 8, a drop of 4
just under 40 percent, again, very similar to the 5
total CDF reduction for all hazards for FLEX.
6 And again, Keith mentioned too that the 7
reasons we still have a fairly good margin to the QHO 8
here is some of the same reasons that we have a lot of 9
margin from the early fatality risk. And that's the 10 fact that we really don't have many sequences that 11 provide dose -- the frequency of sequences that 12 provide dose in the early phase are very low, okay, 13 it's those bypass sequences that are very low 14 frequencies.
15 And effective protective actions are taken 16 that minimize how much exposure you receive at a later 17 phase too, essentially to the habitability criterion.
18 So it's not until people move back later that they get 19 really that exposure.
20 They're not getting exposed right after 21 the accident from the radioactive cloud, let's say.
22 It's more exposure when they move back later in time.
23 So that's why you still have a fairly good margin to 24 latent cancer fatality QHO.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
133 Okay. But for latent fatality risk, 1
there's also a number of key assumptions that were 2
made or boundary conditions that we wanted to look to 3
see what the impact of those would be. So looking at 4
this chart, the first two columns you see base case 5
and the 2020 FLEX case that was -- the circa 2012 case 6
has a margin of around 30 to the QHO. The 2020 FLEX 7
case, that increases to around 50.
8 So we also wanted to look at the accident 9
termination timing. Remember, that was the thing we 10 were discussing back in Level 2 space where we run the 11 accident analysis out to seven days, but we also 12 looked at a couple of shorter timeframes.
13 So we wanted to see what was the impact on 14 latent fatality risk if we terminated the severe 15 accident analysis, therefore the releases, at 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> 16 after SAMG entry. And so that's this third bar that 17 you see. And in doing so, latent fatality risk, now 18 you have a margin of about 150 to the QHO. So it 19 definitely makes an impact.
20 It doesn't look as big on here, because 21 it's a logarithmic scale, but it's a significant 22 impact if you terminate those accidents at 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> 23 after SAMG entry.
24 And then the other thing we really wanted 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
134 to look at was the low dose model. In the base case, 1
what we used is the linear no-threshold, LNT model for 2
low dose exposure. And that is something that is 3
consistent with how the NRC, you know, that we 4
generally use the LNT in regulatory applications 5
involving dose modeling.
6 But there's also a lot of -- there isn't 7
really consensus in, let's say, the health physics 8
community or the technical community about how you 9
should be treating low doses. And is there some 10 threshold below which it's not meaningful to expedite 11 a number of statistical cancers that may increase and 12 statistical cancers that may occur?
13 And so we did an example where we found a 14 threshold, we actually picked one from a 2010 Health 15 Physics Society paper. And it had some thresholds for 16 low dose to low dose modeling, and we applied those.
17 And here you see a very substantial reduction in 18 individual latent fatality risk.
19 And the reason, it's not surprising 20 because, as I was just mentioning, the risk really 21 comes, the latent cancer fatality risk really comes 22 when people move back. And they spend the rest of 23 their lives with a very minimal increase in background 24 radiation.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
135 So it doesn't take much of a threshold to 1
eliminate those increases in statistical cancer 2
deaths. So if you do have some kind of threshold, 3
even if it's a fairly low one, you can have a very big 4
impact on the calculations.
5 CHAIR DIMITRIJEVIC: Yes, I find this also 6
very interesting, Alan. And I appreciate that you 7
guys say, no, we got material from Scott and to show 8
the Commissioners, they're not ready to consider this.
9 You know, they're staying with a non-threshold models.
10 But you pointed in those volumes that 11 those, you know, this alternative, those truncation 12 have a scientific merit and should be considered. And 13 this is very interesting since utility study is 14 showing the, you know, two order of magnitude is 15 reducing latent risk. So I appreciate that you 16 considered that. That was very interesting from my 17 point of view too.
18 I see that Keith has his hand raised.
19 MR. COMPTON: Yes, thank you. This is 20 Keith Compton from the Office of Research. And I did 21 want to emphasize one of the things. I think this 22 particular sensitivity analysis can be very 23 insightful.
24 It can be a bit challenging to interpret, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
136 because it is mathematically a threshold model, but 1
it's not a model that's -- it's not implying that 2
there is a biological threshold. It's simply a model 3
where we don't quantify the risk below the threshold, 4
if that makes sense.
5 In other words it's not -- and that's kind 6
of the challenge, is that what it's really telling you 7
is how much of the risk is coming from, you know, 8
doses that are significantly higher than background.
9 How much of it is coming from moderate or high doses.
10 So I just wanted to --
11 CHAIR DIMITRIJEVIC: Right, yes. And 12 that's, as Alan pointed out, they're just really 13 important, because this is where you are allowed to 14 return. You know, background radiation is acceptable 15 for people to return to the zone. And also, so it's 16 basically low. The big influence comes from those 17 long term long doses.
18 MR. COMPTON: That's exactly right.
19 CHAIR DIMITRIJEVIC: Yes. And also I 20 notice that you also for this, the late fatalities, 21 you considered this release category which is 22 basically intact containment, you know, so basically 23 just based on the tech specs leakage. It also has 24 contribution to this latent fatality risk. You know, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
137 obviously this is very interesting question to 1
discuss.
2 Ron, you have your hand raised and --
3 MEMBER BALLINGER: Yes. It's likely to be 4
way, way too early, but is there any data that's 5
coming out with respect so Fukushima and Chernobyl 6
where people have moved back into the area? Is there 7
any data that suggests that there'll be a big 8
reduction?
9 (Simultaneous speaking.)
10 MR. KURITZKY: Keith, I don't have any 11 information on that.
12 MR. COMPTON: Is the question is there 13 information from Chernobyl or Fukushima suggesting 14 that there would be a big reduction in kind of cancer 15 risk?
16 MEMBER BALLINGER: Right.
17 MR. COMPTON: So yes, so I guess the 18 question would be, or a way to put that is, you know, 19 tracking what the cancer fatality risk coefficients 20 are for exposures that are in those, you know, low 21 dose ranges, whatever works.
