摘要
为提高粒子群优化算法的收敛速度和求解精度,本文基于无视速度影响的简化粒子群优化算法,引入随迭代次数自适应调整的非线性惯性权重和异步学习因子,以此平衡粒子的全局搜索和局部开发能力。同时融合遗传算法的精英保留策略,确保每一代进化中最佳个体得以保留,助力粒子逃离局部最优。最后,通过5种测试函数比较了基本粒子群优化算法、本文改进算法以及其他经典改进算法的性能,实验证明,本文改进算法在收敛速度和求解精度等方面有显著的提升。
To enhance the convergence speed and solution accuracy of the Particle Swarm Optimization(PSO)algorithm,this study proposes an improved PSO algorithm based on the simplified PSO algorithm that disregards velocity influences.The proposed approach incorporates a nonlinear inertia weight and an asynchronous learning factor,both adaptively adjusted with respect to the iteration count,aiming to balance the global exploration and local exploitation capabilities of particles.Additionally,the algorithm integrates the elite retention strategy from genetic algorithms to ensure the preservation of the best individual in each generation,facilitating particle escape from local optima.Finally,the performance of the basic PSO algorithm,the proposed improved algorithm,and other classical enhancement algorithms are compared by five benchmark test functions.Experimental results show that the improved algorithm significantly improves the convergence speed and solution accuracy.
作者
马钰
魏文红
MA Yu;WEI Wenhong(School of Computer Science and Technology,Dongguan University of Technology,Dongguan 523808,China)
出处
《东莞理工学院学报》
2025年第1期41-47,共7页
Journal of Dongguan University of Technology
关键词
简化粒子群优化算法
非线性惯性权重
非线性异步学习因子
群体智能
simplified particle swarm optimization
nonlinear inertial weights
nonlinear asynchronous learning factor
swarm intelligence
