ML20235P082
| ML20235P082 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 07/08/1987 |
| From: | INTERNATIONAL TECHNICAL SERVICES, INC. |
| To: | NRC |
| Shared Package | |
| ML20235P067 | List: |
| References | |
| CON-FIN-D-1350 NUDOCS 8707200334 | |
| Download: ML20235P082 (16) | |
Text
. - - - - -
a Tedmical Evaluation of BG&E 'Itpical Report:
REIFAN Oanputer N Reactor Transient Analysis Model Qualification erry emw-w e arans:rmep a m ausca w m a r -
%7 International Tedmimi Services, Inc.
420 Iexington Avenue New Yodt, New Yodt 10170 i
j 8707200334 D70708' CF ADOCK O'5000317
?
CF ;3
i i
'IABIE OF COtTTENTS Page 1.0 sumary............................
1 2.0 Introduction...........................
2 3.0 Topical Objectives.......................
2
]
i 4.0 Ocmputer Modeling.......................
2 4.1 REIPAN02/)DD03.
2 4.2 Medal kations.....
3 4.3 Models..........................
4 5.0 Plant Test Ocmparison.....................
5 5.1 Multiple Secondary Side Malfunction Event.........
5 5.2 Four Punp Coastdown frcm 20% Power............
6 5.3 Reactor Coolant Operation Pump Combination Flow Tests at Hot Zero Power.............
7 i
5.4 One Pump Ocastdown frun 80% INxer.............
7 5.5 Total Ims of Flow / Natural Circulation 'Ibst 4
frcza 40% Power....................
8 l
6.0 TRAC Analysis Comparison....................
9
]
6.1 Cooldown to RHR Entry Using AINa and AIS.
9 6.2 Runaway Main Feedwater to One Steam Generator...... 10 7.0 IDFT Test Comparison...................... 10 1
7.1 Ifa-1 'Ibst......................... 11 7.2 IE>-3 Test......................... 11 8.0 Conclusions and Re m =rilations................ 13 9.0 Refererces........................... 14 i
i
)
l
)
1 I
1.0 Sumrv l
'Ihe Baltimore Gas and Electric Company (BG&E) submitted a topical I
report [1] for the purpose of &ventation of the BG&E best-estimate REIRAN l
thermal-hydraulic analysis capability.
BG&E presented camparison of five plant transients: these are (1) Mitltiple Secondary Side Malfunction Event, (2) Four Pump Coastdwn fran 20% Power, (3) Reactor Coolant Operation Pump Combination Flw Tests at Hot Zero Power, (4) One Pump Ocastdown frun 80%
1 Power ard (5) 'Ibtal loss of Flw/ Natural Circulation Test from 40% Power).
BG&E also preserited two TRAC analyses; (6) Coc,ldown to RHH Entry Usiry ADVs and (7) APS and Runaway Main Feedwater to One Steam Generator.
Two IOF1'
)
tests simulated for comparison with REIRAN calculations are 16-1 and 16-3.
1 i
It is our cjeneral conclusion that the applicant has presented thorough j
analyses, considering sensitivities to nodalizations, models and physical I
parameters.
We feel that they have exhibited a good understardirg of the l
l l
REIRAN code and the manner in which its applications to simulate plant transients is conducted, ard demonstrated that the staff inenbers who I
performed these analyses have adequate skills to analyze the results of such applications. Although their use of the control systems was not examined in detail ard nom of the transients analyzed tested their ability to use the control systems
- fully, their skill ard urderstandi2g of the code I
demonstrated in the discussion of the results in the report gives us reasonable assurances that the M&E staff also pem the ability.to model the control systems with the code, i
l l
1 1
u
)
t
,.Y' i
n.
1 J
2.0 Introdtetion -
l l
The Baltinc"a Gas and Electric Ocupany (BG&E) satanitted a topical report [1,2,3] for the purpose of hn=ntation of the BG&E best-estimate i
REIRAN thermal-hydraulic analysis capability.
BG&E presented canparison of five plant transients, two TRAC analyses and two IOFT tests with REIRAN calculations.
We have reviewed the applicant's efforts to qualify RErRAN nodels for use in the analysis of the Cahert Cliffs plant.
Cur evaluation is sommarized-in this report.
l l
3.0 Taoical Report Obiectives It was M&E's stated intent to deudokate (i) the capability of BG&E analysts to properly develop RETRAN models of the Calvert Cliffs plant, a Ctanbustion Engineering desion,, (ii) to perform calculations with these models to simulate realistic plant transient resperse, and (iii) to ocupare the results to measured plant data, another best estimate ocuputer code (TRAC), and experimental data (IDET).
