Multi-objective optimization of a PWR core loading pattern by backtracking search algorithm

In this study, the core loading pattern of the initial core configuration of a typical Pressurized Water Reactor has been optimized through the Backtracking Search Algorithm (BSA). The multi-objective fitness function is based on a trade-off between minimization of the power peaking factor (ppf) and maximization of the cycle multiplication factor (k(eff)) simultaneously. Neutronic computations are performed using the PSU-LEOPARD (Pennsylvania State University-Lifetime Evaluating Operations Pertinent to the Analysis of Reactor Design) and MCRAC (Multiple Cycle Reactor Analysis Code) codes. The PSU-LEOPARD generated assembly data have been fed to MCRAC and it calculates normalized power profiles for all fuel assemblies with a specific loading pattern. The BSA generates best loading patterns by optimizing the multi-objective function. The implementation of the BSA scheme resulted in slight enhancements in the first cycle length (similar to 10.1 %). The BSA demonstrates rapid convergence, high efficiency and robustness for the core loading pattern optimization problem.