22 MEMBER BALLINGER: Yes. You would have an 23 expectation based on an LMT model. And then you could 24 compare it with what actually happens.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
138 MR. COMPTON: Right. So I'm going to be 1
careful, because I used the results of the health 2
physicists and the radio biologists, but I'm not an 3
expert on it. But I have not -- I know that, you 4
know, there's a lot of factors that or there's a lot 5
of cohorts that do contribute to generating the cancer 6
risk coefficients.
7 I don't know off the top of my head if 8
Fukushima or Chernobyl cohorts have been studied and 9
their doses quantified, and their cancer risk tracked 10 enough --
11 (Simultaneous speaking.)
12 MEMBER BALLINGER: The Fukushima ones 13 might have been, but I don't know.
14 MR. COMPTON: Yes. So I don't know. But 15 the way that it would feed in, again, to me, because 16 I would want to be careful about staying within my 17 subject matter expertise, that would feed through the, 18 you know, we would look for -- is there consensus 19 guidance on what the cancer risk coefficients would 20 be, if that makes sense. So I would still follow, 21 kind of, the guidance of federal guidance reports and, 22 you know, those who are qualified to make those 23 judgements.
24 MEMBER BALLINGER: You know the old 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
139 saying, if you have the data use it. If you don't 1
have the data use color.
2 (Laughter.)
3 MR. COMPTON: We do have a lot of colors 4
in our report, so we've at least done that.
5 CHAIR DIMITRIJEVIC: Okay, Dennis, you 6
have a hand up.
7 MR. BLEY: I couldn't get my mic open.
8 This is follow-up probably for Keith, maybe others.
9 Some time ago I thought I'd seen a draft Br8 report 10 from the National Academy. But there was never a Br8 11 published. Do you know anything about that? Is there 12 something underway? I'm not sure of that.
13 And the second thing is who are the dose 14 experts at NRC? I'm not sure I know.
15 MR. COMPTON: Well, again, I'll be 16 cautious. I've not heard anything about a Br8. I do 17 know that there was a Br7.
18 MR. BLEY: That's been over ten years ago.
19 MR. COMPTON: Yes. And the question is 20 who much do things, you know, change. They changed a 21 lot back in the early decades, and then they're 22 continuing to get reexamined.
23 We're basing our models of some reports 24 that were done by Keith Eckerman for the SOARCA 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
140 project, which are kind of circa 1990s, early 2000 1
vintage, that were used for Federal Guidance Report 2
- 13. So in answer to -- so hopefully that somewhat 3
responds to that question.
4 But in terms of who are the people at the 5
NRC, I will say that the person at the NRC that I 6
typically talk to, to kind of make sure that we're 7
staying reasonably abreast with the professional 8
community, would be Terry Brock. I know that there is 9
others that also have some expertise, but Terry is the 10 one that I kind of rely on to keep me honest.
11 MR. BLEY: Okay. Thanks. You said 12 Federal Guidance Report 13? Is that the name of it?
13 MR. COMPTON: Federal Guidance Report 13, 14 it's an EPA publication. It's a federal guidance 15 report published in 1999. Some extra detail came out 16 a few years later. And, again, I need to be careful, 17 to not misspeak. But I think it's, I'll just say 18 broadly consistent with ICRP 60.
19 So it's, you know, it's produced by the 20 same community, the same technical community that 21 works on the ICRP reports. And I think that there's 22 probably overlap. And, you know, that community is 23 much more tied into the BR community.
24 So the sources, communities of expertise 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
141 would be, to me, would be ICRP, EPA's Federal Guidance 1
- Report, they produce WISC coefficients for 2
radionuclides, and then also the BR community. So we 3
try to, you know, keep an eye out to see if there's a 4
big shift in those. But I'm not aware of any, you 5
know, kind of, no pun intended, no seismic shifts --
6 (Laughter.)
7 MR. COMPTON: -- in how that, you know, in 8
the numbers or how that would be done. But we're 9
always trying to make sure we're not surprised by 10 anything.
11 MR. BLEY: Okay, thank you. And I 12 personally have never seen that EPA report, so I'll go 13 look for that. Thank you.
14 MR. COMPTON: Yes. Federal Guidance 15 Report 13, and then, yes, it's in the references. It 16 should be in the references. I should just double 17 check that.
18 MR. KURITZKY: Yes, I think in Volume 19 4(e). I think Volume 4(e) or certainly Volume 3(d),
20 I think you'll see a reference to those, to federal 21 guidance.
22 MR. COMPTON: Yes. And that would also 23 give the reference to, I think, the Eckerman report 24 that we, you know, so you can follow the reference 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
142 chain back from what we did to where it came from.
1 CHAIR DIMITRIJEVIC: Okay. Scott has a 2
hand raised. Scott, please?
3 MR. MOORE: Yes. Thank you, Chairman.
4 Just in answer, Dennis, to your question, there are 5
radiation protection experts, senior level radiation 6
protection experts, HPs, throughout the agency in the 7
office of Research, in NMSS, and in NRR. And in some 8
of those organizations, they're actually sub-units 9
that specialize on Rad Pro. So it's throughout NRC.
10 MR. BLEY: Thank you.
11 CHAIR DIMITRIJEVIC: Thanks. Okay, next 12 slide, Alan?
13 MR. KURITZKY: So this is the last slide 14 for this presentation. It's just really a roll up of 15 the results that we have discussed over the previous 16 couple of hours. So this summarizes the Level 1, 2, 17 and 3 results. We have core damage frequency, we have 18 the LRF, the L-R-F, individual early fatality risk and 19 individual latent cancer fatality risk for both the 20 circa 2012 and 2020 FLEX cases.
21 And you can see from this chart that the 22 core damage frequency, the large release frequency, 23 and the individual latent cancer fatality risk all 24 reduced by just about the same amount. And again, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
143 that goes back to what I was saying before.
