基于反向策略和柯西分布的粒子群优化算法

Particle Swarm Optimization Algorithm Based on Reverse Strategy and Cauchy Distribution

针对粒子群算法容易陷入局部最优的特点,采用混沌和反向策略产生初始解、用柯西密度函数和分布函数分别对惯性权重和位置更新公式进行改进.实验首先测试单一改进的有效性,接着融合多种策略,并对4个经典的测试函数进行实验,结果显示改进算法求解精度高、解的稳定性优良,特别是在多峰值函数中表现优越.

According to the characteristics that particle swarm algorithm is easy to fall into local optimal solution, the chaos and the reverse strategy of initial solution, Cauehy density function and distribution function are used to improve the inertia weight and the location update formula. The first test is carried out on the validity of the single improvement, then the integration of a variety of strategies, and experiments on 4 classical test functions are conducted. The results show that the improved algorithm has high accuracy and good stability of solutions, especially in the muhi peak function performance.