Multi-objective design optimization of a fast spectrum nuclear experiment facility using artificial intelligence

Abstract Nuclear experimental design is often a balance between competing objectives or goals for the experiment. The Fast Neutron Source (FNS) at the University of Tennessee would be a unique platform for building targeted integral nuclear cross section measurement experiments to reduce nuclear data uncertainty on future reactor designs. In this paper we apply a genetic algorithm using non-dominated sorting, elitism, and crowding distance heuristics to explore the design space of a simplified 1D approximation of the FNS, and from this data, we present correlations in the final suite of designs which will be the basis of heuristics for more complex optimizations and geometries in the future. In this analysis, we show that a multi-objective genetic algorithm can be used to design nuclear experiments with the dual objective of spectrum matching and flux maximization.