1 Essentially the profile of accidents contributing to 2
these are pretty much the same as mostly these station 3
blackout sequences. You have the same general effect 4
on, you know, the FLEX has generally the same effect 5
on these, all three of these metrics.
6 For large early release frequency, it's a 7
little bit lower. It's 29 percent. Again, that has 8
to do with some of the fact that, as we mentioned 9
before, you have contributions from some of the techs 10 in sequences that FLEX is not really designed for, and 11 even moreso for individual early fatality risk where 12 you're really being driven here by those containment 13 bypass, you know, the interfacing system LOCAs that 14 don't -- for which the FLEX and the extended turbine 15 driven aux feeder don't have much impact.
16 So I think that the bottom line is that 17 all these results show that, when you consider this 18 plant at this site, there was substantial margin to 19 the QHOs. There was those green numbers you see down 20 in the bottom two rows.
21 But you also have to recognize that, 22 looking at the purple numbers up above, that there is 23 a lot less margin to the surrogate risk metrics, you 24 know, CDF and LRF. Now what you make of that and what 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
144 you want to do with that information is not part of 1
this project. But it is certainly interesting 2
information, as I think, Dr. Dimitrijevic, you had 3
mentioned before. It's certainly interesting insight.
4 CHAIR DIMITRIJEVIC:
- Exactly, very 5
interesting insight. This slide is actually telling 6
us a lot. Well, okay. Great, I want to thank you so 7
much. I just want to point something which was 8
missing from our discussion.
9 You know, you said there were numerous 10 meetings, and Hossein has summarized history of our 11 interaction. But we only wrote the one letter after 12 the first meeting when this project was approved. And 13 there we said that -- one the things that ACRS said in 14 this letter that this project needs to have an 15 extensive characterization and quantification of 16 uncertainties.
17 And we have not touched uncertainties in 18 this presentation. So I hope we will change that when 19 you present to us in the full committee in November.
20 Because it is -- some very interesting things happened 21 there which are really mind-boggling for me. And I'm 22 very curious what's going on.
23 And it's one that you have concluded that 24 too many -- including too many basic events, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
145 uncertainties, masks events, masks parameters which 1
breed uncertainties. Because all uncertainty results, 2
as you presented with those volumes, are very narrow.
3 There is no large uncertainty which is really strange, 4
and especially I know that you address some modeling 5
uncertainties through the sensitivity study, so 6
alternative studies.
7 But these conclusions, the number of the 8
basic event uncertainties which include the masks, the 9
large uncertainties is extremely interesting to me.
10 So I hope that we will have a chance to have a 11 discussion on your uncertainty results and what does 12 this actually mean.
13 And also I'm very curious, you know, how 14 did you consider the modeling uncertainties. And are 15 the sensitivity studies on the different assumptions 16 enough to address that. So if you want to just tell 17 me this conclusion on the too many uncertainties mask 18 the large ones, ha, ha, I am very curious about it.
19 MR. KURITZKY: Okay, so fair enough.
20 We're at a loss for time, but I don't want to go into 21 a lot of detail. And honestly I probably -- even if 22 I had more time there's only so much detail I can go 23 into. This is not my area of expertise.
24 But in discussing this at the time I was 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
146 with Saleem, who knows much more about it than I do, 1
but our pre-supposition, you know, we were trying to 2
figure out why we were having such a tight parameter 3
on uncertainty distributions.
4 And we had supposed that it might be due 5
to, as you were mentioning, the fact that there are so 6
many basic events, and when you're sampling from so 7
many events, that it tends to mask, even if you were 8
to pull something from the tail of one event, you have 9
other things that have to fail in the cut-set and you 10 can generally pull more from the -- statistically 11 you'll pull more from the center of those 12 distributions, and it kind dampens out, you know, the 13 tails on the distribution.
14 If that's true, why is that not true for 15 every PRA? That one I have a hard time answering. So 16 is it --
17 CHAIR DIMITRIJEVIC: Because logically it 18 doesn't make sense, because we could just do 19 uncertainty distribution just based on those 20 requirements, then don't consider the other ones which 21 we don't think they contribute. And then we try to 22 reach results of two.
23 You know, you don't have to have 1,000 24 basic events. If we are concerned, like your 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
147 parameter for ELAP, and to the agreement extended to 1
the driven operation, this P obviously has a large 2
uncertainty. It's probably uniformly distributed 3
between zero and one.
4 That itself has a big impact on the result 5
distribution. Do we gain by including the 2,000 other 6
basic events. I mean, you know, there are so many 7
questions to address that.
8 MR. KURITZKY: Right. And the whole idea 9
of uncertainty analysis, and in this particular case 10 for parametric uncertainty analysis, it is something 11 that would be its own project. It's not something 12 we're going to resolve here. We just notice that the 13 results are very tight. We took a little look into 14 it. We tried to figure out, you know, guess why we 15 thought it was happening.
16 There are a couple of things that we 17 thought about. The one about the basic events, we did 18 do a, you know, Saleem had done a little bit of 19 checking that by -- you know, the whole model for 20 internal events, I think, the ratio from the 95th to 21 fifth percentile was like around a factor of eight.
22 And Saleem went and did it for just the weather 23 related loop events. So we greatly reduced the size 24 of the model.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
148 And when he ran the uncertainty analysis 1
for that, he got a ratio from the 95th to the fifth 2
percentile almost a factor of six -- around a factor 3
of 16, so twice as wide a distribution. And so that, 4
in that one case, which could be an anomaly, but in 5
that one case it made us, like, hey, the more you 6
reduce the size of the model, now you're starting to 7
get that bigger spread in the results.
8 But that's far from a scientific proof.
9 We don't have anything written up on that, because, 10 you know, it's just one example. And we have no idea 11 whether that would hold in a more broad sense. So 12 unfortunately we're not, as far as this project, going 13 to have a good answer for that.
