ImprovedCatch Fish Optimization Algorithm with Personalized Fishing Strategy for Global Optimization

Catch Fish Optimization Algorithm (CFOA) is a new meta-heuristic optimization algorithm with human behavior. In this algorithm, search agents simulate the process of rural fishermen fishing in the pond. Therefore, the CFOA generally consists of two phases of the update: the exploration phase and the exploitation phase. However, it still falls under the local optimum and has a low convergence rate. To this end, we propose an improved catch fish optimization algorithm(ICFOA) based on personalized fishing strategies. First, the adaptive Gaussian perturbation is adopted to the exploration stage process to increase the global search capability, expand the search range, and improve efficiency while avoiding falling into the local optima. Then, based on the personalized fishing strategy, the personal position of fishermen is updated by randomly selecting “freehand fishing” factors or “using fishing net” factors to accelerate the algorithm's convergence speed. Furthermore, comparative experiments were performed using the CEC2020 test suite to compare the performance of ICFOA and other excellent meta-heuristics. Finally, Wilcoxon's rank-sum test was used to verify the validity of our statistical experimental results. Moreover, the performance of ICFOA in reducer design also indicates that ICFOA can get the optimal solution in solving practical engineering optimization problems. The results show that ICFOA has a more competitive performance than the original CFOA.