In addition, model enhara.am:ad. and appropriate sensitivity studies were perforne.d by BG&E to provide grea&er insight arxl a better understanding of modeling Calvert Cliffs with RElRAN.
One of M&E's objectives in documenting their analyses and results in this topical' report was to present their effort in performing a code verification in accordance with NRC Generic Letter No.83-ll_ [4]
I' l
4.04 Q3 ratter Modelinc l
l 1
4.1 RETRAN02/ MOD 03 j
i Taa applicant used REIRAN02/FDD03 to perform analyses presented j
l
_____.._________..j
1 j
i in their topical report.
'Ihis version, which has not been approved, contains corrections of code errors diso:vered durirg the review of R N IO2,40D02 [5] which were implemented at the request of the NRC.
However, in addition to error wauction, some nodel modifications were also l
made to this version [6).
The NRC staff has not yet reviewed the code to
{
assure that these model revisions were appropriate, accurate and properly implemented.
However, the use of this version by DG&E is acceptable in the context of this"ccpical report.
4.2 Nodabtation
!r&E developed three nodalizations for Calvert Cliffs: (1) a detailed ime-loop model for transients with nearly symmetric loop corditions; (2) a split-core two-loop model (two cold legs were lumped
[
together) for trarsients which result in asymmetric plant conditions; and (3) a "four-loop" model which in fact has only two loops, but whi<t nodels all four cold legs discretely to study flow patterns during flow tests or transients which can be impact.ed by ora or rare RC pumps on a coastdtm.
'Ihe applicant has selectively and appropriately used one of these nodalization scheme 3 for each transient depenitry upon the expected plant behavior and has, in addition, provided acceptable justifications for such selection, in com cases based upon sensitivity analysis.
'Ihe one-loop model was used in a four pump coastdown and a comparative study with TRAC for couldown to RHR entry using ADVs ard ApS.
An otvious advantage of this nodilization is the computer code speed obtained by a simplified ' raial, while an obvious disadvantage is lack of detail. BG&E der.onstr#wd an awareness of these trade-offs.
i Use of a split-core rodel is a state-of-the-art technique for an asymmetric plant behavior which can influence the core physics due to reactivity feedback from the asymmetric temperature distribution caused by different cold leg fluids ard the mixing of 'these fluids in the core. 'Ihis nodalization was used by B3&E for analysis of the multiple secondary side
)
3 4
1
malfunction event, a natural circulation test frm 40% power, ard a runaway main fc::dJ1ter to one steam generator event as part of a set of PIS-limiting j
system transient analyses.
)
J The four-loop model explicitly models each of the four cold legs, including two cold legs per loops, a unique problem to CE plants.
This rodel is used when individual cold leg /RCP transients are analyzed such l
as three Calvert Cliffs start-up tests; a four pxp coastdown frm 20%
- power, RC operational pump combination flow tests at hot zero power, and a one pump coastdown frm 80% power.
In all three of these plant roaels, the RETPRi non-equilibrium pressurizer was used in combination with a single-node representation of the pressurizer.
In order to overcome the inaccuracy intWM during the computation of a rapid insurge/outsurge transient, BG&E intends to use the
'IWperature Transport Delay Time Model (TIUT) available in RETRAN. BG&E has decinated that they are aware of the lilaitations of the TIUT model and the non-equilibrium model ard of their ranges of applicability.
The steam generator secondary side is modeled as a four volume recirculating steam generator with a best estimate recirculation ratio in the two-loop and four-loop plant nodalizations,
- 111e the one-loop ncdalization only uses a one volume steam generator.
Transients for which EG&E intends to use the one-volume model have been discussed thoughtfully and are listed in Table 1.
I l
In addition, BG&E obtained from Energy Incorporated and modified l
a IGT REIRAN-01 II)-5 deck to model the IDPr facility using RETRAN-02/ MOD 03 l
for simulation of two tests; IE>-1 and II>-3.
1 l
l l
l 4.3 Models l
IG&E has qualified a portion of'their IETRAN code modeling for l
l the Calvert Cliffs plant, and in addition, they have provided Mrownians l
4 1
l l
l
indicatirq their urrierstandiry of the details and limitations of mrtain of those models, including the non-equilibrium model, the bubble rise scriel, the TIUT model and they have Mcemera thoughtfully the impact of various steam generator nodalizations.
'Ihese discussions are important aspects of denanstrating their ability to use the code and analyze the results.