14 I did take a look at to see what type of 15 spread you had 95th to fifth for some other periods.
16 I looked at the NUREG 1150 models, and those, for the 17 PWRs you were, you know, Surry, Sequoyah and Zion, you 18 were looking at numbers that were in that ballpark 19 somewhere between eight and 20. And we had, you know, 20 our spread was eight.
21 For the BWRs, Peach Bottom and Grand Gulf, 22 you had much bigger ones. You had, I think, 40 for 23 Peach Bottom, and I think 70 for Grand Gulf. And they 24 have much lower CDFs, but also they have much bigger 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
149 spreads. So I don't know exactly why ours is on the 1
low end of that. But it's not unrealistic compared to 2
what we've seen for the other PWRs and NUREG 1150.
3 I couldn't find that information in the 4
NUREG 1560 in the IPEs but --
5 CHAIR DIMITRIJEVIC: I know that we 6
haven't too much time now left for us, but I think 7
that's one of the discussions I would like to have, 8
because I think the uncertainty treatment in the PRA 9
is one area which definitely can benefit from the many 10 insights.
11 You know, I noticed that you didn't really 12 consider the uncertainties in max input parameters, in 13 that, you know, the uncertainties connected with, 14 like, containment failure location, size, things like 15 that. I'm very curious how they can be considered.
16 Sometimes we do, like, it will be nice to 17 see the summary sensitivity cases. The sensitivity 18 cases just show us something sensitive or not. But it 19 doesn't really address uncertainty associated with 20 that.
21 Because they somehow stand on the side of 22 the -- and this is one of the issues which I have in 23 many of our reviews, is what is the good uncertainty 24 analyzed, you know, if you do the 20 sensitivity 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
150 cases, each for each other, but they don't show as 1
uncertainty contributor, you know.
2 Because we see a lot of the passive 3
systems which we don't really have enough data. But 4
then we show very -- model uncertainly distribution.
5 And there is a -- I think the uncertainty analysis can 6
contribute a lot from this project and others maybe.
7 So maybe that's too much burden for this project. But 8
anyway, okay.
9 So Members, do we have any additional 10 comments?
11 Should we call for public comments? The 12 people on the public line, if you would like to make 13 a comment, then unmute yourself, and introduce 14 yourself. I see two people, Edwin and Victoria.
15 Edwin Lyman, please?
16 MR. LYMAN: Yes, hi. This is Ed Lyman 17 from the Union of Concerned Scientists. Can you hear 18 me okay?
19 CHAIR DIMITRIJEVIC: Yes.
20 MR. LYMAN: All right, thanks. Yes, a 21 couple of comments. The first, on the issue of the 22 LRF versus the QHOs, and it seems to me, I didn't hear 23 this brought up, maybe I wasn't listening, but the 24 working definition of LRF, and I'm reading it here 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
151 from NRC document, is that it's core damage accidents 1
that can lead to large unmitigated releases from 2
containment before effective evacuation of the nearby 3
population.
4 But it sounds like, from Dr. Compton's 5
response, that they are considering essentially an 6
effective evacuation and crediting it. So if that's 7
true, it seems like there's an inconsistency between 8
what they call LRF and what the standard definition 9
is. So I'd appreciate some clarification on that.
10 MR. KURITZKY: So if I could, just on the 11 one first, Dr. Lyman, so what Keith is talking about 12 is in the --
13 CHAIR DIMITRIJEVIC: Alan, we don't really 14 necessarily respond here --
15 MR. KURITZKY: Oh, okay.
16 CHAIR DIMITRIJEVIC: -- to outside 17 comments. We will just accept them. But we don't 18 really get in discussions.
19 MR. KURITZKY: Okay, sorry.
20 CHAIR DIMITRIJEVIC: Sorry about that, and 21 sorry I didn't -- So Victoria?
22 MR. LYMAN: No, I'm sorry, ma'am. I have 23 a couple more questions.
24 CHAIR DIMITRIJEVIC: Oh, sorry.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
152 MR. LYMAN: Sorry. You know, on the use 1
of thresholds and the general application of risk 2
coefficients, it seems like this work is rapidly going 3
in the direction of the dinosaurs.
4 Well, first of all, as it was pointed out, 5
the Commission itself rejected a
petition on 6
revisiting the meaning of threshold model. And it 7
would seem like why are any offices in the NRC still 8
continuing to use that, even in sensitivity cases.
9 I'd also like to point out that a recent 10 study, a very large study in the British Medical 11 Journal, the INWORKS study, is now suggesting not only 12 is there no apparent threshold but the use of a dose 13 and dose rate effectiveness reduction coefficient may 14 not be appropriate. It's not being seen in the data.
15 And that is pretty much uniformly tied in 16 the MACCS models that are used to estimate these 17 risks. So there may be something like a factor of two 18 already that's being underestimated.
19 And finally, given that the agency is 20 supposed to be taking a harder look at environmental 21 justice issues, and I've raised this in other venues 22 before, the use of these average risk coefficients 23 which make assumptions about the ratio of mortality, 24 cancer mortality to cancer incidents, are very much 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
153 dependent on the population that you're considering.
1 And there are disadvantaged groups where the potential 2
for a cancer mortality, given cancer incidents, could 3
be much greater than is assumed on the average.
4 And by not taking these factors into 5
account, the NRC is really becoming out of step with 6
the greater emphasis on environmental justice in 7
regulatory analysis for our federal government.
8 One other factor is cardiovascular risk 9
which is not being considered, but again, there is 10 emerging data. Another British Medical Journal study 11 is indicating that the cardiovascular mortality risk 12 from low level ionizing radiation may be on the same 13 order of magnitude as the cancer mortality risk.
14 So there are several factors that aren't 15 being accounted for. And given what's already been 16 pointed out, the safety margin, you don't know what 17 the safety margin is -- if you don't really know or 18 quantify the uncertainties, these safety margins are 19 meaningless unless you have better uncertainty 20 quantification, also taking into account these other 21 emerging factors.
