摘要
在基本遗传算法(SGA)的基础上,引入生物学中小生境的概念和正交设计技术,制定了初始种群生成方法,以保证个体的多样性,并在遗传算法选择、交叉和变异等操作中做出相应改进,从而形成可用于电力系统实时无功优化的小生境正交遗传算法。应用此算法时,通过采用一种典型小生境技术结合最优保存策略的小生境最优选择方法来进行种群的选择,减少了迭代过程中陷入局部最优解的可能性,并进一步减少迭代次数。仿真算例结果标明:在优化条件相同时,该算法的迭代次数明显少于基本遗传算法,显著提高了无功优化的收敛速度。
Based on BGA, introduce concept in middle small ecology environment of biology and orthogonal design technology, established primary racial mass to ensure individual diversity, and make out corresponding improvement by selecting genetic algorithm, operating in across and aberrance etc. thereby forming up the orthogonal genetic algorithm on small ecology environment for power system＇s real time reactive power optimization. Used the algorithm, the typical small ecology environment technology combined with best selecting way by excellent saving policy, which is adopted in this paper, it reduces possibility to run into best practical result during iterated process, and further more reduced reiteration time. Artificial calculating result indicates, under same optimization conditions, the reiteration time used the algorithm is less than Basic Genetic Algorithm obviously, it increases convergence speed in reactive power optimization remarkably.
出处
《广东输电与变电技术》
2009年第2期8-12,共5页
Guangdong Power Transmission Technology
关键词
电力系统
无功优化
小生境
遗传算法
正交设计
Power systems Reactive power optimization Niche Genetic algorithm Orthogonal design
