刘国志. 基于信赖域技术的局部收缩的微粒群算法[J]. 云南大学学报(自然科学版), 2008, 30(1): 21-26.
引用本文: 刘国志. 基于信赖域技术的局部收缩的微粒群算法[J]. 云南大学学报(自然科学版), 2008, 30(1): 21-26.
LIU Guo-zhi. A new local constriction particle swarm optimization algorithm based on trust region searching[J]. Journal of Yunnan University: Natural Sciences Edition, 2008, 30(1): 21-26.
Citation: LIU Guo-zhi. A new local constriction particle swarm optimization algorithm based on trust region searching[J]. Journal of Yunnan University: Natural Sciences Edition, 2008, 30(1): 21-26.

基于信赖域技术的局部收缩的微粒群算法

A new local constriction particle swarm optimization algorithm based on trust region searching

  • 摘要: 为了改善标准的微粒群优化算法(SPSO)的性能,给出一个新的速度更新策略——局部收缩策略,且把信赖域技术引入PSO算法中进行惯性权重的动态调整,提出一个新的微粒群优化算法——基于信赖域技术的局部收缩的微粒群算法.新算法(NPSO)保持了PSO算法结构简单的特点,改善了PSO算法的全局寻优能力,提高了算法的收敛速度和计算精度.利用10个测试函数测试新算法的性能,并分别与SPSO、与混沌相结合的微粒群算法(PSOC)、具有被动聚集的微粒群算法(PSOPC)、SPSO的全局版本及带有收缩因子的微粒群算法(CPSO)比较,实验结果表明,新算法(NPSO)大大地改善了实例测试函数的表现.

     

    Abstract: Particle swarm optimization(PSO) algorithm is one of the most powerful methods for solving unconstrained and constrained global optimization problems.Thinking of PSO algorithm is easy to trap into local minima in solving multimode function,it is incorporated new update velocities and trust region technique into the PSO algorithm,and proposed a new particle swarm optimization algorithm(NPSO).The proposed algorithm has not only the simple for implement of algorithm PSO,also the fast convergence and high computational precision.Simulation results and comparisons with the PSOC,standard PSO and CPSO show that the NPSO can effectively enhance the searching efficiency and greatly improve the searching quality.

     

/

返回文章
返回