22 Thank you, those are my comments.
23 CHAIR DIMITRIJEVIC: Thank you. Victoria?
24 MS. ANDERSON: Yes, Victoria Anderson, for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
154 the Nuclear Energy Institute. I wanted to reflect 1
some of the feedback we'd gotten from our members in 2
the nuclear industry. One thing that we notice is 3
that a lot of the insights from this project were 4
achieved without exercising the Level 3 portion of the 5
study. So in other words, they were produced during 6
the Level 1 and Level 2 PRA portions of the work.
7 This really illustrates to us that there 8
may not be any insight to be gained from devoting 9
resources to doing a Level 3 PRA for an operating 10 reactor at this time.
11 We also noted that the insights from the 12 study can necessarily be applied on a generic basis, 13 particularly the FLEX insights. And I appreciate that 14 the committee picked up on that. But I think it is 15 extremely important that, as they say, this is just 16 one plant, one study. And we can't necessarily draw 17 wide conclusions on it.
18 CHAIR DIMITRIJEVIC: Thank you. Thank you 19 very much. Any more comments from the members?
20 Okay. Well, actually this was all of the 21 technical prat. Alan, you still have a slide on our 22 future interactions, right, if I'm right.
23 MR. KURITZKY: Yes. Yes, we do.
24 CHAIR DIMITRIJEVIC: Yes, let's --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
155 MR. KURITZKY: So I guess first off, 1
there's a
number of meetings that we, both 2
subcommittee and full committee meetings that we 3
discussed previously about adding, with the whole 4
project being now condensed into essentially one more 5
year.
6 So it's going to get tough to do all these 7
meetings. So we may have to do some further co-8 actions. You know, we don't have to come up with a 9
decision now, but it might be that we want to take 10 Volumes 5 -- just like they were going to do with the 11 full committee, combine Volumes 5, 6, and 7 all 12 together and then have Volume 8 and 1 together, just 13 to reduce the number of meetings. But that we can 14 interact and discuss later, though I'm open to any 15 comments you have.
16 But the one thing I do want to get out 17 before we run out of time is for next week, or not 18 next week, but our November 1st full committee 19 meeting. You know, I think it's going to go for two 20 hours2.314815e-4 days <br />0.00556 hours <br />3.306878e-5 weeks <br />7.61e-6 months <br /> which is not that much different than this 21 meeting. Well, it's definitely shorter than this.
22 So did you want -- what would you be 23 looking for for that meeting? Do you want a 24 presentation similar to the one that I had for this 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
156 meeting? Are there things that we should leave out of 1
that meeting? I know you wanted --
2 CHAIR DIMITRIJEVIC: Yes.
3 MR. KURITZKY: -- a board discussion.
4 CHAIR DIMITRIJEVIC: I would sort of 5
concentrate on the, I mean, obviously some of the 6
summary slides, you know, the sky level summary 7
slides. But I would like more to concentrate on how 8
you're planning to satisfy your Goal Number 2, what is 9
the important insights from those.
10 Okay. So one of things is uncertainty 11 analysis which was one of our first comments which we 12 didn't touch in this in this meeting.
13 Another thing is the important insights 14 and maybe important sensitivity
- studies, more 15 concentrating on the things which will make to your 16 summary report from those volumes.
17 MR. KURITZKY: Okay. So we can definitely 18 talk about some of that. Much of that information we 19 will not have yet. Much of that information is not 20 going to be generated until we start doing the summary 21 of --
22 CHAIR DIMITRIJEVIC: Right. But you have 23 it through those reports. You have an important 24 insight, you have things which have not been done, you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
157 have things which require future work. I notice these 1
things all, you know, sprinkled, as I said, through 2
the report.
3 MR. KURITZKY: Right.
4 CHAIR DIMITRIJEVIC: So if you can sort 5
of, like, from those sections -- I am not interested 6
in errors made in the things like that, but some 7
things which, in your opinion, require the future 8
work, which are important insights, and some of the 9
important conclusions from dose alternative around 10 what your sensitivity runs.
11 MR. KURITZKY: Okay.
12 CHAIR DIMITRIJEVIC: And then uncertainty, 13 you know, even in this uncertainty, so not finish in 14 one direction and removing.
15 So, you know, the thing is maybe this will 16 require some effort. And I'm sorry about that, but 17 definitely will be helpful for you also when planning 18 your summary report.
19 MR. KURITZKY: No, it definitely will.
20 CHAIR DIMITRIJEVIC: Yes.
21 MR. KURITZKY: The timeframe --
22 CHAIR DIMITRIJEVIC: So something that 23 would not be, you know, worked on in vain, you know, 24 something which would be useful for you too in that.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
158 And also when you are writing the shutdown 1
spent fuel pool you will see we can see what type of 2
insights and conclusions are coming from here. And 3
that will be, you know, maybe we can discuss that next 4
week also. And I may send some email to --
5 MR. KURITZKY: Hossein?
6 CHAIR DIMITRIJEVIC: Hossein. So I 7
already sent one, but it was too late for this. I 8
only sent this after I saw the slides, because in this 9
moment discussing that difference between, you know, 10 base model and FLEX, you know, we saw some summary 11 results. But now I'm more interested in how are we 12 going to meet the Goal Number 2, you know, to expect 13 new insights and enhance the general knowledge of a 14 PRA. Okay.
15 MR. KURITZKY: Okay. So thank you for 16 that. That'll help us prepare for next weeks meeting.
17 Again, with the limited time involved, we will --
18 CHAIR DIMITRIJEVIC: Right.
19 MR. KURITZKY: -- start to dig up some 20 that information.
21 CHAIR DIMITRIJEVIC: I'm sorry about that.
22 And if you don't have a time to do this too much, 23 that's all right. I mean, we will plan, you know, to 24 write the letter saying anything further, how much you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
159 can do good, so will be interesting to have a 1
discussion about it.
