ML13196A422
| ML13196A422 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 07/15/2013 |
| From: | Atomic Safety and Licensing Board Panel |
| To: | |
| SECY RAS | |
| References | |
| 50-443-LR, ASLBP 10-906-02-LR-BD01, RAS 24821 | |
| Download: ML13196A422 (4) | |
Text
NextEra Energy Seabrook, LLC (Seabrook Station, Unit 1)
License Renewal Application NRC Staff Answer to Motion for Summary Disposition of Contention 4B ATTACHMENT 4B-G
Operations THE NUCLEAR NEWS INTERVIEW Karen Vierow: Severe accident code analysis urdue University researchers Three computer programs used to simulate severe P are working on a project to compare the three leading se-accidents at nuclear plants are themselves vere accident programs, or codes, analyzed for comparison.
used by the nuclear power industry in the United States. and never has the relative state of modeling been pursued The codesMELCOR, MAAP4, and SCDAP/ this thoroughly.
RELAP5, all developed for different approaches and for Karen Vierow, an assistant professor in the School of Nu-different purposeshave been tested at Purdue using a clear Engineering at Purdue University, is leading the re-hypothetical accident scenario (station blackout with no search, which is being sponsored by the Nuclear Regulato-recovery of auxiliary feedwater) at a four-loop pressur- ry Commission. She has worked on the project with Yehong ized water reactor based on the now closed Zion nuclear Liao and Jennifer Johnson, graduate students at Purdue; power plant. Conservative analysis conditions were used Mark Kenton, a MAAP4 developer currently with Creare, to investigate the integrity of the Inc.; and Randy Gauntt, a MELCOR steam generator tubes and other developer from Sandia National Labo-components during the accidents ratories. The accident simulations have progression. Despite considerable been performed, and Vierow is now in differences in the codes themselves, the process of analyzing the data.
test results show that the codes are MELCOR was developed by Sandia similar in terms of thermal-hydraulic National Laboratories; MAAP4 (Mod-and core degradation response. ular Accident Analysis Program) by To date, plant data for an actual se- Fauske & Associates, Inc.; and vere accident at a nuclear power plant SCDAP/RELAP5 (Severe Core Dam-exists only from the Three Mile age Analysis Program/Reactor Excur-Island-2 incident in 1979. Code simu- sion and Leak Analysis Program) by lations for the TMI-2 scenario have the Idaho National Engineering and been carried out in the past with Environmental Laboratory.
SCDAP/RELAP, MELCOR, and Vierow talked with Rick Michal, MAAP4, but never have the results of Nuclear News Senior Editor, about the Vierow: The codes have undergone the three codes for the same hypothet- significant upgrades over the years and are code research work.
ical accident been compared in detail, becoming more best-estimate in nature. Interview begins on next page March 2005 N U C L E A R N E W S 23
INTERVIEW: VIEROW How did you get involved in this research?
About four years ago, I was doing some validation of the MELCOR code. I then started studying MAAP4 and SCDAP/
RELAP5 to look into the current state of the technology. I found that the three codes all have their own unique features and that they can learn from each other. That was the rea-son I extended my work to researching all three codes.
Who are the main users for each of the codes?
MELCOR and SCDAP/RELAP5 are used by regulatory agencies and research institu-tions to evaluate hypothetical severe accident events, such as a station blackout or the po-tential for a steam generator tube rupture.
MAAP4 is the severe accident code most widely used by nuclear utilities and vendors because of its short run time and reduced re-quirements for code expertise. The Electric Power Research Institute and many utilities also use it for the NRCs Significance Deter-mination Process and other analyses.
In addition, MAAP4 can be used by an The onset of hydrogen production (start of fuel damage) is predicted to occur at nearly existing plant to simulate how a proposed the same time by all three codes, and roughly the same amount of hydrogen is produced.
modification would affect plant operations.
