Hybrid whale optimization algorithm based on symbiosis strategy for global optimization

Maodong Li,Guang-hui Xu,Liang Zeng,Qiang Lai

DOI: https://doi.org/10.1007/s10489-022-04132-9

IF: 5.3

2022-12-15

Applied Intelligence

Abstract:The whale optimization algorithm (WOA) is a simple structured and easily implemented swarm-based algorithm inspired by the unique bubble-net feeding method of humpback whales. Past studies have shown that WOA performs well in a number of optimization problems. However, it is difficult for WOA to completely free itself from the problems of insufficient convergence accuracy and premature convergence when solving global optimization problems. To address these issues, a hybrid whale optimization algorithm based on symbiotic strategy (HWOAMS) is proposed in this paper. The main idea of the proposed method is to combine the improved symbiotic organisms search algorithm (SOS) with the whale optimization algorithm thus enhancing the search ability of WOA. First, an improved symbiotic phase based on Lévy flight and chaos strategy is introduced into the exploration process to enhance the global search capability; Second, an improved mutualism phase based on Brownian motion is used instead of the original shrinking encircling phase to achieve better local exploitation. Third, an improved parasitic phase based on a modified global optimal spiral operator strategy is embedded in the spiral updating position phase to help the algorithm further improve the exploitation efficiency and convergence accuracy. Finally, a global search strategy is proposed to help the algorithm better balance exploration and exploitation. To establish the effectiveness of the new algorithm, extensive simulation experiments are conducted on HWOAMS using the classical function test set, the CEC 2019 function set and four classical engineering problems. Numerical evaluation results indicate that HWOAMS outperforms 18 other algorithms in terms of local optimum avoidance ability and convergence accuracy in a majority of cases, and has better search performance.

computer science, artificial intelligence