ML20023B950
| ML20023B950 | |
| Person / Time | |
|---|---|
| Site: | 05000470 |
| Issue date: | 04/30/1983 |
| From: | ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY |
| To: | |
| Shared Package | |
| ML19301C195 | List: |
| References | |
| NUDOCS 8305090372 | |
| Download: ML20023B950 (39) | |
Text
{{#Wiki_filter:i ,I I l ENCLOSURE 1-NP TO LD-83-037 V. Response to NRC Questions on CESSAR-F Statistical Combination of Uncertainties
- In Thermal Margin Analysis for System 30 1
REACTOR DESIGN APRIL, 1983 COMBUSTIONENGINE_ERING,INC. WINDSOR, CONNNECTICUT
.~
+ s 8305090372 830503 PDR ADOCK 05000470 E PDR - __ _. _ __ _ . _ . . _ _ _ . . _ _ . - _ _ . _ - ~ . . _ _ _
LEGAL NOTICE This report was prepared as an account of work sponsored by Combustion Engineering, Inc. Neither Combustion Engineering nor any person acting on its behalf:
~
A. Makes any warranty or representation, express or implied including the warranties of fitness for a particular purpose or merchantability, with respect to the accuracy, completeness, or usefullness of the information contained in this report, or that the use of any information, apparatus, method, or process disclosed in this report may not infringe privately owned rights; or
~
B. Assumes any liabilities'with respect to the use of, or for damages resulting from the. use of, any information, apparatus, method or process dis [losed in this report. l l I l i l I l
OUESTION 1 It is incorrect to interpret a non-parametric tolerance limit as a mean value plus a constant times the standard deviation. In Section 2.3.2, the non-parametric "K " is calculated from equation 2-4 by using the determined one-sided tolerance limit and the known mean error. Provide justification for this
= approach to treat non-normal error distribution.
. RESPONSE The modeling uncertainty computed in Section 2.3.2 is combined with the.CECOR ~
(Fxy) uncertainty described in Section 2.4.1.2. A [. _. _ . i. .. [____ jis used to combine these two major uncertainty components. Some departure from normality can exist in both the modeling and CECOR uncertainty distributions. For this reason effective variances and associated K-factors are computed for use in the[ .
. j procedure. The confidence limits computed for the combined uncertainties in this manner have been verified to be more limiting thanthosecomputedusingamorerigorous[ ~ ] nethod. The verification method has been discussed in the response to Question 9.
~ --
6 l I l i l l 9
OUESTION 2 Provide justification for your evaluation of the DNB-0PM modeling error described in Section 2.3.3. ^ RESPONSE . The ratio of the predictions of two codes such as CETOP-D and CETOP-2 is ~~' ~ completely determined at a given point in state variable space. (,_ _ ,
~ ~~ ~
], ~ The[ ..i ~
~ ~ ~ ~--
.[~- _ _ __ _
j.. _
" * ~" ' "
- 7 .""F.'" M
* *1T - ee M
_4m.w *
= -m. eh#** +=w %*** w 4 # .h,,
_ . A.a m.
*.,eggg
----e=-e +- m. ,_g,g,m + - - -
m e
- . - . ,~.m .~. -- . ~~.. men .m . , . . . .
, , . . . e-e-,.~ ,
w -m-- -=
*""1'W**-.gg2. ,, ,, , , .g, _,,. 7 1
* "M m
6 b 9
QUESTION 3 Section 2.4.1.2 states that the Fxy used by CPC are verified by a CECOR calculation of Fxy during startup testing. Why wasn't the CECOR Fxy error and standard deviation evaluated for each time-in-life?
~
, RESPONSE The CECOR Fxy measurement error is mainly due to the hardware properties of the in-core detector instrument and was determined based on operational data from a 1arge sample of fuel cycles and reactors (5 reactors with a total of 11 cycles covering 170 time points with 30 to 40 instrumented fuel assemblies at each
; time point).
As justified in Ref. 3.1, the CECOR Fxy measurement error will remain applicable to all future reactors and cycles employing rhodium, self powered, fixed in-core detectors. Therefore, the CECOR Fxy measurement error is not evaluated for each time-in-life. Reference 3.1 Combustion Engineering, Inc., " INCA /CECOR Power Peaking Uncertainty," CENPD-153-P, Revision 1-P-A, May, 1980. i _ l i 5 l f
QUESTION 4 With regard to the penalty factors for the CPC power distribution algorithm (Section 2.4.1.3), provide a detailed description on how the sensitivity factors associated with RSF, TSF, SAM and BPPCC were obtained. In i particular, how many rod configurations were used, how many RSF (as well as TSF, SAM and BPPCC) values were used per configuration and how wasAR
, determined?
, RESPONSE The detailed calculational procedures to determine the penalty factors associated with RSF, TSF, SAM and BPPCC uncertainty are described below. The cases used in'this analysis and the bases for startup test acceptance band uncertainty are also given.
All values of' RSF, SAM, BPPCC and TSF used in the CPC data base are verified during startup testing (B0C conditions). It is assumed that these parameters will not be remeasured at any time later in the cycle. Therefore only the RSF values at BOC (nominal value) and RSF values representing the effer.fs of measurement error and depletion were used per CEA configuration. A Cmeastfrement variation in the nominal RSF conservatively bounds the effects of
. error and depletion. Thus, three values of RSF per CEA configuration were used in the calculation of the maximum sensitivity factor.