And plant designers could use any of the runs at a moderately fast speed and has a I found that it was difficult to reconcile all codes to predict the performance of future large number of mechanistic models. In of the input for each of the codes. When I plants if a certain set of conditions were im- early applications, MELCORs spatial dis- say input, I mean the geometric description posed on those plants. cretization of a nuclear power plant con- of the plant, the initial conditions, the The codes have undergone significant up- sisted of roughly 10-30 control volumes, boundary conditions, etc. I tried to make grades over the years and are becoming and a large number of parametric calcula- them as consistent as possible for all three more best-estimate in nature. MELCOR tions could be run in a short time. With the codes, but sometimes it was easy to miss was originally intended to be a probabilis- more complicated calculations that are now something.
tic risk assessment tool; the initial objective being demanded of it, MELCORs run There are many modeling options and a for the MAAP4 code was to predict severe times have been increasing and are roughly lot of things that are compared. Its not just accidents, using simple models based on equivalent to those of typical SCDAP/ the same geometry, but also the same first principles; and SCDAP/RELAP5 be- RELAP5 calculations. physics models that must be analyzed. So, gan as a best-estimate code with physics- I have to run the codes many times to get based models. While still different, the ca- What kind of accident simulation did you the parameters to be comparable.
pabilities and applications of the three run for your research? As for computing power, Ive found that codes have been converging over the years. The information input into the three codes MAAP4 runs much faster than the other prior to the accident simulation was made two codes. With MELCOR and SCDAP/
Could you provide an overview of the as similar as possible, and consistent condi- RELAP5, it depends on what type of event codes? tions were placed on all of the analyses. Of is being run. Sometimes the test will go for All three codes are capable of modeling course, to arrive at a set of conditions that a few minutes, but sometimes it will take reactor coolant system response, core ma- the three codes could cover, various as- several hours. Its not in terms of months as terial chemical reactions, core heat-up, sumptions had to be made that rendered the it used to be for other codes. The increase in degradation, and relocation, heat structure calculation unrepresentative of plant behav- computer power definitely shortens our response, and other severe-accident phe- ior after some point in the calculation. Since simulation time.
nomena. the analysis scope of SCDAP/RELAP5 is Modeling of fission product release and limited to the failure of the primary side Why did you pick a four-loop PWR, based transport and containment phenomena are pressure boundary, the test focused on in- on the Zion plant, to simulate an accident?
integrated into the MELCOR and MAAP4 vessel severe accident phenomena, with a Zion was chosen because it is a represen-codes. SCDAP/RELAP5 is characterized special interest in the steam generator tube tative four-loop plant in the current PWR by its detailed, mechanistic models of response to thermal transients during this fleet. We are also evaluating the codes for severe-accident phenomena; however, the period. Its important to remember that our other plant types, including boiling water calculations can be rather time-consuming. objective was to compare the three codes, as reactors and Babcock & Wilcox OTSG SCDAP/RELAP5 typically uses on the or- opposed to validating them. (Once Through Steam Generator) PWRs.
der of hundreds of hydrodynamic compo- Since Zion was one of the plants used in the nents to model the primary system. How many simulations of the same accident famous NUREG-1150 assessment of severe MAAP4 calculations require minimal com- were performed? accident risks, input decks that we could putation time with simplified geometry I ran the analysis probably 20 times for start with were already available. A lot of models. MELCOR falls in between these MELCOR and SCDAP/RELAP5. Marc safety systems and normal features were two codes, being much closer to SCDAP/ Kenton performed the MAAP4 analysis disabled so that the core would melt and the RELAP5 in terms of nodal complexity. It because I do not have this code at Purdue. physics models in the codes would be tested 24 N U C L E A R N E W S March 2005
INTERVIEW: VIEROW through core degradation, relocation, and the developers of each have their own model here or there, or make an existing many other possible serious events. knowledge and their own way of thinking. model more detailed. In fact, I am working It was difficult to choose the right physics now to develop new physics models for the What are the results of the simulations? models for these analyses so that each code codes and modify current ones to improve The results show that the thermal- was given a fair chance to calculate the their prediction capabilities.
hydraulic phenomena and major in-vessel same event.
severe accident phenomena are in good Could the codes be used to test reactor de-agreement for the three codes. Also, the in- What were your conclusions at the end of signs that may be used in a future hydrogen tegral effect of diversified core models in the research? economy?
