基于和声算法的含DG配电网故障定位研究

发布时间：2019-06-01 08:25

【摘要】：随着能源结构的变化和用户对电能需求的日益增加,具有清洁环保、能耗低、供电灵活等优势的分布式发电技术得到了快速的发展。但是随着分布式电源的接入,配电网由传统的单电源辐射状网络发展为多电源复杂的供电网络,潮流方向随机变化,给快速、准确地进行故障定位带来了新的问题,研究新的算法来实现含分布式电源配电网的故障定位对于提高供电可靠性,减小停电范围具有重要的实际意义。本文通过理论分析、应用Matlab/Simulink搭建含DG配网模型对分布式电源接入配电网后的故障定位算法进行研究。分析了分布式电源接入配电网后对短路电流的影响,得出分布式电源的接入对短路电流产生分流、助增和逆流等影响；建立了一种适应DG配电网复杂结构的开关函数模型,引入“树”的区域划分方法,剔除了无故障电流的无源树枝,降低了解的维数,提高了搜索效率；利用具有快速、准确特点的和声算法进行全局寻优,通过具体的算例分析,验证了该算法的准确性与快速性。本文对和声算法的收敛速度和容错性进行了研究。和声算法相比于传统的优化算法具有概念简单,容易实现等优点,但该算法的和声记忆库是随机产生的、具有不确定性,且对记忆库的初始化具有很强的依赖性。为了提高该算法的定位速度和容错性,利用具有高度并行、鲁棒性强等优点的遗传算法(GA)产生和声库的初始解；同时改进和声库中新解的产生方式,克服了对初始和声记忆库依赖性强和参数随机选择不确定性的弊端；最后利用Matlab编制相应算法的程序,分别对相同配电网模型进行寻优运算和分析比较,结果表明优化后的和声搜索算法能够有效地提高算法收敛速度和故障定位精度,在信息缺失、畸变等情况下具有良好的容错性。
[Abstract]:With the change of energy structure and the increasing demand of users for electric energy, distributed power generation technology, which has the advantages of clean and environmental protection, low energy consumption and flexible power supply, has been developed rapidly. However, with the access of distributed power supply, distribution network has developed from traditional single power radiation network to multi-power supply complex power supply network. The direction of power flow changes randomly, which brings new problems to fast and accurate fault location. It is of great practical significance to study a new algorithm to realize the fault location of distribution network with distributed power supply in order to improve the reliability of power supply and reduce the range of power outage. In this paper, based on the theoretical analysis, the fault location algorithm of distributed power supply connected to distribution network is studied by using Matlab/Simulink to build a distribution network model with DG. The influence of distributed power supply on short circuit current after being connected to distribution network is analyzed, and the influence of distributed power supply access on short circuit current shunt, auxiliary increase and countercurrent is obtained. A switching function model suitable for the complex structure of DG distribution network is established. The region division method of "tree" is introduced, the passive branches of fault-free current are eliminated, the dimension of solution is reduced, and the search efficiency is improved. The harmony algorithm with fast and accurate characteristics is used for global optimization. The accuracy and rapidity of the algorithm are verified by an example. In this paper, the convergence speed and fault tolerance of harmony algorithm are studied. Compared with the traditional optimization algorithm, the harmony algorithm has the advantages of simple concept and easy implementation, but the harmony memory database of the algorithm is random, uncertain and dependent on the initialization of the memory library. In order to improve the location speed and fault tolerance of the algorithm, the genetic algorithm (GA), which has the advantages of high parallelism and strong robustness, is used to generate the initial solution of the sound library. At the same time, the generation mode of the new solution in the sound database is improved, which overcome the disadvantages of strong dependence on the initial and acoustic memory database and uncertainty of random selection of parameters. Finally, the program of the corresponding algorithm is programmed by Matlab, and the optimization operation and analysis and comparison of the same distribution network model are carried out respectively. the results show that the optimized harmony search algorithm can effectively improve the convergence speed and fault location accuracy of the algorithm. It has good fault tolerance in the case of lack of information, distortion and so on.
【学位授予单位】：长沙理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM727

【参考文献】