Approximately 700 cases with the CEA configurations listed in Table 4-1 were used at each time-in-life. .The SAM, BPPCC and TSF constants are independently i predicted / measured parameters based on unrodded core conditions. Therefore,
.-the number of CEA configurations used and_ number of SAM, BPPCC and TSF values used per CEA configuration is not applicable.
(1) Rod Shadowing Factor (RSF) - In the calculation of a penalty factor associated with the RSF uncertainty, the effects of depletion and feedback are considered. Figure 4-1 shows the calculational procedure for the penalty factors. Approximately twelve hundred cases (1200) of CPC LHR and DNB-OPM at each time-in-life (BOC, MOC, E0C conditions) are calculated with the nominal RSF values (R's) (Approximately 700 cases use the CEA configurations given in Table 4-1). By changing power distibutions with different core power levels, CEA configurations, power maneuvers, and xenon concentrations, these cases (1200) at each time-in-life were used to encompass steady state and quasi-steady state plant operating conditions throughout the cycle life-time in accordance with CPC design requirements. The CPC LHR and DNB-0PM for each case are compared with those of the reactor core simulator.
- , The error for LHR and DNB-0PM for case i is defined as
- <
l (Zo )1 = (Yg c)j __;y9F )i, i = 1,2,---N (4-1) l (Yhi o
where (Yo c)i = CPC LHR or DNB-0PM for the 1-th case F _ (Yo )1
= Reactor core simulator LHR or DNB-0PM for the i-th case N = Total number of cases (approximately 1200)
Next, the RSF for a given rod configuration presented in Table 4-1 is changed from the nominal value (R +/1R) and the same twelve hundred cases
- of CPC LHR and DNB-0PM are again calculated. The errors for the CPC LHR and DNB-0PM with the changed RSF are calculated by comparison to the reactor core simulator results (LHR, DNB-0PM).
~
This procedure is repeated for all CEA configurations (i=1,2,---6) given in Table 4-1 and the sensitivity used for penalty factor calculations due to RSF uncertainty is then obtained from: i i {^
~ . , .--
lfor all CEA configurations, the - By employing [ penalty factor due to RSF uncertainty is obtained as follows:
- ~
The RSF startup test band uncertainty [~ ~ ~ ~]was cbnservatively chosen based on the maximum error between predTcted (base) and measured RSF values from previous startup test power ascension test results. (2) Temperature Shadowing Factor (TSF)
. The temperature sensitivity of the ex-core detector signals is required to correct the CPC neutron flux power calculation. This correction is based on the temperature correction factor, Tcorf, as used below:
P core =P det *T copf
._. -. . - _ = _ . . .
where Tcorf = 1 + CT (T co -Tin) P core = core power inferred directly from the ex-core detectors, 2 C T
= temperature shadowing factor (TSF) which is determined to give the best linear fit to the detector sensitivity data T eo = nominal core inlet temperature, and
~
= core inlet temperature.
Tin L The temperature calculated ex-core changes detector (TSF) response for System 80 NSSS sensitivity is Cr = ~ [to inlet 3 Thecoolant
^
calculated (i .e. ,6C7=maximum TSF uncertainty for System 80 isT ACT ~due to' the
]) . In the determination of a penalty factor ~~
4 TSF uncertain [ty, an uncertainty ofACT=[ ~~ was conservatively applied to the CPC LHR andlNB-OPM calculations. It is assumed that the maximum deviation of inlet coolant temperature from the nominal inlet temperature which can occur between CPC core power ~ calibrations the temperature duringcorrection steady state operation factor (6 conf) is[~islculated ca from:The uncertainty on ATeopf = ACT * (T eo -Ty=[i_ j l_ T [_]
. lis applied to the LHR and Therefore, DNB-0PM to compensatea[ ' penaltyforf. the actorunc[ iFtainty in'the ex-core detector response sensitivity to inlet ' coolant temperature changes. The predicted value of CT is verified during power ascension testing. ,
~
(3) Shape Annealing Matrix (SAM) - The CPC Shape Annealing Matrix (SAM) used in the power synthesis algorithm is verified during power ascension testing. The predicted SAM elements are calculated by simulating a free unrodded xenon oscillation similar to SAM startup test measurement procedure. The predicted SAM elements are then determined from a regression analysis of the ex-core signals and the corresponding bottom, middle and top third integrals of the core peripheral power. In the calculation of a penalty factor associated with the SAM uncertainty, ex-core /in-core detector measurement errors, depletion, and shape annealing error effects are considered. An ex-core detector measurement error is randomly selected from a standard normal error distribution and applied to the ex-core detector signal. The in-core detector signal measurement is randomly selected from a standard normal error distribution and applied to the core average peripheral power distribution. The one-third core peripheral power integrals are obtained by integrating the core average peripheral power distribution. By using a least square fitting technique k
with the ex-core detector signals and the corresponding one-third core peripheral powers, the SAM elements with detector measurement error effects are determined. Depletion and shape annealing error effects are considered by simulating a
-- xenon oscillation at B0C, M0C and E0C conditiDns and using different shape annealing functions. Through regression analysis, SAM elements are calculated for different core burnup conditions and for variations in shape annealing.
