摘要
机组组合优化问题是一个大规模、离散、非线性的混合整数规划问题,所以求解比较困难,不容易找到理论上的最优解。在基本粒子群算法的基础上,使用一种空间收缩策略,加快了算法的收敛速度。同时为了避免算法出现“早熟”现象,让粒子不仅根据自身和同伴中的最好个体进行调整自己的飞行速度,并且向其他个体学习。通过该算法进行仿真计算,证明了该算法的有效性。
Unit commitment is a large - scale, dispersed and mixed - integer non - linear programming problem, so an improved PSO mechanism is suggested to deal with the equality and inequality in the conventional PSO is preserved. And a dynamic search - space reduction strategy is used to accelerate the optimization process. Also adjustment operator and adjustment sequence are introduced to reconstruct PSO algorithm by using the ideas of single node regulating algorithm. Numerical simulation results show the effectiveness of the proposed method.
出处
《四川电力技术》
CAS
2007年第3期16-18,共3页
Sichuan Electric Power Technology
关键词
粒子群优化算法
机组组合
电力系统
particle swarm optimization
unit commitment
power system
