采用收缩因子的粒子群优化算法

Particle Swarm Optimization Algorithm Using Shrinkage Factor

智能算法在解决函数优化问题时往往存在早期收敛和陷入局部极值等问题。本文提出一种基于收缩因子和自适应惯性权重的粒子群优化算法,对粒子群的速度更新公式进行经典的自适应惯性权重调节,同时采用收缩因子对整个速度向量进行精确压缩,使搜索出的解趋近更优的位置,提升算法的收敛性能。通过4个经典函数的仿真实验测试,结果表明本文提出的KPSO算法具有很好的搜索效果和寻优能力。

Intelligent algorithms often have problems such as early convergence and falling into local optima when solving function optimization problems.In this paper,a particle swarm optimization algorithm based on shrinkage factor and adaptive inertia weight is proposed.The speed update formula of the particle swarm is adjusted by classical adaptive inertia weight.At the same time,a shrinkage factor is used to accurately compress the entire velocity vector to make the searched solution approach better solution,which improves the convergence performance of the algorithm.Simulation experiments were carried on four classic functions.Simulation results show that the proposed algorithm has good search effect and optimization ability.