. Figure 4-1 shows the calculational procedures for determining penalty factors due to SAM uncertahlies. Using SAM values with and without detector measurement error. depletion and shape annealing error effects, approximately twelve hundred (1200) cases at each time in life were run to calculate values of CPC LHR and DNB-0PM. .The cases used in this analysis
, include changes in power distribution due to changes in burnup, core power, CEA configuration, load maneuvers and xenon concentration. The CPC LHR and DNB-0PM values are compared to those calculated by the reactor core
- simulator and the LHR and DNB-0PM error is calculated for each case. By analyzing the difference in these error distributions resulting from the
. use of SAM values with and without detector measurement error, depletion and shape annealing error effects for each case, a LHR and DNB-OPM penalty i
factors due to the SAM uncertainty are calculated based on the most non-conservative error difference at a 95/95 probability / confidence level. (4) Boundary Point Power Correlation Coefficients (BPPCC) ! The CPC Boundary Point Power Correlabon Coefficients (BPPCC) used in the i power synthesis algorithm are verified during power ascension testing. The l predicted BPPCC are calculated by simulating a free unrodded Xenon oscillation similar to the SAM measuiement procedure. The predicted BPPCC are then determined from a regression analysis of the top and bottom one-third core average power integrals and the boundary point powers at the top and bottom of the core. In the calculation of a penalty factor associated with BPPCC uncertainty, in-core detector measurement error and depletion effects are considered. The in-core detector signal error is randomly selected from a standard normal error distribution and applied to the normalized core average axial power distribution. Depletion effects arp considered by simulating a xenon oscillation at B0C, MOC, and E0C conditions. The top and bottom boundary point powers and the top and bottom one-third is integrals of the core average power distribution are then calculated. By using a least square fitting technique using these boundary point powers and one-third core average integrals, BPPCC values with detector signal measurement error and depletion effects are determined. Figure 4-1 shows the calculational procedures for penalty factors due to BPPCC uncertainty. Approximately twelve hundred (1200) cases at each time-in-life of CPC LHR and DNB-0PM are used to calculate the BPPCC values with and without in-core detector signal measurement error and depletion effects. The cases used for the BPPCC penalty factors include changes in power distribution, depletion, power level, CEA configuration, power l maneuvers, and Xenon concentration. The CPC results are compared with j those of the reactor core simulator and the difference in LHR and DNB-0PM
error is then calculated for each case. By analyzing the resultant distributions, penalty factors for LHR and DNB-0PM due to BPPCC uncertainty are calculated based on most non-conservative error difference at a 95/95 probability confidence level. Reference
- 4.1 Combustion Engineering, Inc., " Assessment of the Accuracy of PWR Safety System Actuation As Performed By the Core Protection Calculators", CENPD-170-P and Supplement, July,1975.
I e m.
- m 6
6 1 l l D l
Table 4-1 Rod Configuration For The Calculation Of Penalty Factors Due To Rod Shadowing Factor Uncertainties a Part Length Case Regulating Rod Group Rod Groups Safety Rod Groups
'~
1 All Rods Out - - 2 Bank 5 - - 3 Bank 5 + 4 - - 4 All Rods Out 1 + 2(1) - 5 Bank 5 1 + 2(1) _ 6 Bank 5 + 4 1 + 2(1) - (1) All part length rod groups (13 PLR CEA) are treated as a single operating groups.
~ .
mm i l I i
---n, - - --- - - - - - - , , - . - ,, , , - - - .,, - - - - - - - - -
Figure 4-1 Calculational Procedure for Penalty Factors due to RSF, SAM, and BPPCC Uncertainty 1
~
CPC CPC run base run (R) with R + AR Reactor core ~ Simulator F , DNB-0PM q CPC F , DNB-0PM calculate F , DNB-0PM q q for 1200 cases for 1200 cases with base value .
, with R +/M v _
T- u
~
Calculate error Calculate error for each case for each case and analyze error and analyze error distribution distribution v u Tolerance Limit Tolerance Limit (TLO) (TL) O Penalty Factor
QUESTION 5 The procedure used throughout the report for evaluating variable sensitivity is determined by evaluating _ L a(%Y) J DIUi ) i where y = f (X , X2 , X3 , ---- X ) and X is the variable whose sensitivity isn being 1 determined. N Demonstrate 1 that the sensitivity does not change for different values of X 2, X 3, etc.
