关键词: 鲸鱼优化算法, 动态自适应探索转换策略, 鲸鱼个体聚集度跟随策略, 邻域解变异增强策略, 工程优化

Abstract: In order to solve the shortcomings of the whale optimization algorithm (WOA), such as low convergence accuracy and the tendency to easily fall into local optimum, a multi-strategy improved whale optimization algorithm (MSWOA) is proposed. Firstly, a dynamic adaptive exploration conversion probability is designed to replace the random exploration probability in the original algorithm, so that the excellent individuals close to the optimal individual can easily guide the global search, which is conducive to enhancing the quality of the solution and preventing the algorithm from falling into the local optimum. Secondly, the concept of individual whale aggregation is introduced, and when whales fall into a state of aggregation, large steps are used to update the position to prevent the population diversity from decreasing in the later iteration stage. Finally, a neighborhood demutation enhancement strategy is designed to consider the influence of the current individual and its neighbors on the position of the next generation of individuals, so as to prevent the population from entering the aggregation state and improve the ability of the algorithm to jump out of the local optimum. Based on 29 test functions in the CEC2017 and 10 test functions in the CEC2019, the simulation experiments explore the influence of the three improved strategies on the exploration and development of the algorithm, the impact on population diversity and the convergence of the algorithm. Furthermore, MSWOA is applied to the design of pressure vessel and reducer, and the effectiveness and reliability of MSWOA in solving practical problems are verified.

Key words: whale optimization algorithm (WOA), dynamic adaptive exploration transformation strategy, whale individual aggregation degree following strategy, neighborhood solution variation enhancement strategy, engineering optimization