基于改进加权Voronoi图和遗传算法的变电站规划  

Optimal Substation Planning Based on Improved Weighted Voronoi Diagram and Genetic Algorithm

作　　者：曹昉[1] 孟琦斌 苗培青 李越 

机构地区：[1]华北电力大学电气与电子工程学院,北京市昌平区102206 [2]国网山东省电力公司经济技术研究院,山东省济南市250021

出　　处：《电网技术》2015年第2期511-516,共6页Power System Technology

基　　金：国网山东省电力公司经济技术研究院项目(SDJYY20130301)

摘　　要：针对城市电网规划中的变电站规划优化问题,提出一种基于改进加权Voronoi图和遗传算法的规划算法(improved weighted Voronoi diagram-genetic algorithm,IWVD-GA)。通过改进定权重和引入变权重对加权Voronoi图进行改进,并利用元胞自动机(cellular automata,CA)动态生成改进加权Voronoi图,在生成过程中,权重可根据变电站负载率和供电半径约束动态调整,使供电范围的划分过程可控且更为合理。通过交替定位算法(alternate location algorithm,ALA)对新建站的位置和供电范围进行局部深度寻优,通过遗传算法进行全局寻优,实现了变电站规划优化。为解决CA在元胞个数较多的情况下,在每个演化过程都需要对所有元胞进行判定而导致计算速度较慢的问题,根据元胞状态改变发生在扩张边界的特点,确定需要进行判定的元胞,有效减少了元胞判定次数。算例结果验证了所提算法用于变电站规划的有效性。In allusion to the optimal substation planning problem in urban network planning, a method based on Improved Weighted Voronoi Diagram and Genetic Algorithm （IWVD-GA） is established. The weighted Voronoi diagram is improved by improving the fixed weight and introducing the variable weight, and the Cellular Automata（CA） is proposed to generating the improved Voronoi diagram, during which the weights can be dynamically adjusted according to the constraints of substation load-rate and power supply radius, then the process of dividing power supply areas is controllable and more reasonable. The new substation sites and power supply areas divisions are locally optimized in depth by Alternate Location Algorithm（ALA）, and globally optimized by Genetic Algorithm（GA）. To solve the problem of the low speed of CA caused by judging all the cellulars in every evolution process when there are many cellulars, according to the property that the cellular state transitions appear in the expanding boundary, the cellulars which need to be judged can be determined, thus reducing the number of cellular judging. The results verified the efficiency of the proposed method.

关 键 词：遗传算法 加权Voronoi图 变权重 元胞自动机 变电站规划 

分 类 号：TM721[电气工程—电力系统及自动化]