. RESPONSE (1) Sensitivity for RSF Startup Test Acceptance Band Uncertainty A detailed calculational procedure to determine the sensitivity factors associated with the rod shadowing factors (RSF) is described in answer to Question #4. The sensitivity factor values used for RSF startup test acceptance band uncertainty are not generated from any specific set of operating conditions, but are generated from cases (1200 cases at each time-in-life) chosen to encompass steady state and quasi-steady state conditions in accordance with System 80 CPC design requirements. Therefore, the sensitivity factor values used in the RSF startup test acceptance band uncertainty are not changed for different values of other important operating parameters (i.e., core poweg level, power distributions, CEA configurations, xenon co'ncentration, etc.). Refer to que,stion 4.1 for further details. ,
(2) Partial Derivative for Dynamic Pressore Uncertainty In order to determine dynamic pressure uncertainty, the C ~ ~1 partial derivative of DNB-0PM with respect to pressure is used and determined by [. lcalculated_usingthefollowingequation: 4 (5-1) t
.. ._)-
l In deriving [ ] thirty six hundred (3,600) cases of CETOP-D l calculations are performed over the range of System 80 operating conditions l listed in Table 5-1. The CETOP-D results with different pressures but with I I the same operating conditions for other parameters are paired and the partial derivative of DNB-0PM with respect to pressure is calculated using equation 5-1. As discussed above, the partial derivative for dynamic pressure uncertainty is not generated from any specific set of operating conditions,[~
~ ~ ~
~ ~ ~ ~ ~ ~ ~]conditionswhich l
span the CPC operating space. Therefore, the value used for DNB-0PM partial derivative with respect to pressure does not change for different values of other operating parameters (temperature, flow, etc.). l
(3) Sensitivity for DNBR Computer Processing Uncertainty The DNBR computer processing uncertainty is determined by using thef DNBR sensitivity and the DNBR error. The[ ]DNBRsensitivityis
]
obtained from the DNBR sensitivity defined in the following equation:
. (5-2) -
~ ~~
In order to calculate [. CETOP-D calculations are performed over
- the range of System 80 operatin]g conditions listed in Table 5-1. From the CETOP-D r.esults obtained for the 3,600 cases, cases with different DNBR values but the same operating conditions for other parameters were paired . and the partial derivative of DNB-0PM with respect to DNBR was calculated using equation 5-2.
The sensitivity value used for DNBR computer processing uncertainty is calculated from the set of operating conditions [
.] Therefore, the sensitivity value used[. lat other operating conditions given in Table 5-1.
(4) Sensitivity for CECOR Fxy Measurement Uncertainty
~
The[ ] sensitivity used for CECOR Fxy measurement uncertainty is calculated from CETOP-D calculations for the operating conditions given in Table 5-1. From the CETOP-D results obtained for 3,600 cases, cases with different radial peaking factor values but the same operating conditions for other parameters were paired. The partial derivative of DNB-0PM with respect to radial peaking factor was 3hus calculated using the following equation: . (5-3)
. ,i .
Since the sensitivity value used for CECdR Fxy measurement uncertainty is joperatingconditionsgiveninTable5-1
- -~ ~' '
- the[
over the range of System BU operating space, these values do not change for different values of other operating parameters. e
TABLE 5-1 Operating Conditions for System 80 NSSS
. Parameters Symbol Unit Operating Conditions
. Mass Flow Rate G 106 lbm hr-ft 2 ;
Inlet Temperature T *F Pressurizer Pressure P psia . . _ _ _ _ .., Radial Peaking F None , Factor Axial Shape Index ASI None - -
~
+
_~ - m 5 P
OUESTION 6 Justify, derive or provide a reference for equation 2-9 in Section 2.4.1.3. What, if any, are the restrictions on the Pi's necessary for the validity of the equation?
. RESPONSE The startup test acceptance band uncertainty (PS) is made up of penalty factors
. due to uncertainties on Rod Shadowing Factors (RSF), Temperature Shadowing Factor (TSF), Shape Annealing Matrix (SAM), and the Boundary Point Power Correllation Coefficients (BPPCC). The penalty factors take into account the possible non-conservatism of the CPC LHR and DNB-0PM calculations due to uncertainties in the CPC power distribution constants.
The penalty factors are obtained from equation 2-8 in Section 2.4.1.3 of Part II. The second term of equation 2-8a,f ] stands for a fractional penalty factor (<1.0) due to startup-test acceptance band uncertainty. Equation 2-9 is used to combine the individual penalty factors to provide a single overall penalty due to uncertainties on RSF, TSF, SAM and BPPCC. The basis for equation 2-9 is a[ _
.] technique . This is appropriate as the individual Pi's are obtained via independent measurements and the functional relationship describing power distribution in the CPC is essentially a linear function, therefore higher order derivatives need not be included in equation 2-9 (Reference 6.1). The startup test acceptance band uncertaintyfactors(PS)will.be[ =
~ ~~] included in the net LHR and DNB-0PM penalty factors.
The Pi's are calculated-based on the start p test acceptance criteria bands which affect the CPC/COLSS LHR and DNB-0PM calculations. Therefore, the determined Pi's are restricted such that no thanges greater than acceptable criteria are allowed. At' Reference 6.1 G. J. Hahn and S. S. Shapiro, " Statistical Models in Engineering," Chapter 7, John Wiley & Sons, January 1967. D 0
OVESTION 7 Section 2.4.1.4 discusses the treatment of uncertainties associated with axial fuel densification, fuel rod bow, computer processing and engineering factors. _ Explain why the axial fuel densification uncertainty factor is handled differently from the other factors.
