摘要
本文提出了一种量子粒子群算法解决电力系统无功优化问题。量子粒子群算法采用实数编码,融合了量子进化算法的概率性并进行计算和粒子群算法的更新策略,在全局寻优能力和保持种群多样性方面表现出了较大优势,而且寻优速度快。另外,为了避免大量不可行初始解,本文采用倾斜分布式启发方法得到初始种群。IEEE-30系统证明了量子粒子群算法的高效性和鲁棒性。
A new evolutionary algorithm called Quantum Particle Swarm Optimization Algorithm （QPSO） is proposed in this paper to solve reactive power optimization in power system. QPSO is based on real-number coding, which combines both probabilistic parallel computing of Quantum Evolutionary Algorithm （QEA） and update policy of Particle Swarm Optimization （PSO）. QPSO have greater superiority in global search ability and proper population multiplicity, as well as speedy convergence, In addition, an inclining initialization is adopted in order to avoid a great quantity of infeasible solutions. IEEE-30 system proves efficiency and robustness of QPSO.
出处
《电气技术》
2012年第2期15-19,共5页
Electrical Engineering
关键词
无功优化
量子粒子群算法
概率性
倾斜分布式启发
reactive power optimization
Quantum Particle Swarm Optimization Algorithm （QPSO）： probabilistic
inclining initialization
