关键词: 水波优化算法, 种群间相互作用策略, 概率扰动策略, 复形法, 投资组合优化

Abstract: A multi-strategy improved water wave optimization is settled to address the issues of low convergence accuracy and susceptibility to local optima in water wave optimization (WWO). In the propagation stage of the algorithm, an inter population interaction strategy is introduced to achieve information exchange within the population, thereby improving the convergence accuracy and stability of the algorithm. This paper designs a probability perturbation parameter that replaces the position update mechanism refracted in the original algorithm when the random number is less than the probability perturbation parameter, so that the new position of the individual will be randomly determined within the search space range, avoiding a decrease in population diversity in the later stages of iteration. The reflection operation of the complex method is applied to the individuals with the worst fitness to improve the ability of algorithm to escape from local optima. The simulation experiment is based on 30 dimensional and fixed dimensional test functions. At the same time, convergence analysis and Wilcoxon test are conducted on the MSWWO and the other algorithm, proving that MSWWO has better convergence performance and robustness. The application of MSWWO in solving portfolio optimization problems has verified its effectiveness and reliability in solving practical problems.

Key words: water wave optimization algorithm, inter population interaction strategy, probability perturbation strategy, complex method, portfolio optimization