RESPONSE
~
The axial fuel densification factor is not statistical in nature but represents the shrinkage of the pellet stack due to heating and irradiation. In principle, this factor could be directly included in the Average Linear Heat Generation Rate (ALHGR) by assuming that the effective fuel rod length is the nominal fuel rod length divided by the axial densification factor. However, neither CPC/COLSS directly includes this factor in the ALHGR calculation. Therefore, the axial fuel densification factor is applied as a multiplier on the CPC and COLSS LHGR calculations. e em 6 1 e e l 1'
l OUESTION 8 Provide a detailed description on how the axial fuel densification uncertainty, fuel and poison rod bow uncertainties, and the engineering factor uncertainty are determined. What, if any, are the restrictions of these uncertainties (i.e., plant specific or generic)? , RESPONSE
- a. The axial fuel densification uncertainty is calculated from the following equation (Reference 8.1)
F = 1/ ((1+E/100) (1 -Ae/200)) ! , where, E= percent linear thermal expansion and,Ae= percent change in i pellet density. This uncertainty is calculated on plant specific basis, even though the variations between plants are negligible. i
- b. The fuel rod bow uncertainty is calculated based on the methodology described in Reference 8.2. The important parameters used are given below.
- 1. Hot pin average heat flux which corresponds to reactor over power trip and design not pin radial peak,
- 2. Reactor pressure which coresponds 40 reactor over pressure trip,
- 3. Gap width,
- l l
- 4. Cladding eccentricity, -{ -
l 5. Cladding geometry,
- 6. Grid spacing,
- 7. Grid-rod interference force,
- 8. Fuel non-uniformities and fuel clad interaction,
- 9. As fabricated rod bow ,
- 10. Burnup
- 11. Enrichment The power changes in the center rod due to the bowing of the surrounding rods in each of eight directions were calculated to be, linear functions of bowing displacement.
1
~
The standard deviation of rod-to-rod channel closure (in inches) as a function of burnup (BU) at the " hot" condition is expressed as, Se = 1.2 (a + b(BU)1/2)
where: a = standard deviation of as fabricated channel dimension (in.), a b= span-dependent burnup coefficient, in (MWD /MTU)1/2 BU = assembly average burnup (MWD /MTU). The a 95%increase in power probability (to% of the fuel will not exceed is given by:gfgg) due to fuel ro that 95 t95/95 = i BSj c where, B is the linear heat rate augmentation coefficient, which is empirically derived by least squares technique from measured data and analytical calculations, and expressed as, B = a + b E + C(BU) + d (BU)2: , a, b, c and d are constants, E = fuel enrichment (w/o U-235)- BU = fuel exposure (MWD /MTU). For the purpose of generic evaluation, a conservative upper limit is calculated corresponding to 4.0 w/o fuel enrichment and 35,000 MWD /MTU burnup. = The poison rod bow uncertainty which is the change in linear heat rate due to poison rod bow is found to be very nearly a linear function of bow magnitude and the largest power increase for a given displacement accompanies bowing in the lateral direcfion. Using this information and . the lateral displacement, a coefficient B* is computed which conservatively relates poison rod displacement (S95/95) and linear heat rate augmentation (t*95/95) t*95/95 = B*S95/95 A conservative maximum value is calculated, and it is generically applied. Refer to Reference 8.2 for a detailed description of fuel and poison rod bow.
- c. The engineering factor uncertainty is an allowance provided to account for the effect of variations in the fuel pellet and clad manufacturing process. These result in variations in the quantity of fissile material and variations in the gap conductance, and therefore, affect the fuel stored energy. Variations in pellet density, diameter, and enrichment are included in this allowance, as are variations in clad diameter and
. thickness. This factor is generic to all 16 x 16 fuel.
l t i u .--_-._. . , . - . . . -_ -
1 References 8.1 Fuel Evaluation Model, CENPD-139-P-A, July 1974. 8.2 Fuel and Poison Rod Bowing, CENPD-225-P, October 1976. 6 le 1 e % i 6 9
QUESTION 9 Provide a justification for the method used to combine the quantities referred to as "Kar" values in equation 2-14 of Section 2.4.1.5. Show that the result can indeed be used to obtain a 95% probability /95% confidence tolerance limit.
. RESPONSE The justification for using the[. _ .
] method for combining the "Ka" values in equation 2-14 is provided by performing an alternate The tolerance limits computed using comp the [utation using[. method have in every case examined proved to be more limitin similar verification procedure is used for equation 2-26 and is repeated on a case-by-case basis. A comparison of the DNB and FQ tolerance limits as computed by both techniques is given in the following table.
~
[ _ ]
~
DNB0PM FQOPM . NOTES: 1. Upper tolerance limit 95/95.
- 2. Lower tolerance limit 95/9E.
In future applications.C-E will use thef I' ~--
~ ' ~~~
~ __
,to evaluate the tolerance limits described ts equations'2-14 and 2-26.
b l i I l i s.- .- .,,..c,---,__ -
QUESTION 10 Describe the wide ranges of radial peaking factors and axial shape indices (ASI) used in determining the dynamic pressure uncertainty in Section 2.4.2.2. [ RESPONSE
~
The dynamic pressure uncertainty is expressed as (Equation 2-20, Reference 10.1) follows: - , where . 7 . . _ , The dynamic pressure compensation offset, APD is the maximum non-conservatism pressure bias during transient events, due to CPC process timing, , pressure sensor delays, and RPS delays between trip signal and event turnaround. [. __ _
~
]is calc 01ated for the range of plant operating
. conditions given in Table 2.2 of Reference 10.1 and the wide ranges of radial peaking factors and axial shape indices given below:
~
Radial peaking factor: ,
~
Axial Shape Indices: _. The CPC will prevent reactor operation by ond these ranges by providing a range trip. - Reference l 10.1 Enclosure 1-P to LD-83-010,' Statistical Combination of Uncertainties, Part II. i
- l .
