Convergence Analysis of the Particle Swarm Optimization Based on Stochastic Processes
-
摘要: 分析了粒子群优化算法 (PSO) 的全局收敛性. 在已有文献的假设前提下和随机系统理论基础上, 对 PSO 进行算法分析推导, 给出了其动力学系统依均方收敛的一个充分条件, 从而有效地避免了已有文献基于线性时变离散系统研究 PSO 收敛性的不足. 通过对所得的粒子运行轨迹图和已有文献相比较, 得到了更好的结果和判据. 通过仿真实验分析研究, 验证了该结论的有效性.Abstract: This paper analyzes the global convergence of the particle swarm optimization algorithm (PSO). Based on the assumption of the previous articles and the theory of stochastic processes, this paper presents a sufficient condition for the system mean-square to be stable. The proposed condition overcomes the disadvantage of the previous studies on PSO for the linear time-varying discrete systems. Compared with the methods in the previous literature, the proposed method achieves better results. Simulations demonstrate the validity of the proposed method.
-
计量
- 文章访问数: 2549
- HTML全文浏览量: 58
- PDF下载量: 2305
- 被引次数: 0
下载:
