Evaluation of Eigenvalue Separation by the Monte Carlo Method

Abstract

A new technique for evaluating an eigenvalue separation by the Monte Carlo method is proposed. This technique uses a fission matrix evaluated by an ordinary Monte Carlo calculation and enables us to evaluate higher order eigenvalues and eigenvectors that have not been able to be determined by Monte Carlo calculation. To validate it, the proposed technique was applied to the analysis of the modeled BWR core and the coupled-core reactor assembled at the C-core tank of KUCA (Kyoto University Critical Assembly). Consequently, it was found that this new technique gave good results regarding eigenvalue separation, although regional division needs to be noted in a system for evaluating a fission matrix.