OVESTION 11 Explain why the chosen partias derivative of DNB-0PM with respect to the pressurizer pressure is used for pressure sensitivity and provide additional details on how that value is determined. The same question also applies to other uncertainty factors such as DNBR computer processing uncertainty, fuel . rod bow uncertainty and system parameter uncertainties, etc.
RESPONSE
Dynamic Pressure Uncertainty The uncertainty for the dynamic pressure is represented as follows: where, [ _ _ . _ _ . , _ __ .
,___3 -
The dynamic pressure compensation offset, APD, is the maximum non-conservatism pressure bias during transient events due to CPC process timing, pressure sensor delays, and RPS delays between trip and event turnaround. This APn is calculated and the maximum value_is taken. L , _ _ __
'~
_ _3- - - -
~
determined as given below: ~~'
- 1. The DNB-0PM is calculated for different cases in the wide range operating conditions as follows:
Parameters Units Range p Temperature 'F _. 1 Flow 106 1bm/hr-ft 2 ___ , Radial Peaking Factor None l ASI None DNB-0PMs for approximately 3600 cases were obtained within the wide range. l 2. The cases with different pressures, but with otherwise the same operating conditionswerepairegandthepartialderivativesofDNB-0PMwithrespect
-to pressure for the i case i_s calculated using the following equation:
l
The[ _ 3-DNBR Computer Processing Uncertainty The DNBR computer processing uncertainty (Kr)cp is calculated by using the following equation: , (Note: Reference 11.1 inadvertently expressed the above as where , [' '. ' ~ ~
]DNB-OPM' sensitivity to DNBR, an@NBR is the DNBR error between the CPC Fortran Simulator computer (CDC-7600) and the CPC's.
[ ]DNB-0PM sensitivity to DNBR is calculated as follows: DNB-0PMs are calculated for different sets of operating conditions (within the CPC range) for a minimum DNBR of 1.250. For the same set of operating conditions, and for a minimum DNBR of 1.350, the DNg0PMs are calculated. The partial derivative of DNB-0PM to DNBR for the i case is calculated by using the following equation:
~
]
~
The DNBR error between th'e of'f-line computer CDC-7600 and the on-line computer is calculated as follow'ie The DNBR's are calculated for a set of operating conditions using the CDC-7600 computer. For the same set of operating conditions and with the same algorithm, the DNBRs are calculated using the on-line computer. The error in predicting the DNBR for the i th case is expressed as: ADNBRj = DNBR (on-line) - DNBR (off-line) DNBR (off-line)1, A statistical evaluation is made on the errors thus obtained.~~ThegNBR ' [ _ .-_ _ _ _ J. e 0 w
Funl~ Rod Bow Uncertainty The fuel rod bow uncertainty is directly applied on the Linear Heat Rate and there is no conversion involved. System Parameter Uncertainties The system parameter uncertainties and how they are applied are described in " Statistical Combination of Uncertainties - Combination of System Parameter Uncertainties In Thermal Margin Analyses for System 80", Enclosure 1-P to L2-82-054. Reference
. 11.1 " Statistic al Combination of Uncertainties", Parts II and III, Enclosuresl-P and 2-P to LD-83-10, January 1983.
! e p 1 W 0
QUESTION 12 Referensce origin of values for the secondary calorimetric power measurement error, the secondary calorimetric power to the CPC power calibration allowance, and the thermal power transient offset as described in Appendix B. Provide r justifications for these values.
RESPONSE
Secondary Calorimetric Power Measurement Error and Calibration Allowance The secondary calorimetric power measurement error consists of the uncertainty components for the following parameters.
- 1. Feedwater Flow
- 2. Feedwater Temperature
- 3. Secondary System Pressure
- 4. Pressurizer Heaters
- 5. Reactor Coolant System Losses
- 6. Coolant Pump Heat i
- 7. Component Cooling Water .
A typical analysis of the secondary calor:imetric power error, based on the above uncertainty components and secondaty instrument accuracies, is provided in Figure 12-1. The values used in Reference 12.1 are expected to bound the design secondary calorimetric power error. Verification of the secondary calorimetricpowererrorisperformedduringstartuptesging. The secondary calorimetric power to the CPC power calibration allowance is l based on Technical Specification allowances. Table 4.3-1, Note 2 of the CESSAR l Technical Specification states that adjustments are to be made to the Linear Power Level, CPCAT Power, and CPC nuclear power signals if the absolute difference with the secondary calorimetric power calculation is >2%. This allowance is consistent with that for other CPC plants. . Thermal Power Transient Offset The thermal power transient offsets on CPC DNB and LPD calculations 'are evaluated to assure that the CPC Design Basis Events (DBEs) are adequately modeled. The DBEs that are limiting for the determination of these offsets are those which involve single CEA misoperations. The limiting DBE for the thermal power transient offset on the CPC DNB calculation is the single CEA withdrawal from full power, which gives the most nonconservative CPC calculation of heat flux. The thermal power transient !~ offset on the CPC DNB calculation was determined as[ ]which covers the
maximum nonconservatism involved. The tharmal power transient offset is us d in the evaluation of the uncertainty bias for the thermal power in CPC DNB calculation, BERRO, to correct the thermal power. Since the neutron flux power is not credited for any of the single CEA DBEs in CPC, the uncertainty bias for the neutron flux power for DNB, BERR2, does not require a transient bias offset component. The limiting DBE for the thermal power transient offset on CPC LPD calculation is the single full-length CEA drop. This event gives the most nonconservative power for CPC calculation of LPD. Th thermal power transient offset on CPC LPD calculation avas determined as which covers the maximum nonconservatism involved. This thermal power tr sient offset on CPC LPD calculation is used in the evaruation of uncertainty bias for the power used in CPC LPD calculation, BERR4, to correct the power. Reference 12.1 Combustion Engineering, Inc., " Statistical Combination of Uncertainties, Part II, Uncertainty Analysis of Limiting Safety System Settings, C-E System 180 Nuclear Steam Supply Systems," Enclosure 1-P to LD-83-110, January 1983. Mb
.~
m M l i l
,,, i. , i.
