Abstract:

In view of the problems of low convergence accuracy and falling into local optimum when solving complex problems, a sparrow search algorithm based on adaptive t-distribution and random walk is proposed in this study. In the initialization process, the algorithm uses reverse learning to generate reverse solutions from which excellent individuals are selected to form the initial population. In the original sparrow search algorithm, the adaptive t-distribution strategy and Gaussian random walk strategy are used to improve the optimization ability of the sparrow individuals, and can prevent the algorithm from premature. The simulation results show that the proposed algorithm improves the convergence accuracy and convergence speed when compared with the comparison algorithm.

Key words: sparrow search algorithm, adaptive t-distribution, opposition-based learning strategy, random walk strategy, function optimization, local optimum, global optimum, optimistic algorithm