ML20003F874
| ML20003F874 | |
| Person / Time | |
|---|---|
| Issue date: | 02/18/1981 |
| From: | Minogue R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| RIL-114, NUDOCS 8104230798 | |
| Download: ML20003F874 (3) | |
Text
_
"R t L' FI L E
, 'y
}
[
v UNITED STATES g
o,,
NUCLEAR REGULATORY COMM!sslON E
5, g
j W ASHINGTON. D. C. 20555 o
a
"%,,,,, +#
FEB 161981 3
MEMORANDUM FOR: Harold R. Denton, Director Office of Ntclear Reactor Regulation h
FROM:
Robert B. Minogue, Director
(
Office of Nuclear Regulatory Research
SUBJECT:
RESEARCH INFORMATIM LETTER NO.114 "PIPIMG BENCHMARK PROBLEMS" "5
INTRODUCTION AND
SUMMARY
- - - - - 4 h
Dynamic analyses of piping systems are normally perfomed using computer s
programs to handle the complex geometries and various static 'or dynamic loading conditions. Applicants for nuclear pow 6r plant licenses are
):
i.
required to provide confimation of the adequacy of their programs. The requirement to provide this confirmation is given in the Standard Review Plan, Section 3.9.1 and Appendix B,Section III of 10CFR50. The Office f
-? -
of Nuclear Reactor Regulation has indicated a need for verification of these programs.
p 5
This Research Infomation Letter (RIL) transmits the results of efforts at Brookhaven National Laboratory (BNL) to develop benchmark problems and solutions, of sufficient complexity to check the adequacy of the 2
{
applicant's computer programs. This work represents one task of the program, " Mechanical Piping Benchmark Problems."
Seven piping problems and solutions were selected as a benchmark for the verification of computer programs using unifam support motion and the response spectrum method of analysis. The benchmark problems and solutions d
are presented in the report, NUREG/CR-1677, " Piping Benchmark Problems,
}'.
Volume 1."
All the benchmark problems were used in some fonn'to evaluate y
the seismic method of analysis of the five plants which were shut down
?)
'in 1979 in response to IE Bulletin 79-07. The computer programs, EPIPE.
which were used for generating solutions to the benchmark problems, and
?
.E PSAFE2, are suitable for the analysis of linear elastic structures
.. f subjected to static or dynamic loads. These were developed and are "A-available on the BNL computer system.
~.
. [-
? -
I
};
8104230
~~
n..
~ ^*
e.
w,.
.~..
(
k]
Harold R. Denton 2
FES 1 g 193]
APPROACH Owing to the lack of problems suitable for benchmarks in the open literature, specific problems were developed. To acc. plish this, an available general purpose finite element code (EPIPE) was modi"ed and cptimized i:
to perform piping analysis of elastic structures in accordance with 5
current NRC practice (response spectrum method, Regulatory Guide 1.92).
The updated code (EPIPE) was then verified against available closed form j
solutions, limited experimental results, and solutions developed from n
,C existing docestic (ANSYS) and European (SPANDLE) computer programs.
f, These verification studies are described, in part, in the report BNL-t f NUREG-2i241, dated January 1976, by M. Reich, T. Y. Chang and S. Prachuktam.
4 The benchmark problems and solutions were next developed and transmitted f.
to various nuclear power plant liensees for preliminary evaluation.
ff The subsequent matching of results (between the benchmark problem solutions
!y.,
and those submitted by the licensees) corroborated the correctness of the solutions, and further verified the EPIPE Code.
c The computer program, EPIPE, is a completely verified, full feature, k,?<
finite element piping code. It permits the static analysis of piping
!I systems subjected to grtvity, concentrated loads, thermal expansion, i.i pressure and anchor / supports moment loads. EPIPE incorporates additional fT options to allow dynamic analysis via the response spectrum method, time
!F history method using modal superposition, and the time-history method jj using direct integration. For all analysis options, it perforns stress 5
evaluation in accordance with accepted engineering formulae.
I ?..
(
EPIPE is serving as the backbone for a more sophisticated and usaful
- 2 program called PSAFE2. Eventually, PSAFE2 will incorporate the features of EPIPE and serve as the primary computer code for static and dynamic gy analysis. PSAFE2 was developed to verify applicant compliance with ASME i
..l '
Code Criteria, a feature not available in EPIPE. In addition PSAFE2 allows for analysis of piping systems on a production basis. PSAFE2 2.
performs stress evaluations of ASME class 2 and class 3 piping in accordance
?
with the appropriate ASME criteria, and allows the evaluation of systems
> vb.-
subjected to multiple support excitation. The code is continually g V' modified to expand its capabilities and to comply with the benchmark p
problem development needs as defined by NRR. Continuing verification of 2
the codes is accomplished through their use in developing confirmatory
' h' problem solutions to assess the acceptability of applicant solutions and i {.
analysis methods.
- i
- ?
RECOMMENDATIONS t Gi The benchmark problems should be incorporated into the licensing procedure.
Each applicant for a nuclear power plant license should be required to solve the benchmark problem set using its computer programs. NRC (i?.
evaluation of these results will verify the applicant's methods.
' }i ~
,.n
'l P
i
~
~
. J
~
~
a
~
(.
- ==
10 OI Harold R. Denton 3
The elastic anhlysis code, EPIPE, should be used to perfonn independent evaluations and confirmatory analyses of selected applicant problems.
These evaluations will further verify the applicant's computer programs while also serving as a check on applicant implementation.
i
!j CONCLUSIONS The benchmark problems developed to date have a restricted. range of.
application.
They can only verify the accuracy of the analytical solution of computer programs used for the seismic analysis of linear elastic piping structures subjected to uniform support motion and evaluated using the response spectrum method of analysis. A future set of benchmark
=-
p' problems will be developed to extend the ra.1ge to evaluation of linear j
elastic piping structures using time-history methods, and also, independent multiple support excitation reponse spectrum methods.
In addition, i
j major efforts are underway to extend the benchmarks to assess the degree of applicant ccepliance with ASME criteria for both class 1 piping in accordance with ASME-Boiler and pressure Vessels Code, Sectson III, y
Subarticle NS-3500, and class 2/3 piping in accordance with ASME, y,
Section III, NC/ND-3600. The information, results, and recommendations reported in this RIL were reviewed for appropriateness and application j
~ to the licensing process by the cognizant NRR techincal monitor for this
, y, program.
W
^
Robert B. Minogue, Director I
Office of Nuclear Regulatory Research
'.~;. -
cc:
T. E. Murley, DST
. f,,f.
.G. Knighton, RSCB s -
l
- ?;.,,
1-
' R*'s
.; Y 1
l
'ff-f
.II J
'*y:
.O 3
--- +-,-,,... -,, -.,
,c-
~
-