Analysis on influence of multigroup nuclear data uncertainty based on stochastic sampling method

Stochastic sampling method is used in this paper to study the influence of the uncertainty of multigroup nuclear data on reactor physics calculation. Random values for the nuclear data such as multigroup microscopic cross-sections are generated by sampling the AMPX multigroup library using the self-developed SAMP module and the covariance matrix of SCALE software. Lattice calculations and the steady core calculations are then performed by SCALE and PARCS respectively. The uncertainties of lattice and core calculation results are obtained by statistical analysis accordingly. Taking the fuel assembly and first cycle core of Almaraz PWR plant as the objects, the uncertainties of lattice calculation results such as effective multiplication factor, kinetic parameter, nuclide concentration and two-group macroscopic cross section, and the uncertainties of steady core calculation results like core power distribution are analyzed. The results indicate that the assembly parameters are dependent on the fuel burnup, and the maximum uncertainty of steady core radial power distribution is about 1.9%.