DSGWO: An improved grey wolf optimizer with diversity enhanced strategy based on group-stage competition and balance mechanisms

Abstract

Grey wolf optimizer (GWO) is a popular algorithm in the field of meta-heuristics with fast convergence speed and strong exploitation ability for solving different types of optimization problems. However, there are two problems presented in the optimization process of GWO. Problem (1): GWO algorithm uses three leading wolves to direct the population for search, resulting in poor population diversity. Problem (2): Individuals selected in the optimization process of GWO have the current optimal fitness value, resulting in poor global search capabilities of the algorithm, and the population may tend to converge prematurely and fall into local optimum. In order to solve the above problems, this paper proposes the Diversity enhanced Strategy based Grey Wolf Optimizer (DSGWO), which combines two mechanisms to improve the performance of the GWO algorithm, including the group-stage competition mechanism and the exploration–exploitation balance mechanism. The experimental results are evaluated by using IEEE CEC 2014 benchmark functions and two engineering problems. The Wilcoxon rank-sum test is also conducted to measure the performance of DSGWO. The results demonstrate that the DSGWO algorithm can efficiently obtain the optimal solution of the dataset used in this paper, preserve more population diversity, and enhance the global search capability. In addition, through the application of two engineering problems, it is verified that DSGWO is suitable for solving engineering design problems.