为了改善粒子群算法求解多峰优化问题的能力,提出了一种新的利用粒子在种群中序号来构造Species的方法,同时还设计了适应性局域搜索算子和触发式重新初始化策略分别用于增强算法的开发能力和探索能力.通过对一组标准多峰测试函数的仿真实验,表明所提出的算法能够有效的求解多峰优化问题. In order to improve the ability of Particle Swarm Optimization (PSO) to solve multi-modal optimization problems, a new method of constructing Species using population numbers in a population is proposed. At the same time, adaptive local search operators and triggered reinitialization strategies To enhance the ability of development and exploration of the algorithm.Through a series of standard multi-peak test function simulation experiments show that the proposed algorithm can effectively solve the multi-peak optimization problem.