基于竞争侵略策略的粒子群算法

Particle Swarm Optimization based on competitive aggression Strategy

针对粒子群优化算法(Particle Swarm Optimization,PSO)容易陷入局部最优及算法收敛速度慢的问题,提出基于竞争侵略策略的粒子群算法(Particle Swarm Optimization based on competitive aggression Strategy,CAPSO)。凭借高自由度的侵略性,自由粒子群中粒子与择优进化群中粒子进行比较,得出竞争最优,进而通过竞争池指导择优进化群更新,使CAPSO算法快速跳出局部最优,且提高算法收敛速度。使用8个标准测试函数分别对4个算法以及CAPSO算法进行仿真,对寻优结果进行分析。结果表明,CAPSO算法无论是在单峰函数问题还是多峰函数问题上,总体拥有优秀的寻优结果。

In order to solve the problem that particle swarm optimization(PSO) algorithm is easy to fall into local optimization and slow convergence speed,a particle swarm optimization algorithm based on competitive aggression strategy is proposed.By comparing aggressive free particles with high degrees of freedom with optimized particle swarm optimization particles,competitive optimization is obtained,and then particle swarm optimization update is guided by competitive pool,which makes CAPSO algorithm jump out of local optimization quickly.And improve the convergence speed of the algorithm.Eight standard function tests are used to simulate the four algorithms and the CAPSO algorithm,and the optimization results are analyzed.The results show that the CAPSO algorithm has excellent optimization results in both unimodal function problems and multi-peak function problems.