COPERNICUS: A multi-cycle optimization code for nuclear fuel based on parallel simulated annealing with mixing of states

COPERNICUS is the Studsvik code for performing nuclear fuel optimization over a multi-cycle planning horizon that provides for an implicit coupling between traditionally separate in-core and out-of-core fuel management decisions. These decisions include determination of: fresh fuel region size; sub-region enrichments and bundle designs; exposed fuel re-use; and core loading pattern. The COPERNICUS methodology is based on a parallel implementation of the Simulated Annealing optimization algorithm, modified by the technique of Mixing of States, that allows for deployment in a processor scalable environment. COPERNICUS utilizes the 3-D licensing grade code SIMULATE for evaluation of all core loading pattern constraints, such as those involving reactivity and thermal margin requirements. Results are presented for a transition cycle design that compares performance of multi-cycle optimization to successive, single cycle optimization with regard to reducing levelized fuel costs.