Optimization procedure for radial BWR fuel lattice design using genetic algorithms

An optimization procedure based on the Genetic Algorithms method was developed for the optimization of enrichment and gadolinia distributions in boiling water reactor fuel lattices. The optimization process was linked to the HELIOS code to evaluate and qualify all the investigated solutions. The goal is to search for the optimal fuel lattice distribution, which has the lowest average enrichment between two defined values, and the minimum local power peaking factor; which satisfies an average gadolinia concentration target and a k-infinite multiplication factor, also target. The weighting factors, used to give the relative importance of the evaluation parameters in the objective function, take two different values depending on the values of the evaluation parameters. This strategy helped the process to rapidly guide the search toward configurations which satisfy all the constraints. The genetic algorithm uses one crossover point, and two offspring. The main contribution of this work is the development of an efficient procedure for BWR fuel lattice design, using a powerful algorithm and an objective function which saves computing time, because it does not require lattice burnup calculations. The results show the good performance of this procedure; fuel distributions were found with low enrichment, low radial power peaking factor and good reactivity performance during the lattice life. (authors)