2 MR.
KURITZKY:
Okay.
- Yes, we'll 3
definitely try, we'll have that in. And as far as the 4
existing presentation, do you want me to still include 5
the project status information for the full committee, 6
you know, at the beginning at the presentation?
7 CHAIR DIMITRIJEVIC: No, no, no.
8 MR. KURITZKY: Okay, don't need that.
9 CHAIR DIMITRIJEVIC: We have distributed 10 it.
11 MR. KURITZKY: Okay. And then for the 12 other stuff just kind of, just cut it down to just 13 have the high level insights, right, that was -- for 14 all the --
15 CHAIR DIMITRIJEVIC: Yes, high level 16 insights, yes.
17 MR. KURITZKY: Okay. And that's all I 18 had. If there's anything else that the subcommittee 19 wants to mention --
20 MEMBER REMPE: Vesna, just to make it 21 clear, since we are writing the letter on Volumes 3 22 and 4, he should include these higher level -- I 23 support what you're saying big time about the higher 24 level insights, but he should think about that for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
160 Volume 3 and 4, right?
1 CHAIR DIMITRIJEVIC: Well, I just connect 2
it to this work which is the, you know, Level 1, Level 3
2 for internal events and hazards.
4 MEMBER REMPE: Yes.
5 CHAIR DIMITRIJEVIC: Just for the Level 3 6
and Level 4, I mean the Volume 3 and Volume 4.
7 MR. KURITZKY: Yes. So, Dr. Rempe, yes, 8
the answer is yes to that question, yes.
9 MEMBER REMPE: That's what I heard Vesna 10 say, but I just wanted to make sure we're all on the 11 same page. Thank you, Alan.
12 CHAIR DIMITRIJEVIC: I just thought that 13 would also help thinking for the next volumes. You 14 know, when you think of documentation and organization 15 you will see how easy it will be to extract this now 16 for this one. And you say all right, maybe we can do 17 the better job in Volume 5.
18 All right. Well, thank you so much. For 19 me that was very enjoyable. And I really appreciate 20 your incredible work in this presentation. So thank 21 you very much.
22 MR. KURITZKY: Our pleasure, and thank you 23 very much. Thanks to the subcommittee for all your 24 feedback.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
161 CHAIR DIMITRIJEVIC: Okay. All right, 1
guys. So enjoy your afternoon, all right.
2 MR. KURITZKY: Take care.
3 MEMBER REMPE: So members, could I ask you 4
to stay on the line for just a minute for a discussion 5
about this afternoon. But I'll be careful to make 6
sure that someone else is listening that we don't say 7
anything we shouldn't. But it shouldn't take more 8
than a couple minutes. I just want to make sure we're 9
all on the same page of what we're going to do this 10 afternoon, okay.
11 CHAIR DIMITRIJEVIC: All right.
12 MEMBER REMPE: So as you probably know, we 13 have two retreat items we're going to be discussing.
14 And I don't want to say what they are. But there's 15 one that Alicia sent out that started actually at 16 noon.
17 And I guess I'd like to make sure members 18 are okay. Do you want to take a break for an hour and 19 come back at 1:00 and we'll start with the new member, 20 the solicitation discussion? Or do you want to --
21 MR. MOORE: Joy, can I interject for a 22 second, please?
23 MEMBER REMPE: Yes.
24 MR. MOORE: There are people online that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
162 are not members, or NRC, or ACRS employees. So if 1
you're still online, please log off.
2 MEMBER REMPE: And the court reporter, of 3
course, we're done for the day for you.
4 MR. MOORE: Thank you, court reporter.
5 CHAIR DIMITRIJEVIC: Thank you.
6 (Whereupon, the above-entitled matter went 7
off the record at 12:05 p.m.)
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
Level 3 PRA Project Overview for Internal Fires, Seismic Events, and High Winds Advisory Committee on Reactor Safeguards Reliability and PRA Subcommittee October 19, 2023 Alan Kuritzky Division of Risk Analysis Office of Nuclear Regulatory Research (301-415-1552, Alan.Kuritzky@nrc.gov)
Outline Level 3 PRA project (L3PRA project) status L3PRA project public reports L3PRA project overview report for reactor, at-power, internal fires, seismic events, and high winds Future interactions 2
Acknowledgements NSIR, NRR, NMSS, Regions, TTC National Laboratories (INL, SNL, PNNL, BNL)
Commercial Contractors (ERI, ARA, IESS)
PWR Owners Group Westinghouse EPRI ACRS 3
Caveat The L3PRA project adheres to the state-of-practice for most technical aspects; however, due to limitations in time, resources, or plant information, some aspects of the study were subjected to simplifications or were not fully addressed.
As such, inclusion of approaches in the L3PRA project documentation does not necessarily imply endorsement of these approaches for regulatory purposes.
4
5 Project Status Overview
6 Generic Process for PRA Model Development Develop Documentation Develop Initial PRA model Perform Internal Reviews Revise PRA Model Perform PWROG-led Peer Review Revise PRA Model Finalize Model Staff Self-Assessment Internal Technical Review Management Review Perform TAG Review Brief ACRS Re-Perform Internal Reviews Phase 1 Phase 2
7 Level 3 PRA Project Status Level 1 Level 2 Level 3 2020-FLEX*
Reactor, at-power, internal events Complete Complete Complete Complete Reactor, at-power, internal floods Complete Reactor, at-power, internal fires Complete Complete Complete Complete Reactor, at-power, seismic events Complete Complete Complete Complete Reactor, at-power, high winds Complete Complete Complete Complete Reactor, at-power, other hazards Complete N/A N/A Reactor, LPSD, internal events Complete Phase 2 L3PRA management review Phase 2 Revised model/documentation Phase 1 Initial model/documentation Spent fuel pool (all hazards)
Phase 2 L3PRA management review Phase 1 Initial model/documentation N/A Dry cask storage (all hazards)
Phase 2 L3PRA management review N/A Integrated site risk (all hazards)
Phase 1 Initial model/documentation N/A
- Not part of original project scope.
