一种采用抽样策略的PSO 算法

原始粒子群优化算法(PSO) 和各种改进方法存在着参数取值固定、收敛精度低等问题. 为此, 提出一种采用抽样策略的粒子群优化算法(SS-PSO). 通过拉丁超立方抽样(LHS) 策略更新粒子速度和位置, 以加快收敛速度; 提出一种基于随机采样的最优位置修正方法, 以微调全局最优; 提出“双抽样”LHS 局部搜索方法, 以提高收敛精度. 与其他新近提出的两个算法进行对比, 结果显示SS-PSO 在一定程度上提高了算法的性能.

Particle swarm optimization algorithm via sampling strategy

There’re some issues such as fixed parameters’ value and easy to fall into local optimum in classical PSO algorithm and many improvements. Therefore, an improved PSO algorithm via sampling strategy(SS-PSO) is proposed. Firstly, replacement of the particles’ speed and location via Latin hypercube sampling(LHS) is proposed for speeding up the convergence process. Then, correction of the global best location via random sampling is proposed for fine tuning the global best location. Finally, “double sampling” LHS is proposed for local search to improve the convergence precision. Two new improvements are used to compare with the SS-PSO algorithm. The results show that the SS-PSO algorithm can improve the PSO’s algorithm performance in some extent.