Accelerated grey wolf optimiser for continuous optimisation problems

Abstract

Grey wolf optimiser (GWO) is a relatively simple and efficient nature-inspired optimisation algorithm which has shown its competitive performance compared to other population-based meta-heuristics. This algorithm drives the solutions towards some of the best solutions obtained so far using a unique mathematical model, which is inspired from leadership behaviour of grey wolves in nature. To combat the issue of premature convergence and local optima stagnation, an enhanced version of GWO is proposed in this paper. The proposed algorithm is named accelerated grey wolf optimiser (A-GWO). In A-GWO, novel modified search equations are developed that enhances the exploratory behaviour of wolves at later generations, and the exploitation of search space is also improved in the whole search process. To validate the performance of the proposed algorithm, set of 23 well-known classical benchmark problems are used. The results and comparison through various metrics show the reliability and efficiency of the A-GWO.