Tuning of Control Parameters of Grey Wolf Optimizer using Fuzzy Inference

Abstract

The Grey Wolf Optimizer (GWO) is a recent metaheuristic that can be explored in many applications. This paper proposes a mechanism to tune the control parameters that influence the hunting process in the GWO in order to improve its efficiency. This adjustment is made by a fuzzy inference system that uses the normalized fitness value of each wolf and the hunting mechanism control parameters of the GWO. The proposed fuzzy mechanism is tested and compared with the conventional GWO and another version that uses a fuzzy system as input information the ratio of the current iteration number and the maximum number of iterations. For performance analysis of the proposed fuzzy mechanism, all tested optimizers in ten benchmark optimization functions ran 1000 times. Simulation results show that the proposed fuzzy mechanism improves the convergence of the conventional GWO and it is competitive in relation to another fuzzy version adopted in the GWO design.