Applicability and validation of TULIP code based on fast spectrum benchmarks

Abstract

BackgroundNECP-SARAX is a neutronics analysis code system for advanced reactors developed by the Nuclear Engineering Computational Physics Laboratory team of Xi'an Jiaotong University. In the past few years, a considerable amount of verification and validation work has been done based on CEFR, PHENIX, SUPERPHENIX, JOYO MK-I, ZPR, and ZPPR reactors. The results indicate that NECP-SARAX offers high performance for fast spectrum reactor analysis. Meanwhile, the fuel and control rod assemblies of these reactors are used for verification of the cross-section generation code TULIP. While TULIP has demonstrated promising preliminary results in fast spectrum system analysis, a comprehensive systematic verification and validation process remains essential.PurposeThis study aims to validate the applicability of TULIP code for various fast spectrum systems.MethodFirstly, a total of 147 critical experiment benchmarks were selected from ICSBEP and used for analysis. The initial results demonstrated that the keff bias between TULIP and Monte Carlo codes exceeded 10-2 for an experimental benchmark with a thick reflector. Then, a homogeneous two-nuclide problem simplified from the HMF021-002 benchmark was subsequently used to analyze this phenomenon, and the intermediate-weight nuclides had resonance-like fluctuating scattering cross sections above the resonance energy. Finally, to address this phenomenon, the TULIP code was undergone enhancements, mainly focusing on optimizing the resonance calculation strategy and method using ultra fine group to deal with thee self-shielding effect of resonance-like cross sections in the non resonant region under high loading of intermediate-weight nuclides.ResultsIn a fast spectrum system with a large amount of structural material, the self-shielding effect of the resonance-like cross section of the intermediate-weight nuclides above the resonance range becomes non-negligible. The optimized TULIP method reduces the keff bias to within 3×10-3 for these benchmarks with a thick reflector.ConclusionsNew numerical results indicate that the enhanced TULIP code has good performance for various fast spectrum system analyses.

Keywords

• tulip code
• icsbep benchmark
• intermediate-weight nuclide
• nonresonance energy
• resonance-like cross section