Improved sparrow search algorithm with adaptive multi-strategy hierarchical mechanism for global optimization and engineering problems

Aiming at the problem that the sparrow search algorithm (SSA) does not have high optimization accuracy and is prone to fall into local optimum, an improved sparrow search algorithm with adaptive multi-strategy hierarchical mechanism is proposed (ISSA). Firstly, in the initialization phase, the population is created by combining the triangular topology and Logistic Chaos mapping, and an elite dynamic reverse learning strategy is used to enhance the population diversity and balance the local and global search performance. Secondly, an adaptive multi-strategy hierarchical mechanism is applied to the population, where adaptive dynamic adjustment strategy is applied to the discoverers to improve the flexibility and search efficiency of the algorithm; differential mutation operation is applied to the followers to generate a mutated subpopulation, which enhances the ability of the algorithm to jump out of the local optimum; and vertical and horizontal crossover strategies are applied to the vigilantes, where horizontal crossover enhances the global search ability, and vertical crossover maintains the population diversity and prevents the algorithm from falling into local optimality. Finally, the classical benchmark functions as well as the CEC2020 and CEC2022 test functions are selected for simulation and analysis, and the ISSA is compared with other optimization algorithm, and the ANOVA analysis, the Wilcoxon rank-sum test, and the Friedman test are performed. The simulation results show that the ISSA proposed in this paper achieves significant improvement in both convergence accuracy and convergence speed. Meanwhile, the application of ISSA to engineering problems fully verifies its practical value and significant advantages in the field of engineering problems.