Multistrategy boosted multicolony whale virtual parallel optimization approaches

Highlights：

• We propose a hierarchical decision-making model based on information sharing, which should effectively coordinate global exploration and local exploitation.

• Strategies of information dissemination among whale individuals in different subcolonies are designed to improve the efficiency of information exchange and space search according to their learning ability, information dissemination capability and relationship distance.

• We apply the proposed MSMCWOA to address 23 unimodal and multimodal function optimization problems and two engineering optimization problems.

• The proposed approach has a better optimization accuracy (average value and best value), solution stability (Std.), convergence rate and computational cost than state-of-the-art approaches.

摘要

•We propose a new multistrategy boosted multicolony whale optimization algorithm (MSMCWOA) motivated by the advantages of information exchange, mutual learning, and cooperation of complementary elements, which can better simulate the social behavior of whales. At the same time, parallel collaborative optimization among subcolonies is formed, which can diversify the population and avoid premature convergence.•We propose a hierarchical decision-making model based on information sharing, which should effectively coordinate global exploration and local exploitation.•Strategies of information dissemination among whale individuals in different subcolonies are designed to improve the efficiency of information exchange and space search according to their learning ability, information dissemination capability and relationship distance.•We apply the proposed MSMCWOA to address 23 unimodal and multimodal function optimization problems and two engineering optimization problems.•The proposed approach has a better optimization accuracy (average value and best value), solution stability (Std.), convergence rate and computational cost than state-of-the-art approaches.