摘要
粒子群优化算法是一种随机全局优化算法,但它容易陷入局部最优和早熟,为了克服其缺陷,本文提出了一种多样性驱动的自适应粒子群优化(DDA-PSO)算法。本算法包括吸引阶段和驱动阶段。吸引阶段利用惯性权重线性递减机制加快粒子收敛,驱动阶段利用多样性驱动速度策略提升种群多样性。两个阶段相互自适应转换,粒子能跳出局部最优和防止早熟,算法的勘探与开拓获得自适应平衡。DDA-PSO算法与其他已有算法进行了比较,实验结果表明,DDA-PSO算法提高了收敛速度和精度,全局搜索能力得到显著提高。
Particle swarm optimization algorithm is a stochastic global optimization algorithm,but it is easy to fall into local optima and premature.In order to overcome its defects,a diversity driven adaptive particle swarm optimization(DDA-PSO)algorithm is proposed in this paper.The algorithm includes the attraction stage and the driving stage.In the attraction stage,the linear decreasing inertia weight mechanism is used to accelerate the particle convergence,and in the driving stage,the diversity driving speed strategy is used to improve the population diversity.The two stages adaptively switch to each other,the particles can jump out of the local optima and prevent premature,the algorithm obtains an adaptive balance between the exploration and exploitation.The experimental results show that DDA-PSO algorithm improves the convergence speed and accuracy,and the global search ability is significantly improved.
作者
宗敏
杨玉群
徐刚
ZONG Min;YANG Yuqun;XU Gang(Department of Mathematics,NanchangUniversity,Nanchang 330031,China;Affiliated Middle School of Nanchang University,Nanchang 330047,China)
出处
《南昌大学学报(理科版)》
CAS
北大核心
2022年第4期386-391,共6页
Journal of Nanchang University(Natural Science)
基金
国家自然科学基金资助项目(61866023)。
关键词
粒子群优化
多样性
勘探与开拓
全局优化
Particle Swarm Optimization
Diversity
Exploration and exploitation
Global optimization