In addition, the transients analyzed required sufficient use of the RETRAN control systems modeliry that BG&E has adeqdately demonstrated its ability to use those input nodels to represent the plant's control systems.
TAB E 1.
B3&E Nodalization Selection RETRAN SG SEODNDAITI SIDE MODEL Ch.14 UPDATED FSAR EVEffr SINGLE NODE MULTIP E LODE RCSD X
IDF X
SR X
CEAD X
CEAW X
CEAE X
EL X
IOL X
IDW X
FW4 X
IDAC X
MSIB X
SGTR X
WIB X
5.0 Plant Transient Comparisons 5.1 Multiole Secondary Side Malfunction Event A reasonable agreemert. Was obtained between the RETPAN 5
I calculated results and data of the asynetric cooldown during the first 200 seconds of the event.
BG&E, by performnce of a sensitivity study, iMicated that use of the TIUT model r@cM the peak pressure from 2338 psia to 2310.7 psia which is closer to the measured 2306 psia, and in::reased the peak pressurizer level by about two inches closer to plant data. Timing of these occurmnces, however, wexe not affected by the use of this model.
Although, the results began to differ after roughly 200 seconds into the transient, BG&E recognized and dia'W the fact that this was due to the uncertainty in the stuck open TBV position.
A series of sensitivity studies was conducted to determined which parameters would affect the transient results.
In the supporting du-nt presented by the applicant [3), EME analysts demonstrated a thorough understarxiing of the transient by pmviding an explanation of each substantial change in slope of plant parameters and hw they impact each other.
'Ihe applicant should be encouraged to provide future subnittals in this depth.
5.2 Four Pumn Coastdown from 20% Power Measurements were made of the total RCS flow for the first 55 seconds after reactor trip to obtain RCS flw data during a four pump coastdown at 20% of full power.
REIPAN calculated flow was within roughly 3% of measured flow
]
for the. first 35 seconds of the transient where the flow measurement accuracy is 2%.
A closer agreenent was obtained with the four-loop model than with the single-loop model due to the finer ncdalization and more j
accurate representation of R3 flow paths and pressure losses.
M&E indicated an understanding of the accuracy of these models and the uncertainties in plant data.
l 6
3 i
l l
l J
5.3 Reactor Coolant Operation PttnD Combination Flow Tests at Hot Zero Power Fifteen various combinations of flw tests were conducted at hot stand-by to evaluate operating pump combination flw distribution.
The REIFAN four loop model was modified to include a cross flw junction between the two identical dwncmer volumes used for thic analysis, to permit such cross fim to occur when the coastdwn is ssynmetric.
B3&E also input an explicit locked rotor re/erse f1w pressure loss coefficient to improve mode _lirg.
Statistical analysis was performed with the unsured plant data and REIRAN camputed results were campared to ths; mean va2ue of the data. In twelve out of fourteen cases studied, the ocmputed results were within two standard deviations (95.5% of data) of meash data.
One of the remainirs tests had insufficient data points to be meaningful and in the other test, the caputed flw differed from the mean by 4.1%.
5.4 One Pumo Coastdown from 80% Power The objective of this test was to measure single RCP coastdown data and validate the lw RCS flw trip from 80% pwer with one RCP secured.
l The four-loop model was used in the analysis.
Model enhancements including downcomer cross flw and realistic RCP reverse i
pressure loss coefficient were used.
Other plants conditions were adjust'i l
1
\\
l to reflect corx11tions at 80% power.
l i
l l
Although the camputed pressurizer pressure and hot and cold leg i
temperatures originally did not agree well and the total RCS flw under-predicted plant data by 1 to 4 % over the 60 second period, these differences were attributed by B3&E to differences in TBV arri ADV openirg tines.
B3&E then used revised data, adjusted the TSV and ADV 1rput to the 7
l
1 code and obtained better agreement.
1 BG&E performed a sensitivity study in which RCP mament of j
inertia aM rated torque were varied.
It was found that by either l
increasiry pump moment of inertia by 10% or decreasiry pump torque by 10%,
temporal flow rate increased closer to measured data.
It was also fourd that addirq a downcamer cross flow path and realistic RCP reverse flow loss factor iesulted in higher RCS flw than the initial model after 18 seconds into the transient and canpared 1:cre closely with the data (79.1% versus 78%
for the base case and 80% for plant data). For the base case with downcomer l
cross-flow ard RCP reverse flow model coefficients, REIRAN urder-predicted Rs total flow (i.e., over predicts flow reduction in the shutdown loop) by 0,4% to 3.9%.
