Fuzzy Galactic Swarm Optimization with Dynamic Adjustment of Parameters Based on Fuzzy Logic
Emer Bernal, Oscar Castillo, José Soria, Fevrier Valdez · SN Computer Science · 2020
In this work, a fuzzy method for dynamic adjustment of parameters in galactic swarm optimization is presented. Galactic swarm optimization is based on the movement of stars and galaxies in the universe, as well as their attractive influence allowing the use of multiple cycles of exploration and exploitation to solve complex optimization problems. It has been observed in the literature that the utilization of fuzzy systems for dynamic adjustment of parameters in metaheuristic algorithms produces good results when compared to using fixed parameters in the algorithms. In this work, the adjustment of the c3 and c4 parameters is made through the use of fuzzy systems because these parameters have a significant role in the operation of galactic swarm optimization. We tested the fuzzy approach with a set of benchmark mathematical functions and with the fuzzy controller of the water tank problem to measure the performance. Finally, a comparison of the results is presented among the proposed method and other metaheuristics.