On Test Functions for Divergence-based Grey Wolf Optimizer

Abstract

Nature-inspired algorithms have captured the attention of the research community in recent times. Due to the ease of implementation and the advancement in technology, these algorithms have found their niche in the field of optimization. Their applications span from the designing beam in civil engineering to the prediction of diseases in medical sciences. One such widely researched algorithm, published recently, is grey wolf optimizer (GWO); based on the behavior of the grey wolves. This grey wolf optimizer has gone through hybridization and modifications as is natural in this domain. One such recently developed variant is divergence-based grey wolf optimizer (DGWO). This paper details the working mechanism of DGWO and presents the performance based on benchmark functions. For this, 23 well-known benchmark functions implemented in python are used. Seven of the functions are unimodal and six are multimodal and ten are fixed-dimensional multimodal functions. The results for the test functions are presented by using 2D graphs. The results show that the newly developed DGWO works comparably well and is suitable for solving optimization problems.