Article Preview
Top1. Introduction
The swarm intelligence optimization algorithm is inspired by the biological world of nature, simulating the behavior of some things or creatures in the natural world, and searching for the global optimization in the solution space. In recent years, new swarm intelligence optimization algorithms have continued to emerge. Scholars have proposed a series of swarm intelligence optimization algorithms through ants, wolves, birds, moths, whales, and sparrows, such as ant colony algorithm (ACO), grey wolf optimization (GWO), moth-flame optimization (MFO), whale optimization algorithm (WOA), sparrow search algorithm (SSA). Because swarm intelligence optimization algorithm has the advantages of easy operation, strong robustness, and wide application range, many scholars pay attention to it. The sparrow search algorithm was first proposed by Xue in 2020 (J. K. Xue & B. Shen,2020), which is a new type of swarm intelligence optimization algorithm. Compared with other algorithms, the sparrow search algorithm has the characteristics of higher solving efficiency. However, in the later stage of the algorithm iteration, it is still the same as other intelligent algorithms, and it is easy to get stuck in the local extremum.
In order to improve the shortcomings of the swarm intelligence optimization algorithm that is easy to fall into the local optimum at the later stage of the iteration, the global optimization ability is improved, many scholars have proposed different improvement strategies. Oliva used the chaos operator to chaotically map the whale position update probability (D. Oliva, M. A. E. Aziz & A. E. Hassanien, 2017), global optimization performance is improved. The Logistic chaotic mapping is applied to the particle swarm algorithm (W. L. Yang, X. T. Zhou & M. N. Chen,2019), the diversity of the solution is enhanced, and it reduces the probability of the algorithm being trapped in the local space to a certain extent. The convergence speed of the algorithm is accelerated by adding an adaptive weight factor (A. E. Hegazy, M. A. Makhlouf And S. G,2020), and it is successfully applied to the feature selection. Wang et al. proposed an adaptive inertial weight, which changed the speed update method, so that the bat algorithm can dynamically and adaptively adjust the speed during the search process (X. W. Wang, W. Wang & Y. Wang, 2013). The sinusoidal inertia weight factor is added to change the position update method of the gardener bird, the global and local development capabilities are effectively balanced (Y. R. Wang, D. M. Zhang And Y. Fan,2020). The Cauchy mutation operator is introduced to avoid the firefly algorithm from falling into the local optimum (W. H. WANG, L. XU, K, W. CHAU, et al.2020), the global search ability is improved. The convergence speed and accuracy of particle swarm optimization are improved by introducing nonlinear control parameters and Cauchy mutation (J. Li, Y. K. Luo, C. Wang, et al.,2020). The reverse learning strategy is used to the particle swarm algorithm, the diversity of the population is enriched, and the overall exploration ability has been effectively improved (P. F. Xia, Z. W. Ni, X. H. Zhu, et al., 2021). Gaussian mutation and Cauchy mutation are integrated into the moth algorithm, local and global exploration capabilities are enhanced (Y. T. Xu, H. L. Chen, J. Luo, et al.2019). With the help of the Cauchy mutation ability, the cat swarm is approached to the global optimal solution, the occurrence of precocious phenomena is prevented (L. Pappula & D. Ghosh,2017).