Improved Grey Wolf Optimizer Based on Opposition-Based Learning

Abstract

Swarm intelligence (SI)-based algorithms are very popular optimization techniques to deal with complex and nonlinear optimization problems. Grey wolf optimizer (GWO) is one of the newest and efficient algorithms based on hunting activity and leadership hierarchy of grey wolves. To avoid the slow convergence and stagnation problem in local optima, in this paper, opposition-based learning (OBL) is incorporated in GWO for the population initialization as well as for the iteration jumping. In this strategy, opposite numbers have been used to deal with the problem of slow convergence. The proposed algorithm is named as opposition-based explored grey wolf optimizer (OBE-GWO). To evaluate the performance of OBE-GWO, it is tested on some well-known benchmark problems. The experimental analysis concludes the better performance of OBE-GWO.