i FIGURE 12-1 l SECONDARY CALORIMETRIC POWER ERROR i - 3 ) i i 5 - i 4 _ l e E
$ i'i J g3 -
I E
\
u 2 i .
~
1 I I I I I I I l l l 0 10 20 c 30 40 50 60 70 80 90 100
% OF DESIGN RATED POWER i
l
OUESTION 13 Appendix A of the report states that CETOP-1 and CETOP-2 are simplified versions of CETOP-D and perform the on-line thermal-hydraulic calculations for the plant monitoring and protective systems, respectively. Provide a detailed description on the difference between CETOP-1 and CETOP-2 and CETOP-D. Has CETOP-1 been approved by the NRC?
RESPONSE
~
CETOP-D is a fast running, accurate, core thermal-hydraulics calculator. It is used as the setpoint DNBR - Overpower Margin calculator for all CPC/COLSS
~
pl ants. As such, CETOP-D is benchmarked against TORC /CE-1. The general CETOP methodology is described in Reference 13-1. The CETOP-D code is described in detail in References 13-2 and 13-3. CETOP-1 is a version of the CETOP-D code which has been streamlined for use in the Core Operating Limit Supervisory System (COLSS). Tne major difference between CETOP-1 and CETOP-D are the following: . _ _ _ i _ . ~ ... .._ .._ ..__ ,__- .._ . . __ W . ---=+=W-+.*+--h---=e. - + - - - -.e , _ _ - - . _ . _ _a_ _ _ _ . m These differences between CETOP-1 and CETOP-D result in CETOP-1 having a shorter execution time while essentially maintaining the accuracy of CETOP-D. Figures 13-1, 2 and 3 show the CETOP-1 vs. CETOP-D results at B0C, M0C, and E0C for approximately 1000 randomly selected points at each time-in-life. CETOP-2 is also a fast running version of CETOP-D. This ' version has been streamlined in order to meet the timing and core memory storage requirements of the CPCs. CETOP-2 has been described in References 13-4, and 13-5. Figures 13-4, 5, and 6 show the CETOP-2 vs. CETOP-D results at B0C, M0C and E0C for approximately 1000 randemly selected points at each time in life. The primary use of CETOP-1 is in COLSS which is a control grade monitoring system. CETOP-1 was chosen to be used in the reactor core simulator because of its s_ hor _t execution time and very high accuracy compared with CETOP-D and CETOP-
- 2. E . . . _
,.%., ., 4 - - .m+e._-ens-+ -w = = =- uw-.== -- + - -**.=s.*-=**
l l
References 13.1 Dr. Chong Chiu, "Three-Dimensional Transport Coefficient Model and
" Prediction-Correction Numerical Method for Thermal Margin Analysis of ~
PWR Cores," Nuclear Engineering and Design, Vol. 64, No.1, March
- 1981, pp. 103-115.
13.2 "CETOP-D Code Structure and Modeling Methods for San Onofre Nuclear Genefating Station Units 2 and 3", Docket No. 50-361, 50-362, CEN-160(S)-P, Rev. 1-P, September 1981.
-~
13.3 "CETOP-D Code Structure and Modeling Methods for Arkansas Nuclear One - Unit 2," CEN-214(A)-P, July 1982. 13.4 "CPC/CEAC Software Modifications for Arkansas Nuclear One - Unit 2," CEN-143(A)-P, Rev. 1-P, September 1981. 13.5 " Response to Questions on Documents Supporting the ANO-2 Cycle 2 License Submittal", CEN-157(A)-P with Amendments 1-P, 2-P, and 3-P, 1981. . m t m' 6 0 4 F l I l l l
o o o. o e
=
y N
= =
-o ,
sn o w a _,.. o g s ~
\.
=
s -
\ Rm o
o O 1 m o
~ ~
w o cr: s x o ._ C o C
. o -
o ~ o z. r_
~
- o o
U - 55 O _ 3 c0 . 1 c3 I
) '
< 8 m
CL. 1 \ O ! l-- LLJ u - U I i l l i o o D o o o o o o
> a o o a o o m o e o e o
~
N N - "
~
- l. CL.
O F--- ( .01X) LLJ
-001/ucio Ta017]
U
o a w e4
==
~
o hk O N
. \ __
a __ w m.
