基于遗传算法和图论的配电网重构研究

发布时间：2018-01-18 04:34

本文关键词：基于遗传算法和图论的配电网重构研究　出处：《华北电力大学》2014年硕士论文　论文类型：学位论文

【摘要】：配电网络承担着将高压电力网络和用户连接在一起的任务。一般,配电网络的结构大多为闭式网络但其运行时则以开环网络运行,且一般具有多个开环点。由于电压等级相对较低,其运行时一般损耗较大。配电网重构则以及时更换配电开关的运行状态以达到优化配电网运行方式的目的。通常,配电开关有两类：区域联络开关和线路开关,通过开关运行状态的更换可以降低系统网络损耗并使各条线路的负荷率较为均衡,配电网重构(DNRC)是一个减少系统损失、平衡负载、隔离故障、恢复网络正常运行及减小运营成本等至关重要的方式。虽然以上目标可以通过其他不同途径实现,但是配电网重构是一个实现它们的通用途径。众所周知,馈线主体部分的重构是使系统损失最小化的最理想实现方式。论文研究配电网重构问题,具有显著地理论和现实意义。论文的主要研究工作和取得的创新成果如下：配电网重构的目的是为了在经济运行的方式下满足用户的用电需求,因此在对配网重构进行分析时有必要进行潮流计算。由于针对节点众多的配电网,高斯-赛德尔和牛顿-拉夫逊方法以及快速解耦法时常出现潮流计算结果不收敛的情况,在此情况下,由于良好的适应性,前推回代法成为一个较好的潮流计算工具被用于配电网潮流计算。很多研究者采用多种技术方法解决配电网络的重构问题：研究者们惯于使用传统方法、启发式技术、人工智能、进化编程等方式在不同程度上解决了该问题。然而,论文将图论和遗传算法结合起来,用于解决配电网重构问题。图论用于产生初始种群,以确保配电网络的辐射型结构,这可使配电网络的辐射状结构检查工作大大简化。遗传算法用于保证得到配网重构的可行解。本论文采用Matlab编程技术,同时选用IEEE33节点系统进行了算例分析。论文分别选取三种目标函数对算例进行求解：总网损最小、负荷均衡率最高以及将两个目标相结合。同时对结果进行了分析,分析结果显示了所提算法的有效性。
[Abstract]:The distribution network is responsible for the task of connecting the high-voltage power network with the users. In general, the structure of the distribution network is mostly closed network, but it operates as an open-loop network. And generally have a plurality of open-loop points, because the voltage level is relatively low. In order to optimize the operation mode of distribution network, there are usually two types of distribution switch: area contact switch and line switch, while the distribution network reconfiguration changes the operation state of distribution switch in time to achieve the purpose of optimizing the operation mode of distribution network. The switching operation state can reduce the system network loss and make the load rate of each line more balanced. The distribution network reconfiguration (DNRC) is a way to reduce the system losses, balance the load, and isolate the fault. Although the above objectives can be achieved through other different ways, distribution network reconfiguration is a common way to realize them. The reconfiguration of the main part of the feeder line is the most ideal way to minimize the system loss. This paper studies the reconfiguration of the distribution network, which has significant theoretical and practical significance. The main research work and the innovative achievements of the thesis are as follows: The purpose of distribution network reconfiguration is to meet the needs of users in the economic operation mode, so it is necessary to carry out power flow calculation in the analysis of distribution network reconfiguration, because of the distribution network with a large number of nodes. Gauss-Seidel and Newton-Raphson methods and fast decoupling methods often occur when the results of power flow calculation do not converge. In this case, due to good adaptability. The forward-push-back method has become a better power flow calculation tool for distribution network power flow calculation. Many researchers use a variety of techniques to solve the problem of distribution network reconfiguration: researchers are used to traditional methods, heuristic techniques, artificial intelligence. Evolutionary programming has solved the problem to some extent. However, this paper combines graph theory with genetic algorithm to solve the problem of distribution network reconfiguration, and graph theory is used to generate initial population. In order to ensure the radiative structure of the distribution network, the inspection of the radial structure of the distribution network can be greatly simplified. Genetic algorithm is used to ensure the feasible solution of distribution network reconfiguration. This paper adopts Matlab programming technology. At the same time, the IEEE33 node system is selected for the example analysis. Three kinds of objective functions are selected to solve the example: the total network loss is minimum. The load balancing rate is the highest and the two targets are combined. The results are analyzed and the results show the effectiveness of the proposed algorithm.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM732

【共引文献】