Transport equivalent full core neutronics calculations via finite volume method and optimized diffusion parameters


Progress in Nuclear Energy, vol.165, 2023 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 165
  • Publication Date: 2023
  • Doi Number: 10.1016/j.pnucene.2023.104935
  • Journal Name: Progress in Nuclear Energy
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Compendex, Environment Index, INSPEC, Pollution Abstracts, Civil Engineering Abstracts
  • Keywords: Finite volume method, Full core, Genetic algorithm, Krylov subspace methods, Neutronics
  • Hacettepe University Affiliated: Yes


In this study, three-dimensional neutronics analysis of the hot zero power (HZP) situation of the BEAVRS full core benchmark problem is performed by employing a finite volume method (FVM), Krylov subspace methods, and transport equivalent diffusion model (TEDM). TEDM is used for generating transport-corrected diffusion parameters and we use a genetic algorithm for the correction. The study shows that a transport equivalent and accurate neutronics analysis of a highly detailed commercial light water reactor (LWR) core is feasible using only the two-group diffusion theory. The core model in the benchmark is represented by 165 different diffusion parameter sets since the model is highly heterogeneous. The reaction rates, flux, and current distributions obtained from Serpent code, the fuel assembly-based FVM diffusion theory solutions, and a genetic algorithm are used to generate the transport equivalent diffusion parameters. The task of the genetic algorithm used is to minimize the difference between Serpent and diffusion theory solutions by estimating the two group diffusion coefficients and the scattering matrix. The transport equivalent diffusion parameters are then used for the full core neutronics analysis by FVM and Krylov subspace methods. The results of the analysis are compared with the measured data provided by the BEAVRS benchmark and the results of Serpent code simulations and found to be in complete agreement with the Serpent code results and satisfactorily consistent with the measured data provided by the BEAVRS benchmark.