These parametric studies indicate BG&E's ability to seek out the causes of differences between plant data ard computational results.
2;Lal loss of Flow / Natural Circulation Test from 40% Power 5.5 t
l This test was conducted to determine the power-to-flow ratio
]
durirg natural circulation. Key parameters were recorded for only the first l
60 seconds of this transient although the transient lasted much lorger.
Therefore, calculational result camparison was made for this 60 second 4
l period.
For this simulation, the RUIRAN two-loop model was itedified to be l
initialized at the 40% power carditions.
1 Agreement between the plar.t data aM REIPAN results in RCS flow, pressurizer pressure and pressurizer level was good; however, hot and cold leg temperatures and therefore the secondary side pressures did not agree well.
The temperatures differed by about 5'F ard the pressure by mughly 25 psi.
(During this time period, the hot and leg temperatures decreased l
roughly 10'F ard 5*F, respectively and the net charge in the secondary pressure was approximately 75 psi.)
l The topical report dirme the camparison of the onset aM magnitude of the natural circulation flow well after 60 seconds (five 8
I mirutes), and BG&E irdicated that this Mwmecion was based upon the record of the test written by the operator rather than upon recorded measurements.
6.0 TPAC Analysis Comnarison
'Ihe Ics Alamos National laboratory (IANL) developed a finely detailed IRAC-PF1 nodalization of the Calvert Cliff plant including a three dimensional vessel with modelirg of lower plenum mixing pipes.
Both steam generators and the pressurizer were modeled using n11ti-volume nilti-node nodalizations.
The two TRAC analyses presented by BG&E were performed by IANL.
i For sin 11ation of two transients performed by IANL, it appears that l
B3&E's objective was to perform their own best-estimate analysis for each of
)
these transients, and not to sinulate the IANL computation of the transient j
1 by using exactly the same assumptions.
Thus our review of these two I
analyses was not fmW upon the degree of agreement or disagreement between the results obtained by the two codes, but was focused upon BG&E's understanding of their own results and the reasons for differenTs frun the TRAC results.
l 6.1 Cooldown to RHR Entry Usirn ADVs and APS The REIFAN one-loop model was used in the comparison of this transient with the results frcan usiry a finely detailed TRAC nodalization.
Tuo major differences between the initial BG&E calculations and those of IANL using these codes are: (1) the pressurizer emptied in the l
REIRAN analysis while TRAC did not prtdict pressurizer emptying; and (2)
TRAC predicted faster cooldown after 6000 seconds than REIPAN.
Reanalysis of this transient by B3&E after the initial submittal showed that difference (1) was due to different assumptions about the use of charging flow in the analysis ard the time at which chargirq was assumed to be restored for level 9
L_________
)
)
control.
With these changes, BG&E repceted that RETRAN did not predict the pressurizer emptying.
Difference (2) was attributed to the gecznetrical modeling of the AINs and, after BG&E altered its model, differences between the two calculations were said to have been " resolved".
6.2 Runaway Main Feedwater to One Steam Generator
)
'Hiis analysis was performed as part of the NRC sponsored pr==rized thermal shock project.
l
'Ibe FORV reached its setpoint and the pressurizer level rose to the top rtoghly 500 seconds earlier in the REIRAN analysis than in the TRAC t
I calculation becausa the TRAC model did not allow backup pressurizer heaters to reactivate after pressurizer liquid level returned above a propcu-M l
level setpoint. Furthermore, BG&E indicated that the reason RETRAN camputed j
that the pressurizer level rose to the top over 1000 seconds earlier than i
l l
TRAC was because of the large difference in HPSI flow rate which resulted frten the different pressures predicted.
'Ihe minimum downccaner temperature was canputed to be 350*F,by RETRAN, whereas TRAC camputed minim 2m average downocxner temperature of 400*F.
'Ibe hot leg temperatures also show a similar difference between RETRA:1 and TRAC.
Overall, the results exhibited similar harils despite large differences in the computer code models and the degree of detail and capabilities in plant nodalizations.
j 7.0 LOPP Test Carmarison B3&E modified an M-5 REIRAN-01 model supplied by Energy Incorporated l
to model M-1 and M-3.
'Ihese experiments are secorx3ary system initiated j
events: M-1 was a loss of steam load anticipated transient; and M-3.was an l
l excessive load increase anticipated transient.
10 l
___________._______________a
)
'j 7.1 LDFI' Test I6-1
'Ihe calculational results presented (steam generatn. dcme pressure, secondary side liquid level, pressurizer prersure and level and coolant temperature in the intact loop steam generator primary side inlet j
plenum) in the topical report did not agree with the test data.