8 Level 3 PRA Project Public Reports Summary (Vol. 1)
Back-ground (Vol. 2)
Reactor, At-Power, Internal Events and Internal Floods (Volume 3)
Overview-IE/IF (3)
L1-IE (3a)
L1-IF (3b)
L2-IE/IF (3c)
L3-IE/IF (3d)
Reactor, At-Power, Internal Fires and External Events (Volume 4)
Overview-F/S/W (4)
L1-FIRE (4a)
L1-SEIS (4b)
L1-HW/OH (4c)
L2-F/S/W (4d)
L3-F/S/W (4e)
Reactor, LPSD, Internal Events (Volume 5)
Overview-LPSD (5)
L1-IE (5a)
L2-IE (5b)
L3-IE (5c)
Spent Fuel Pool (Volume 6)
Dry Cask Storage (Volume 7)
Integrated Site Risk (Volume 8)
Overview-SFP (6)
L1/L2 (6a)
L3 (6b)
L1-L3 (7)
L1-L3 (8)
9 Level 3 PRA Project Public Reports Summary (Vol. 1)
Back-ground (Vol. 2)
Reactor, At-Power, Internal Events and Internal Floods (Volume 3)
Overview-IE/IF (3)
L1-IE (3a)
L1-IF (3b)
L2-IE/IF (3c)
L3-IE/IF (3d)
Reactor, At-Power, Internal Fires and External Events (Volume 4)
Overview-F/S/W (4)
L1-FIRE (4a)
L1-SEIS (4b)
L1-HW/OH (4c)
L2-F/S/W (4d)
L3-F/S/W (4e)
Reactor, LPSD, Internal Events (Volume 5)
Overview-LPSD (5)
L1-IE (5a)
L2-IE (5b)
L3-IE (5c)
Spent Fuel Pool (Volume 6)
Dry Cask Storage (Volume 7)
Integrated Site Risk (Volume 8)
Overview-SFP (6)
L1/L2 (6a)
L3 (6b)
L1-L3 (7)
L1-L3 (8)
10 Level 3 PRA Project Public Reports Summary (Vol. 1)
Back-ground (Vol. 2)
Reactor, At-Power, Internal Events and Internal Floods (Volume 3)
Overview-IE/IF (3)
L1-IE (3a)
L1-IF (3b)
L2-IE/IF (3c)
L3-IE/IF (3d)
Reactor, At-Power, Internal Fires and External Events (Volume 4)
Overview-F/S/W (4)
L1-FIRE (4a)
L1-SEIS (4b)
L1-HW/OH (4c)
L2-F/S/W (4d)
L3-F/S/W (4e)
Reactor, LPSD, Internal Events (Volume 5)
Overview-LPSD (5)
L1-IE (5a)
L2-IE (5b)
L3-IE (5c)
Spent Fuel Pool (Volume 6)
Dry Cask Storage (Volume 7)
Integrated Site Risk (Volume 8)
Overview-SFP (6)
L1/L2 (6a)
L3 (6b)
L1-L3 (7)
L1-L3 (8)
11 Level 3 PRA Project Public Reports Summary (Vol. 1)
Back-ground (Vol. 2)
Reactor, At-Power, Internal Events and Internal Floods (Volume 3)
Overview-IE/IF (3)
L1-IE (3a)
L1-IF (3b)
L2-IE/IF (3c)
L3-IE/IF (3d)
Reactor, At-Power, Internal Fires and External Events (Volume 4)
Overview-F/S/W (4)
L1-FIRE (4a)
L1-SEIS (4b)
L1-HW/OH (4c)
L2-F/S/W (4d)
L3-F/S/W (4e)
Reactor, LPSD, Internal Events (Volume 5)
Overview-LPSD (5)
L1-IE (5a)
L2-IE (5b)
L3-IE (5c)
Spent Fuel Pool (Volume 6)
Dry Cask Storage (Volume 7)
Integrated Site Risk (Volume 8)
Overview-SFP (6)
L1/L2 (6a)
L3 (6b)
L1-L3 (7)
L1-L3 (8)
Preliminary Schedule for Releasing Draft L3PRA Reports for Public Comment Reactor, at-power, internal events and internal floods (Vol. 2 and Vols. 3x, 3a-3d) (4/22/2022) - comments resolved and currently with ADM for final publication Reactor, at-power, internal fires, seismic events, and high winds (Vols. 4x, 4a-4e) (8/18/2023)
Reactor, low-power and shutdown, internal events (Vols. 5x, 5a-5c)
(Q2-2024)
Spent fuel pool, all hazards (Vols. 6x, 6a-b) (Q3-2024)
Dry cask storage, all hazards (Vol. 7) (Q1-2024)
Integrated site risk (Vol. 8) (Q4-2024)
Summary report (Vol. 1) (Q4-2024) 12
13 Overview of Reactor, At-Power PRA Results for Internal Fires, Seismic Events, and High Winds
Base case model (Circa-2012 case) reflects plant as designed and operated in 2012 2020-FLEX case includes:
New RCP seals (shutdown seals)
FLEX strategies and equipment for responding to an extended loss of AC power (ELAP)
Credit for continued turbine-driven auxiliary feedwater (TDAFW) pump operation given a complete loss of all installed AC and DC power 14 2020-FLEX Case
FLEX strategies for coping with the plant conditions that result from an ELAP event involve a three-phase approach:
Phase 1 - Initially cope by relying on installed plant equipment and on-site resources Phase 2 - Transition from installed plant equipment to on-site FLEX equipment Phase 3 - Obtain additional capability and redundancy from off-site equipment and resources until power, water, and coolant injection systems are restored or commissioned 2020-FLEX case only considers Phases 1 and 2 15 FLEX Strategies
16 2020-FLEX Case Results Level 1 PRA (1 of 4)
CDF by Hazard Category Hazard Category Circa-2012 CDF
(/rcy) 2020-FLEX CDF
(/rcy)
CDF Reduction Internal events and floods 6.47E-05 2.67E-05 59%
Internal fires 6.14E-05 5.34E-05 13%
Seismic events 1.08E-05 8.49E-06 21%
High winds 1.38E-05 4.85E-06 65%
Total 1.51E-04 9.34E-05 38%
17 2020-FLEX Case Results Level 1 PRA (2 of 4)
Internal Events and Floods (42.9%)
Internal Fires (40.7%)
Seismic Events (7.2%)
High Winds (9.2%)
CDF % CIRCA-2012 CASE (1.51E-04/RCY)
Internal Events and Floods (28.6%)
Internal Fires (57.1%)
