摘要
微粒群算法(PSO)来源于对社会模型的模拟,是一种随机全局优化技术。该算法简单,容易实现,且功能强大。文中对PSO进行了改进,引入了“分群”和“灾变”思想,并应用于求解水火电力系统的短期有功负荷最优分配问题。通过具体算例验证了改进PSO算法的有效性,而且其收敛速度比遗传算法(GA)快,求解精度比普通的:PSO和GA的高。
Particle swarm optimization (PSO) algorithm is a stochastic global optimization technique based on the simulation of social behavior. This method is easy to implement and possesses excellent performance. A refined particle swarm algorithm is proposed in which particles are divided into several clusters and the idea of 'Cataclysm' is led in. The refined PSO algorithm is applied to the optimal load distribution in short-term scheduling of hydro-thermal power systems. The effectiveness of the refined PSO algorithm is verified by practical calculation example, the convergence speed of refined PSO algorithm is faster than that of genetic algorithm(GA) and the solutions of refined PSO algorithm are more precise than those of common PSO algorithm and GA.
出处
《电网技术》
EI
CSCD
北大核心
2004年第12期16-19,共4页
Power System Technology
关键词
水火电力系统
短期发电计划
无功优化
短期发电计划优化
微粒群算法
Electric load distribution
Electric power systems
Genetic algorithms
Optimization
Stochastic control systems