Use of the fission matrix method for solution of the eigenvalue problem in a spent fuel pool

In this paper, we examine the use of the fission-matrix method to calculate the eigenvalue for a spent fuel pool. The fission matrix coefficients are calculated using MCNP5. In order to make the method as efficient as possible for real-time calculations, these coefficients are pre-calculated. Certain simplifying assumptions are made based on geometric considerations, greatly reducing the amount of pre-computation, and allowing the coefficients to be used on a variety of problems beyond that for which the coefficients were calculated. This methodology is applied to a reference pool that is being designed for the I2S-LWR project. Typically, the eigenvalue calculation in Monte Carlo is very difficult for loosely coupled problems such as a spent fuel pool due to source convergence issues, which are not present using the fission matrix. The fission matrix method has shown accuracy very close to that of MCNP5, while giving results in several orders of magnitude less time. Total pre-calculation time was less than a single eigenvalue calculation and can be used across many different pool configurations.