Seismic Events (9.1%)
High Winds (5.2%)
CDF% 2020-FLEX CASE (9.34E-05/RCY)
18 2020-FLEX Case Results Level 1 PRA (3 of 4)
Basic Event Name Failure Probability Internal Events Internal Fires Seismic Events High Wind Events F
1-FLEX-FAILS 0.30 0.7 0.7 0.5 S
1-RCS-SDS-FC 0.01 0.01 0.01 0.01 T
1-AFW-SBO-NO-FLEX-FA 0.30 0.715 0.715 0.5 Combined FLEX failure probability (p = F*T) 0.09 0.5 0.5 0.25 FLEX Failure Probabilities by Hazard Category
19 2020-FLEX Case Results Level 1 PRA (4 of 4)
20 Level 2 PRA Results (All Hazards Combined) (1 of 2)
Circa-2012 Case 2020-FLEX Case Risk Metric Reduction LERF 1.9E-06/rcy 1.3E-06/rcy 29%
LRF 1.1E-04/rcy 6.7E-05/rcy 38%
CCFP 0.680 0.764 N/A
21 Level 2 PRA Results (All Hazards Combined) (2 of 2)
Level 2 PRA Surrogate Risk Metric Time at which airborne radiological releases are terminated 7 days after event initiation SAMG entry +
60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> SAMG entry +
36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> LERF 1.9E-06/rcy 1.9E-06/rcy 1.9E-06/rcy LRF 1.1E-04/rcy 3.5E-05/rcy 3.5E-05/rcy CCFP 0.680 0.620 0.235 Level 2 PRA Surrogate Risk Metric Time at which airborne radiological releases are terminated 7 days after event initiation SAMG entry +
60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> SAMG entry +
36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> LERF 1.3E-06/rcy 1.3E-06/rcy 1.3E-06/rcy LRF 6.7E-05/rcy 2.6E-05/rcy 2.6E-05/rcy CCFP 0.764 0.679 0.309 Circa-2012 Case 2020-FLEX Case
22 Level 3 PRA Results (1 of 3)
IEIF: internal events and floods F:
internal fires S:
seismic events W:
high winds ALL: all hazards combined
23 Level 3 PRA Results (2 of 3)
IEIF: internal events and floods F:
internal fires S:
seismic events W:
high winds ALL: all hazards combined
24 Level 3 PRA Results (3 of 3)
Accident truncation - airborne radiological release termination time reduced from 7 days after accident initiation to 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after SAMG entry
Dose truncation - changed from linear no-threshold (LNT) to model based on Health Physics Society position paper, Radiation Risk in Perspective: Position Statement of the Health Physics Society (PS010-2), 2010
25 Summary of Results (All Hazards Combined)
Overall, the results show that the combination of this plant design and site location has substantial margin to the QHOs associated with the NRCs safety goal policy (51 FR 28044),
when considering all hazards combined, though the margins are noticeably less for the surrogate risk metrics of CDF and LERF that were endorsed by the Commission when it approved the issuance of Regulatory Guide 1.174 (SRM-SECY-98-015).
Risk Metric (per reactor-year)
QHO or Subsidiary Risk Metric Circa-2012 Case 2020-FLEX Case Risk Metric Reduction Core damage frequency 1E-04 1.5E-04 9.3E-05 38%
Large early release frequency 1E-05 1.9E-06 1.3E-06 29%
Large release frequency N/A 1.1E-04 6.7E-05 38%
Individual early fatality risk 5E-7 7.5E-13 6.6E-13 12%
Individual latent cancer fatality risk 2E-6 6.5E-08 4.0E-08 37%
26 Future Interactions
Subcommittee meetings Reactor, LPSD, internal events (Vol 5) - TBD Spent fuel pool (Vol. 6) and dry cask storage (Vol. 7) - TBD Integrated site risk (Vol. 8) and summary report (Vol. 1) - TBD Full Committee meetings Reactor, at-power, all hazards (Vols. 3 and 4)
- Nov. 1, 2023 Reactor, LPSD, internal events (Vol. 5), spent fuel pool (Vol. 6),
and dry cask storage (Vol. 7) - TBD Integrated site risk (Vol. 8) and summary report (Vol. 1) - TBD 27 Future Interactions
Acronyms and Definitions (1 of 2) 28 AC alternating current ACRS Advisory Committee on Reactor Safeguards ARA Applied Research Associates BNL Brookhaven National Laboratory CCFP conditional containment failure probability CDF core damage frequency DC direct current ELAP extended loss of AC power EPRI Electric Power Research Institute ERI Energy Research, Inc.
IESS Innovative Engineering & Safety Solutions, LLC INL Idaho National Laboratory L3PRA Level 3 PRA (project)
LERF large early release frequency LNT linear no-threshold LPSD low power and shutdown LRF large release frequency PNNL Pacific Northwest National Laboratories PRA probabilistic risk assessment
Acronyms and Definitions (2 of 2) 29 PWR pressurized-water reactor PWROG PWR Owners Group QHO quantitative health objective RCP reactor coolant pump RCY reactor-critical-year SAMG severe accident management guideline SNL Sandia National Laboratories TAG Technical Advisory Group TDAFW turbine-driven auxiliary feedwater