After a.
sensitivity study with respect to the Interfacial Heat Transfer Coefficient (DfIC) was ps:: formed, BG&E concluded that REIRAN does not permit a varying DfIC to sire'llate this sort of transient involvire a rapid pressurizer insurge followed by an equally rapid outsurge. 'Ibe use of a large value of Df7C (27500 BIU/HR-FIi2_.F) properly calculated the insurge portion but not the outsurge portion because it caused excessive energy transfer fran the steam which resulted in a lower pressure, which in turn caused too much pressurizer liquid to flash during the outsurge.
Conwycelingly, the use 2
of a low value (10 BIU/HR-PT _.F) of DIIC predicted plant data more closely l
during the outsurge portion than the original calculation which used large DfIC; however, pressurizer pressure and level were not well computed during the insurge.
BG&E found that 400.0 BIU/HR-Fr2_.F gave a better overall fit to the data.
In this analysis, the BG&E RCIRAN primary and secondary pressure transient behavior did not follow the plant data precisely. We concur with IG&E's conclusion that this deviation is attributable to lack cf basic data regarding the details of Main Steam Flow control Valve operation during the transient, and that the absence of such data makes it virtually impossible
)
to obtain extsct correlation.
7.2 IDFT Test L6-3 l
l
[
After B3&E added the use of the TIDT option, the primary side parameters, e.g., pressurizer pressure and Ibt leg temperatures agreed well j
l between the test data and REIRAN results.
'Ihe corputed coolant l
11 l
I
-.. =
=_
tenperatare :in the ' intact loop steam generator prinary side inlet ~ plenum lagged the test data ly rux3hly 25 seconds.
This difference was attributad to uncertainty in the data.
The aanputed and test steam gemrator &re pressures began to diverge at about 50 seconds due-to the' earlier' REHRAN reactor trip and the resulting earlier closure of the MSIW, '1ha reason for the pressurizer level diverJence starting at abcut 50 seconds was' attributed to less energy being available to be deposited in the secondary due to the early scram and to a higher terminal steam flow and feedwater flow in the RETRMI calculation.
1 12 l
i
.________________9
~
8.0 Conclusions and Recommendations 1G&E staff have Cencastrated their ability to develop input models for best esthnate catputation of plant transients in the Calvert Cliff 9 Nuclear Power Plant, atx3 to perform the supyJrting SorSitiVity BbXIiCS to aid in detennitetion of the aIprtpriat0 nodel selection arti nasalization. IG&E has
{
developed three nxlalizatiorr Ltich it sur s to us; BG&E has presented ra discussions and jtutifications of their nadalizahluns or certain of the trarmdr_nts (as indicated in the foregoing text), and such T4alizations should therefore be consi& red qualified for future use.
It is rectanmendtvi that in ftiure licencirg applications, rodel selection be jastified or< r transient by transient kusis.
BG&B may, however, rely upon arri lefsence the discussions contained in this application when rakhrJ such rodel relection, D3&E staff have further dencastrated their ability to thoroughly analyze a transient.
It $s re)-- =ded that in future licensing sutnittals D3&E be requested to sulanit analyses of the det.fil and thercucfmess provided in their analysis of Multiple Secondary Malfunction Event.
i 4
1 13 l
n.
1 o.
'i 9.0 Referenogs 1
B%E Topical Report A-85-13, "REIRAN Otmputer Code Reactor System Transient Analyn.is Model qualification," January 31, 1986.
),-
2.
Ietter frun J.A. Tiernan (BG&E) to NPO, dated February 24, 1937, "RETRAN Rsview - SukL11ttal of Additional Infcraf tion".
3.
Ietter frun J.A. Tiernan (BGAE) te NRC, ' dated June 5, 2S87, "RETRAN Review 6thnittal' of Additional L1 formation".
4.
USNRC Generic letter No. 83-11, "Licer.sio Qualification for Performing Stfety Analyscs in Support of Licensiig Actions,"
February 8, 1983.
5.
Istter frt2n P.B. Abramson (Ahl,) to J. Carter (NRC), "Tedhnical Evaluation Repo:t: REIRAtF2/ MOD 02," May M,1983.
6.
Istter frun T.11. Schnatz (UGRA) tp C.O. 'Ihatas (NRC), "RIiTRAN - A Procfram for Trinsient 'Ihernal-Hydraulic Analysis of Otmplex Fluid, Systems," FePomry 4, 1985.
14 l
I i