ML19209A545
| ML19209A545 | |
| Person / Time | |
|---|---|
| Issue date: | 09/12/1979 |
| From: | Brooks W Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19209A542 | List: |
| References | |
| NUDOCS 7910040302 | |
| Download: ML19209A545 (5) | |
Text
.
ENCLOSURE Review of WCAP-9485, "PALADON - Westinghouse Nodal Computer Code" Report No. WCAP-9485 (Proprietary)
Report
Title:
PALADON - Westinghouse Nodal Computer Code.
Report Date: December,1978 Originating Organization: Westinghouse Electric Corporation Reviewed By: Core Perforsance Branch /W. L. Brooks Westinghouse Nuclear Energy Systems has submitted licensing topical report WCAP-9485 entitled, "PALADON - Westinghouse Nodal Computer Code".
The report describes the calculational models employcd in the code, including both the neutronics and ther-nal-hydraulic models, and documents the cualification of the results against both experiment and higher order calculations. Our evalution~of this report follows:
1.
Sumary of Report The methodology used in the PALADON code is discussed. PALADON is an
~
improved version of earlier nadal -codes. The improvements are discussed and the calculational. algorithm is derived. Both neutronic and themal-hydraulic con =ideration are discussed. Feel and burnable poisen depletion calculations are described. The xenon steady state and transient equations are presented as well as the steady state and buildup ecuations for
" samarium.
In addition to the description of the PALADON code, data are cresented to qualify the code for the calculation of cower distributions, control red worths and reactivity coefficients. Ccecarisons are made to measured data from startup tests or to calculations cerformed with the Westinghouse-design diffusion theory code TURTLE. The data base used in the comparison includes data from 15 different olants and 36 different core designs in-
- luding a wide s:ectrum of possible core configurations. The folicwing treas are aedressed:
7910040 J op 1090 265
?00lFOR8EL
, ~
1.
isothermal temperature coefficient comparison to measurement, 2.
control rod worth comparison to measurement, 3.
power distribution comparisons to fine mesh diffusion theory (TURTLE) and measurement, 4.
critical boron concentration comparison to fine mesh TURTLE,
- 5.
dropped bank accident power. distribution comparison to fine mesh TURTLE,
-~
6.
dropped rod accident power distribution comparison to measure ment, and 7.
N-1 red worth comparison to~ measurement.
For each area addressed the data base is presented, including individual comparisons between calculations and standard (measurement or higher order calculation), the mean and standard deviation of the discrcpancy between calculation and standard are presented, and conclusions are drawn regarding the suitability of PALADON to perform the calculations.
2.
Sunnary of Review We have reviewed the description of the PALADON model including the neutronics calculations, the geometric model, and the thermal-hydrualics calculation. The nodal calculation technique has a history of successful application for reactor calculations and the description of the PALADON code is sufficient to permit the conclusion that it is state-of-the-art and is therefore ~ acceptable.
Com:arisens are presented between measured and calculated isotter al tem-cerature coefficients for 3-and *-loop plants for core c:nditions ranging frem unrodded to al' regulating banks (A+5'C+0) in and for three evcies of coeration. A total of 55 data coints for first : Ores and 27 data Ocints for reload cores are : resented. Bascd on these data it may
- e :encluded :na: PALACCN :redicts measured values, wnen a:Orceriate
- 'ases nave :een 1: lied, :: wi nin 3 Ocm/~.
(3x10~I ik/k :er *:)
1090 266
. PALADON control rod bank worth predictions were compared to measured data colTected in Westinghouse's control red worth program. Comparisons were made for 3-and 4-loop plants and for single bank and multiple bank configurations. The data presented support the conclusion that PALADON predicts rod bank worths to the same accuracy as that of' the currently approved design methods.
Comparisons of power distribution have been made between thes calculated by PALADON and those calculated by fine mesh diffusion theory. Ccecarisons were made for both 3-and 4-loop plants and for up to three. cycles of.
operation. Two types of comparisons were made - avcrage :ssembly pcwer and peak red power in an assembly. PALADON does not calculate the peak rod powers directly - rather, they are derived from the assembly average powers and an algorithm for local peaking factor which is derived from fine mesh diffusion theory calculations.