\= c
%,' o w
N I - en w o 5 e u- o o
, o -
. . o s a r C
- o
_- - o, _ c_ _ o o u o L o W I a l i I ! o o o a o o a. o o o o o o o o o e o e o cu N - - e l t cotx3 i 001/Wa0 Id017]
o o o cw S E
-m . o o = - o w o a an ,
cu
. u W -
=
N e O o II in m - e m
- m _
w o ce x
=a o ._
- o
' o 3
- _ o s o r
_. c
- o o
[ C - S o d uJ o C ,
' o o
I n 1 o -
+ =
LLJ U
~ C l I t I C e i U3 o o o o o a-o o
o o o o a O e o o a-en ca -
~
CL_ C F-- ( O ! X. ) g "001/uJO IJ017] U
o o l l w i , C4 i ! ... c .
= -
iJ
. o -
o w o g o E OJ k j N 1 I
\ :
\g
- i o
O v.
\
I v -- A O w o cc -
=> x e -
u . c o a C -
. y a ~
i o n
; C6
; 7, _ - o
~
I c. (.__ - I e I _. c C- i I w CC . o
, i, i r-' ! ; e a
1 tr 3 '
- r. .
C . L- - ig j s - I l l l l l 1 I i 1 C c 0 C c c o c c c-C y c c c c a o s- c z> c x a
- ., c, - .
~ CJ CL C
L___.
'1' COI/udC 2dC133
( _>
o o 4 N m u a .
= "a y a o m ..
o a i, a o m a C'J 9
\k
% o o
w I i - m - 9 m a
~ ~
o w - cc x o o o C o i m o - w a ~ i o n l l '
- o o
O _ c o C
^
i l " E j O ' o o
\ D l
L l C F-- . i -. LLJ j - U i i I l I I I I o I o
< C o o o o o o
> o o o o o o e o e o o o o N - -
~
(N CL O F-( : O!X) W 001/Wa0 2dC133 0
o o i a i i I , N I i I t . I i i ! E ! i 9 i i i. _ . i
.o .l i o t
o ia
; o a m
\
I = i i ,
% I' s
6 m. o a x e, i
\% l I
i M _ m ' i s LAJ i - a: ' o o . e i .
- u. ;
o o
. , o -
t o , i a r
- =
.- - o y) .
g _ I t n
'A u
- i
. I t._.) o 5
! o 1 m x
u
=* -
l tr ,
- 3 t
l 1 i o
! o
'a 1 t'f,
.a, o o o a O o
~ , s o- a o o o o o u . O e o.
w <~ - - m W r-
; -0IX)
W 00i/Wdo ?dO1~3] L.)
QUESTION 14 The report does not provide values of uncertainties and errors and indicates that they will be provided later. Are these values plant specific? What are the generic values? Provide a list of items which are plant-specific for each individual CESSAR plant and describe how these plant-specific items interface with the CESSAR generic submittal.
RESPONSE
' References 14.1 and 14.2 provide uncertainties and errors used in statistical combination of uncertainties. The uncertainties and errors used in the reports (Parts II and III) are listed in Table 14-1. The values of the error components are generic for all CESSAR-F plants. The as-built error components for each CESSAR-F reactor will be compared to the generic data to verify that the generic data is bounding. The impact on the uncertainty factors of any plant specific data which is not bounded by the CESSAR-F values will be addressed on that plant's docket. Therefore, the uncertainty factors derived from the CESSAR-F SCU analysis are generically applicable to all plants referencing CESSAR-F. References for the generic data are also given in Table 14-1.
em 9 9 4 0
- v - -- - -
Table 14-1 A List of Plant Data Error Component
,g--. .
k
-m- -- ~---m m ---
e W '
- m e
- -- - m -
P N
-- maam -e
~' * *- e . _.._
men-wem -- e e.-. Nmular -e + .
.. "h=-ehe 9#O 6
6
-w-w e-- --w- - -. . ____ 6 m
l l l l l i l l l 1 - e l l
I References 14.1 Combustion Engineering, Inc., " Statistical Combination of Uncertainties, Part II, Uncertainty Analysis of Limiting Safety System ~ Settings, C-E System 80 Nuclear Steam Supply Systems," Enclosure 1-P to LD-83-10, January 1983. 14.2 Combustion Enginering, Inc., " Statistical Combination of Uncertainties, Part III, Uncertainty Analysis of Limiting Conditions for Operation, C-E System 80 Nuclear Steam Supply Systems," Enclosure 2-P to LD-83-10, January 1983. 14.3 Combustion Engineering, Inc., " INCA /CECOR Power Peaking Uncertainty", CENPD-153-P, Rev. 1-P-A, May 1980. 14.4 Combustion Engineering, Inc., " Fuel and Poison Rod Bowing", CENPD-225-P and Supplements, June 1978. 14.5 Combustion Engineering, Inc., " Fuel Evaluation Model," CENPD-139-P, October 1974. 14.6 Combustion Engineering, Inc., " Statistical Combination of Uncertainties, Combination of System Parameter Uncertainties in Thermal Margin Analyses for System 80," Enclosure 1-P to LD-82-054, Nby 1982.
~ ==
6
-- ,- -- - - - - , ,}}