基于随机生成树策略的配网重构遗传算法  

Genetic Algorithm Based on Random Spanning Trees in Distribution Network Reconfiguration

作　　者：欧阳武[1] 程浩忠[1] 张秀彬[1] 张节潭[1] 

机构地区：[1]上海交通大学电气工程系,上海200240

出　　处：《高电压技术》2008年第8期1726-1730,共5页High Voltage Engineering

基　　金：国家自然科学基金(50177017)~~

摘　　要：常规遗传算法求解配电网重构不能保证配电网的辐射状结构,会带来大量不可行解,而现有的改进编码方法和遗传策略操作复杂、计算量大、对复杂配电网并不十分有效。为了解决上述问题,引入了随机生成树策略,即根据支路序号的实数编码方式产生随机序列,再利用图论的"避圈法"产生对应的生成树,确保每个重构方案都是树状结构,使遗传进化在解空间内连续进行;同时为避免早熟,提高收敛速度,在变异操作中动态控制变异率。33节点和69节点典型算例验证了方法的有效性和良好的计算性能,表明方法求解迅速、简单有效。Conventional Genetic Algorithm （GA） can not ensure the radial structure of network in distribution network reconfiguration and it brings a large number of unfeasible solutions. Though some improved GA methods were proposed, the above-mentioned problem has not been completely solved due to their complicated operations and a large amount of calculations. In this paper, a new improved GA with the strategy of searching random spanning trees in distribution network reconfiguration is proposed. The real coding is adopted to produce random sequences and available structures of distribution network are then obtained using graph theory, which can guarantee the continuity of the GA evolution. Besides, the strategy of controlling the mutation rate dynamically is used in solving progress. Typical examples of 33-bus and 69-bus network illustrate the effectiveness and advantage of the method proposed.

关 键 词：遗传算法 编码 配网重构 支路 随机生成树 变异率 

分 类 号：TM744[电气工程—电力系统及自动化]