基于等级制度和布朗运动的混沌麻雀搜索算法

针对麻雀搜索算法在迭代后期种群多样性减弱、易于陷入局部最优等问题,提出了一种基于等级制度和布朗运动的混沌麻雀搜索算法(CSSA-HB).首先引入混沌映射调整麻雀搜索算法关键参数;其次引入等级制度,利用父代种群中3个最优个体对警戒者进行位置更新,加强个体间交流,增强种群多样性;然后利用布朗运动可控均匀步长,增强算法的探索能力;当算法陷入停滞时,利用布朗运动策略对个体施加扰动,促使算法跳出局部最优;最后利用贪婪策略保留优势个体,有效加快收敛速度.对12个测试函数进行仿真实验,结果表明混沌映射能有效增强算法性能,迭代映射表现最佳;改进算法具有较强的局部最优规避能力、更快的收敛速度和更高的收敛精度.

A Chaos Sparrow Search Algorithm Based on Hierarchy and Brownian Motion

Aimed at the problems that population diversity is weakening and easy to fall into local optimization in the late iteration by sparrow search algorithm (SSA), a chaos sparrow search algorithm is proposed based on hierarchy and Brownian motion (CSSA HB). Firstly, chaotic map is introduced for adjusting the key parameters of SSA. Secondly, a hierarchy is introduced. The three best individuals of the parent population are used to update the position of the vigilantes, enhancing communication among individuals and increasing population diversity. The uniform step controlled by Brownian motion is used to enhance the exploration ability of the algorithm. When the algorithm is stagnant, the Brownian motion strategy is used to disturb the individual to urge the algorithm to get rid of the local optimum. Finally, the greedy strategy is used to retain the dominant individuals and effectively accelerate the convergence speed. 12 test functions are made by the simulation experiments. The results show that the chaotic maps can enhance effectively the performance of the algorithm, and make the iterative maps perform the best. The improved algorithm is stronger in local optimum avoidance ability, faster at convergence speed, and higher in convergence accuracy.