Improved verification methodology for TRAC-PF1/NEM using NEA/OECD core transient benchmarks

This paper describes the verification methodology of TRAC-PF1/NEM based on the NEA/OECD core transient benchmarks. The necessity of verification of the coupled best-estimate codes capable to predict the transient behavior of light water reactors (LWR) is well recognized. The TRAC-PF1/NEM code, currently maintained in The Pennsylvania State University (PSU), is a coupled three-dimensional (3D) neutronics/thermal-hydraulics (T-H) best-estimate code that allows simulating the postulated design basis accidents for pressurized water reactors (PWR). The nodal expansion method (NEM) is implemented in the neutronics module of TRAC-PF1/NEM. TRAC-PF1 is a system transient analysis code allowing 3D fluid dynamics vessel modeling as well as 1D balance-of-plant representations. Rod ejection/withdrawal accidents, which constitute 3D transient problems with the moving of central or peripheral control assemblies, are analyzed in this paper. It is demonstrated that the obtained results lire in the range of one neutronic node-per-assembly and one T-H cell-per-assembly reference results. it is also shown that the modifications of the thermal conductivity and heat capacity of both uranium oxide and zirconium cladding, introduced into TRAC-PF1 according to the benchmark specifications in comparison with the initial TRAC correlations, do not change the steady state and transient results. The analyzed NEA/OECD core transient benchmarks have been recently added to the developed TRAC-PF1/NEM qualification procedure. These benchmarks help to improve the verification methodology by establishing a well-defined set of test problems for future TRAC-PF1/NEM modifications and distributions. (C) 2000 Elsevier Science Ltd. All rights reserved.