terms of total hydrogen production and to- The key conclusion was that each of the These three existing codes were designed tal core debris mass slumping into the reac- codes has high capabilities, but that some to analyze light-water reactors. In a hydro-gen economy, the reactor would be a high-TIMING OF KEY EVENTS temperature design that would be cooled by Event MELCOR (s) SCDAP/RELAP5 (s) MAAP4 (s) a gas or molten salt instead of water, so the Start of core uncovery 7 680 7 160 9 615 codes would have to be modified to be Core completely voided 11 620 9 950 14 500a valid. The geometry of the high-tempera-5% cladding oxidized 13 780 14 806 13 800 Slumping to lower head 16 189 16 130 21 994 ture plant is also different from that of an LWR. The high-temperature plant could be a
MAAP4 calculated a very slow rate of water level decrease at the bottom of the core, leading to a substantial delay in the voiding of the bottom node. This is, in part, due to continued slow draining of the pressurizer. (s) = second what is called a prismatic design, or a peb-ble bed design that has a lot of graphite in tor vessels lower head are consistent for have physics models that could be incorpo- it. That new code would need to include the three codes. This consistency will prob- rated into the others. So, while all the codes new models for the graphite behavior and ably reduce the codes prediction differ- are impressive, each can benefit from learn- the different geometry of the core. There ences for ex-vessel severe accident phe- ing from the other codeslooking at the key would not be the traditional vertical fuel rod nomena, such as ex-vessel corium water or assumptions made in the other codes and see- assemblies that LWRs have.
corium concrete reaction, hydrogen behav- ing where some of the assumptions are valid These codes are big computer programs, ior in the containment, and containment and where some models could be improved. and they have perhaps 400 000 lines of For-pressure response. There are also some dis- tran. They have a numerical architecture.
crepancies that could be termed as minor Could you elaborate on how and what the The basic equations are all set up, and its and that are possibly due to uncertainties in codes can learn from one another? not an easy task to make major modifica-the numerics and physics models. For example, SCDAP/RELAP5 has the tions or write a new code. Each code has a During the testing, several key assump- most detailed treatment of hydrogen pro- core set of writers that can make major tions were made to account for known dif- duction from oxidation. The hydrogen pro- modifications to it.
ferences in heat transfer modeling and the duction rate is a key indicator of the pro-representation of countercurrent natural cir- gression of a hypothetical severe accident. What kind of effort would be needed to de-culation of hot gases. Given these assump- Once the fuel is damaged, MELCOR has velop codes for Generation IV gas-cooled tions, the three codes predicted similar tem- good physics models to predict the fission reactors?
peratures in the various reactor coolant product release and transport phenomena. Some of this is included in my earlier system components. Future work could fo- These capabilities have been removed from statements on reactors for hydrogen pro-cus on resolving the modeling differences recent versions of SCDAP/RELAP5, and duction. Gas properties would need to be in these few key areas. another code performs these calculations confirmed and the codes would need to be for SCDAP/RELAP5 users. Perhaps the tested for their capability to run without Were there any uncertainties during the simulations? TIMING OF HEAT STRUCTURE FAILURES There is regarding core degradation, in Event MELCOR (s) SCDAP/RELAP5 (s) MAAP4 (s) its late phase, where things are relocating Onset of natural circulation 9 300 9 000 9 720 in the core. We have less detailed knowl- Failure of surge line 16 287 14 955 14 860 edge of how the core would relocate during Failure of hot leg piping on pressurizer loop 16 464 15 720 15 267 Failure of SG tubes on pressurizer loop 16 553 15 210 14 913 a severe accident. The industry has some good ideas, but there is minimal plant data on it because the only actual data comes MELCOR code will incorporate some of the light water coolant/moderator they from TMI-2, when the core partially the knowledge base from SCDAP/RELAP5 were developed for. Changing to a gas melted. Based on that and on experiments, for hydrogen models or use a more paramet- coolant should not be too much of a chal-physics models were developed that were ric approach to account for uncertainties in lenge because gas behavior is much eas-put into the codes. Thats where some of the hydrogen-related phenomena. MAAP4, on ier to predict than steam/water behavior uncertainties come fromnot having a the other hand, is able to complete calcula- with its phase changes between liquid and complete knowledge of the phenomena, tions in orders of magnitude less time than vapor and the complicated steam/water trying to use models based on smaller ex- the other codes due to simplified versions of property tables. In addition to modeling periments and applying the results to a full basic equations and fast-running models the different fuel geometry and incorpo-plant. based on first principles. The assumptions rating severe accident models for phe-made in MAAP4 to allow it to run so fast nomena peculiar to gas-cooled reactors, What was the most challenging task in com- should be further analyzed and applied as events not considered for LWRs must be paring the codes? appropriate into the other codes. modeled. These would include introduc-The hardest part was to compare the tion of air or water into the primary sys-physics models. Were trying to run all Are there any specific improvements you tem that could result in graphite burning.
three codes on an even playing field. There could talk about? We believe the codes are flexible enough are differences in the nature of the codes, There were no major deficiencies in the that they can be modified to model Gen and each has a different philosophy because codes. Each one could, of course, add a IV reactors.
26 N U C L E A R N E W S March 2005