摘要
在简单比较遗传算法、模拟退火算法、Tabu算法、传统粒子群无功优化算法的基础上,提出一种改进的粒子群无功优化算法即自适应强引导粒子群的电力系统无功优化算法,该算法在粒子群算法的基础上引入强引导思想,在搜索初期,利用数学中的外推技巧对粒子位置的更新加以引导,减少算法随机性,利用群体适应度方差判别种群的多样性,并相应调整变异概率作出变异判别实现自适应更新粒子速度和位置,提高全局和局部搜索能力,进一步解决寻优后期粒子可能陷入早熟收敛的问题,可以更有效地搜索到全局最优解。通过对福建某高压配电网进行无功优化,本文算法优化后最优降损率可达15.3%,最低电压从0.8950pu提高到0.9973pu,结果表明本文算法及模型的可行性和有效性。
Reactive power optimization in power system is the effective means to improve the vohage quality, re- duce system losses and ensure secure and economic operation of system. Based on analysis and comparison of genetic algorithm, simulated annealing algorithm, Tabu algorithm, and traditional particle swarm optimization algorithm for reactive power,an adaptive induction-enhanced PSO is introduced to solve the problems of strong randomness and premature convergence in last period evolution of traditional PSO in this paper. During the early search, the strategy of extrapolation in mathematics is used to update the particle position for decreasing the randomness of PSO, and colony fitness variance is used to judge the population diversity. Mutation probability is adjusted for making variant judgment to update the particle position and speed adaptively in order to improve the capability of local and global search. Calculation of a power distribution system in Fujian province shows that, the validity and feasibility of the proposed algorithm is proved. The optimal loss-lowering rate is 15.3% , and minimum voltage increases from 0. 8950pu to 0. 9973pu.
出处
《电工电能新技术》
CSCD
北大核心
2012年第4期24-28,38,共6页
Advanced Technology of Electrical Engineering and Energy
基金
福州大学科技发展基金资助项目(600110)
关键词
粒子群
强引导
自适应变异
无功优化
particle swarm optimization
induction-enhanced
adaptive mutation
reactive power optimization
