A multistrategy hybrid adaptive whale optimization algorithm

Abstract

Abstract To solve the problems of whale optimization algorithm (WOA) with slow convergence speed, low accuracy, and poor global search ability, a multistrategy hybrid adaptive whale optimization algorithm (MHWOA) was proposed. In this algorithm, the logistic–tent hybrid chaos algorithm was used to initialize the population, which could make the population distribution more random and uniform. The opposition-based learning strategy was adopted to expand the population of individuals and complete the population selection. To balance the exploitation phase and exploration phase, a dynamic parameter was constructed based on the sigmoid excitation function, and an active adaptive weight was added to adjust the global and local search, which accelerated the convergence speed also. The perturbation mechanism of the Student T-distribution was introduced with random perturbation to expand the search range and improve the global search ability of the algorithm. In total, 23 benchmark functions were selected to conduct convergence performance and optimization performance experiments of the proposed algorithm. The average value and standard deviation were determined as evaluation indexes. The MHWOA was compared with other improved WOA variants and advanced algorithms. The results showed that the proposed MHWOA had better iterative convergence and optimization performance than different algorithms on the optimization of unimodal functions, multimodal functions, and fixed dimension functions. Meanwhile, the MHWOA was applied to the optimal designs of pressure vessels and springs. The experimental results displayed that the MHWOA obtained better solutions than other meta-heuristic algorithms. This study has practical solid application value, which can be applied to solving various engineering problems.