The results of these comparisons are presented and support the conclusion that PALADON calculates average assembly power to the same accuracy as coes fine mesh diffusion theory. While this is not the case with the calculation of peak Nd pcwers a fine mesh correction factor is acplied to the PALADDN calculation of peaking factors which makes these quantities conservative. This is an acceptable procedure.
Comparisons between fine mesh diffusion theory and PALADON calculations of critical baron concentration at zero oower and full power have been made for up to four cycles of operation for several 3-and 4-1000 plants.
The data presented show that a smell bias exists between PALADON and fine mesh diffusion theory and that the bias is the same for zero power and full ecwer and at beginning and end of cycle. This fact is further evidence that moderator and Occoler thermal feedbacks and fission creduct worths in PALACON are com: arable to fine mesh diffusion theory.
~ ree ; sa: c:ncitions nave been investigated wit! tne U LA~CN :0de. 3ese
' cluce 9e :0wer distributiens resul-ing ' rem a droceed ::ntr:i d bank 1-0 a sing'.e Orc::ed ::n rel rec and ne 1-1 rec wer:n. B e er ::ec bank 200ROR8NAL 1990 267
. comoarisons are made with the results of fine mesh diffusion theory calcula-tions and the latter two comparisons are made with measured values. The comoarisons for drocped control rod bank configurations were done for In all cases the average assembly power, peak assembly power and peak Fg.
PALADON results were equivalent to or conservative with rescect to fine mesh diffusion thecry when the fine mesh correction factor is applied to the peak results.
For the single droeped control red comparisons were made to measurements perfomed during plant startuo tests. These comparisens orovide evidence that these upset power distributions may be calcula' tad by PALA00'N and also provide sufficient data to derive a value for the uncertainty to be apolied to the calculated peaking factors. We conclude that PALACON acceptably calculates dropped rod peaking factors when the aparceriate uncertainty has been added.
The data presented regarding the calculation of the N-1 red worth are from startup physics test for thirteen.eactors. These data demonstrate that PALADCN calculations of the N-1 red worth are of similar accuracy to those of standard design methods and the uncertainty is low cc:.: pared to that assumed for this quantity (10 percentl in safety analysis.
3.
Evaluation Procedure We
.<e reviewed the topica' report WCAP-9485 within the guidelines provided by the Standard Review Plar.. Section 4.3.
We conclude that sufficient infomation has been provided to permit a knowledgeable person to conclude that the PALACCN code is a state-of-the-art nodal analysis code.
We nave reviewed the data presented in succort of the verification of tne
- LACCN code. Sufficient data are : resented and analyses :erfor*ed to emi: tre following conclusions:
'.so:. pal tem: era ure :ef"cients are :reci::ec to.vi:nin 3
- me - wnen a::r::rta e biases a e a::11ec.
2.
- ntrol r
- c bank aor ns are credictec to wi:nin re same accuracy
~
as cresently a:oroved design Te:nocs.
P00R ORGINAL 109o 268
. 3.
Assembly average powers are predicted to within the same accuracy as the cresent design methods.
4.
Assembly peak pcwers are predicted cunservatively when corrected by the fine mesh correction factor.
5.
Critical boron concentrations are predicted with acceptable accuracy.
6.
Dreeced control red bank peaking factors are predicted to the st.1e accuracy as design methods when the fine mesh correction is Laplied.
7.~
Dropped sf.1gle red peaking factors are precicted with accectable accuracy.
S.
The N-1 rod worth is predicted to an accuracy consistent with accroved design methods - much less than the ten percent uncer-tainty assumed in safety analyses.
4 Regulatory Position On the basis of our review we conclude that topical report WCAP-9485 may be
~
referenced ist licensing actions for. calculations of those cuantities described above for which it has been qualified. Further qualification will. be re-quired befcre the PALACON code may be used to calculat4 1xial or three-dimensional ocw~er distribution or perform xenon transient calculations (such as those for constant axial offset control analysis).
1090 269