Pattern optimization of PWR reactor using hybrid parallel Artificial Bee Colony

In this paper, we propose a core reloading of pressurized water reactors technique based on a hybrid Artificial Bee Colony (ABC) algorithm. Our approach integrates the merits of both ABC algorithm and Particle Swarm Optimization (PSO). The neighborhood search scheme of the algorithm is improved by location of personally encountered the most flowers and location of highest concentration of flowers explored by the intact swarm. The probability of selecting food sources by onlooker in the proposed algorithm is enhanced through using roulette-wheel mechanism. To address the drawback of most optimization algorithms, this method has been parallelized, so that the runtimes may be greatly reduced by using a multiprocessor computer cluster. The proposed optimization method is applied to the cycle length maximization of a VVER-1000 core. Simulation results show that the proposed ABC method could have the advantages of original ABC, and is capable of producing low cost, fast, and reasonably accurate solutions.