Comparison of the Baseline USL Calculation Methods for Loosely Coupled and Novel Neutronic Systems

The goal of this study is to evaluate the accuracy of different upper subcritical limit (USL) calculational methods for loosely coupled and novel neutronic systems. This study varied the separation distance over five center-to-center separation distance intervals for four loosely coupled models and explored seven single-system neutronic models. Each of these 27 systems was simulated using MCNP6.2 with 200 randomly perturbed, continuous-energy ENDF/B-VII.1 cross-section files that are in the TENDL 2019 library. The distribution of the values from these perturbed runs was used to calculate stochastic 99/99 USL values for each model iteration. USLs were also estimated for these 20 systems using the Whisper 1.1 code, and the Whisper-identified relevant benchmarks were used to further analyze the relationship between the region-wise USL calculation and the overall system USL calculations. Sensitivity data files were produced using MCNP6.2 and then used with the Oak Ridge National Laboratory TSURFER and USLSTATS methods to estimate USLs for a cross-method USL comparison. A windowing study was performed when using the USLSTATS method to determine the efficacy of the method using datasets with differing degrees of similarity to the given application case. The results show that USLs for each of the loosely coupled system models were higher USLs than the calculated stochastic USLs. The single-system uranium models also displayed a consistently lower stochastic USL as compared to the USL calculational methods, while the single-system plutonium models showed close agreement between the stochastic USLs and the other USL calculational methods.