摘要
针对函数优化的非线性特点,在标准粒子群优化算法的基础上,提出了一种带自适应变异的粒子群优化算法.该算法对惯性权值进行参数设计,建立非线性递减策略曲线模型,通过设置校准系数,改变惯性权值的曲线变化率,使其随迭代过程进行自适应变化.通过在迭代初期选取较大的惯性权值,增强算法的局部寻优能力,加快算法收敛速度,而在迭代后期选取较小的惯性权值,提升算法的全局搜索性能.同时,在算法中引入变异机制,增加种群的多样性,从而更好地提升算法由局部到全局的开放式搜索能力.通过选择基准测试函数对几种算法进行性能测试,证明改进算法收敛速度快、精度高,总体性能优于对比算法.
Based on the nonlinear feature of function optimization,this paper proposes a particle swarm optimization algorithm with adaptive variation which is on the basis of the standard particle swarm optimization algorithm. The algorithm designs the parameter of inertia weight,set up the nonlinear decreasing curve model,and alters the curve of the inertia weight rate by setting the calibration coefficient,and makes it changing adaptively with the iterative process. The algorithm can enhance local optimization ability and accelerate the convergence speed through selecting large inertia weight at the beginning of iteration,and improve global searching performance by selecting low inertia weight in the late iteration. At the same time,variation mechanism is introduced into this algorithm to increase the diversity of population,and improved the ability of the open search performance optimized from the partial to the whole The advantage of fast convergence and high precision in the improved algorithm has been proved by selecting benchmark functions and testing several comparing algorithms. The improved algorithm outperforms the compared algorithms.
出处
《洛阳师范学院学报》
2015年第11期9-12,26,共5页
Journal of Luoyang Normal University
基金
宿州学院一般科研项目(2014yyb03)
宿州学院科研平台开发课题(2014YKF44)
关键词
粒子群优化算法
惯性权值
非线性
自适应
变异
particle swarm optimization algorithm
inertia weight
nonlinear
adaptive
variation
