基于GA-ACO算法的含分布式电源的配电网重构研究

发布时间：2018-02-13 05:21

本文关键词： 配电网重构分布式电源多种负荷方式蚁群算法遗传算法潮流计算　出处：《南京理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：配电网重构是提高电力系统运行的安全性和稳定性的有效方法。而分布式电源(Distributed Generation, DG)的接入,使得原有单电源的网络变成了多电源的网络,同时DG的容量和安装位置直接关系到配电网的电能质量、电压分布、短路电流大小以及线路的潮流分布等,带来的后果就是原有网络重构的目标函数,约束条件不再适用于现有网络,因此研究含DG的配电网重构具有十分重要的意义。本文首先对DG做了一些等效和简化处理,并以简单配电网为例,严格论证和推导了不同容量的DG出力、DG在系统中接入位置对于系统网损的影响。在潮流计算中,依据PV节点、PI节点、PQ(V)节点所具有的不同特点分别建立了相应的潮流计算模型,还提出了一种考虑最大负荷,一般负荷,最小负荷的3种负荷方式的重构模型,在每一种负荷方式下,都进行一次潮流计算,这样使得重构方案更加接近最优解。本文以基本环路中打开的开关为基因位构成染色体,染色体数即为基本环路数,这样可以极大程度的缩小搜索空间,提高算法的搜索效率。通过对遗传算法和蚁群算法优缺点的比较,提出了遗传算法与蚁群算法动态融合的新算法。主要原理是首先利用遗传算法的快速全局搜索能力生成信息素的分布,然后利用蚁群算法的正反馈收敛机制寻找最短路径,求精确解。遗传算法中交叉率PC和变异率Pm的选择和调节可以更快速的选择最优路径,提高算法的收敛代数,在配电网中表现为快速的逃离不可行区域,蚁群算法的正反馈机制快速的使得优质染色体的比例明鲜增加,解的质量得到了极大的提高,而且蚁群算法的融入加快了收敛速度；同时蚂蚁以微小的出错概率从原有配电网中进行选择,从而保证了解的多样性,有效的防止GA-ACO的早熟。通过对IEEE16、 IEEE33节点的仿真,验证了上述算法的可以性。
[Abstract]:Distribution network reconfiguration is an effective method to improve the security and stability of power system. At the same time, the capacity and installation position of DG are directly related to the power quality, voltage distribution, short circuit current and power flow distribution of the distribution network. The result is the objective function of the original network reconfiguration. The constraints are no longer applicable to existing networks, so it is very important to study the reconfiguration of distribution networks with DG. In this paper, we first do some equivalent and simplified treatment to DG, and take simple distribution network as an example, we strictly demonstrate and deduce the influence of DG's access position in the system with different capacity on the system's network loss. According to the different characteristics of PV / Pi node / PQ / V) node, the corresponding power flow calculation models are established, and a reconstruction model of three load modes considering maximum load, general load and minimum load is proposed. In each load mode, a power flow calculation is carried out, which makes the reconstruction scheme closer to the optimal solution. In this paper, the open switch in the basic loop is used as the gene site to construct the chromosome, and the number of chromosomes is the number of the basic loops, which can greatly reduce the search space. Improve the search efficiency of the algorithm. By comparing the advantages and disadvantages of genetic algorithm and ant colony algorithm, A new algorithm for dynamic fusion of genetic algorithm and ant colony algorithm is proposed. Firstly, the fast global search ability of genetic algorithm is used to generate the distribution of pheromone, and then the convergence mechanism of ant colony algorithm is used to find the shortest path. The selection and adjustment of crossover rate PC and mutation rate pm in genetic algorithm can select the optimal path more quickly, improve the convergence algebra of the algorithm, and quickly escape the infeasible region in the distribution network. The positive feedback mechanism of ant colony algorithm can increase the proportion of high quality chromosome and improve the quality of solution greatly, and the integration of ant colony algorithm accelerates the convergence speed. At the same time, ants select from the original distribution network with small error probability, so as to ensure the diversity of understanding, and effectively prevent the precocity of GA-ACO. The simulation of IEEE 16 and IEEE33 nodes verifies the feasibility of the above algorithm.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM732

【参考文献】