基于改进的小生境粒子群算法在函数优化中的应用

摘要

为了克服标准粒子群算法在搜索后期中易陷入局部最优等缺点,提出了一种改进的小生境粒子群算法.通过将小生境技术引入标准粒子群算法中,保证了种群的多样性;同时在惯性权重中引入余弦函数,更改算法中认知项和社会项加速因子,加入迭代因素,并在位置更新策略中加入了飞行时间因子等策略,使其更加贴近粒子群算法的客观规律.通过对5个非线性基准测试函数进行数值仿真实验对比,结果表明改进的小生境粒子群算法在非线性的复杂函数优化中具有更好的寻优能力,避免了"早熟"现象,同时还具备收敛速度快,搜索精度高等特点.%In order to overcome the disadvantages of the standard particle swarm optimization(PSO)in the late search, an improved niche particle swarm optimization(PSO)algorithm is proposed.The niche technology is introduced into the standard particle swarm algorithm to ensure the diversity of the population.At the same time the inertia weight is introduced into cosine function,and the algorithm in cognitive and the social accelerated factors are changed,with iterative factors.And in the location update strategy,the flight time is joined for sub strategies,which is more closed to the objective law of particle swarm optimization algorithm.Through numerical simulation comparison of 5 nonlinear benchmark functions,results show that the niche PSO has better optimization ability in complex nonlinear function optimization,avoiding the"premature"phenomenon,and fast convergence speed and search precision.