Application of genetic algorithm for optimization of control rods positioning in a fast breeder reactor core

Abstract In this work, an integer coded genetic algorithm is implemented for finding the optimal arrangement of twelve absorber rods (control rods) within the active core of a fast breeder reactor, satisfying and optimizing operational and safety parameters. This is a multi-objective, multi constraint combinatorial optimization problem. The problem is having a large search space due to the huge number of control rod arrangements possible. A chromosome in genetic algorithm, which denotes a particular arrangement of control rods, is represented as a set of 12 integers. Each integer stands for one control rod and denotes the position of the rod in the reactor core. The genetic operations, crossover and mutation are modified and adapted for this particular study to take into consideration the geometric positioning of control rods within the core. The results obtained show that the algorithm is able to converge to optimal, feasible configurations by exploring the search space effectively. The performance of algorithm is further enhanced by applying parallel programming techniques in the evaluation of genetic population. The study attempts to validate the effectiveness of genetic algorithms in handling objectives and constraints related to reactor physics as well as engineering design domains.