Preliminary Generation of Windowed Multipole Covariance Data

Abstract

A methodology for the generation of covariance data in the windowed multipole formalism for resolved resonance data was developed. Using this process, covariance matrices were created for oxygen-16, uranium-235, and uranium-238, such that sampled cross-sections would match a user-selected average standard deviation, set at 5% ± 0.5% for this work. Windowed multipole datasets were then sampled and used on a simple critical benchmark to demonstrate the use of WMP data in conjunction with the Total Monte Carlo method of uncertainty propagation.