分布式电源优化配置与配电网重构

发布时间：2018-10-11 07:48

【摘要】：分布式电源（Distributed Generation, DG）作为传统供电方式的补充，以其就地消纳电力，灵活高效发电，减少环境污染等优点在电力系统中扮演着重要角色。DG接入配电网后，使得配电网的结构和运行方式发生改变，原有的网络潮流、网损和节点电压都将受到影响。配电网重构通过开合开关以改变网络拓扑，可有效降低网络损耗并提高电压质量，保证配电网经济运行。因此对并网分布式电源的容量与位置进行有效规划，并利用网络重构实现含分布式电源配电网的经济安全运行具有重要的理论与现实意义。本文主要针对分布式电源优化配置和含分布式电源的配电网重构两个问题进行了相关研究。在对风力发电、光伏发电、燃料电池、微型燃气轮机四种典型DG的运行方式和控制机理分析的基础上，将DG分成PQ、PI、PV、PQ(V)四种节点类型处理，建立了相应的配电网潮流计算模型，给出了含DG的节点分层前推回代潮流计算方法，并对潮流算法的性能和含各种类型DG的潮流计算进行仿真分析。建立了以网络有功损耗和静态电压稳定性为双目标的分布式电源优化配置模型，采用量子粒子群算法进行求解。考虑DG的安装位置与容量限制，提出区域优化的思想，，实现了不同节点类型DG的区域优化配置，并在IEEE33节点系统上进行了仿真测试。另外，在PGE69节点系统上对考虑DG数量限制的不同类型DG区域优化配置和DG位置恒定优化注入容量两种优化情况进行了仿真测试，并对单目标与双目标优化配置性能作了比较分析，证明了双目标区域优化配置方法的正确性与有效性。建立了以有功损耗最小的含分布式电源的配电网重构模型。采用整数型环网编码，对量子粒子群算法的变异策略进行改进，应用改进后的整数编码型量子粒子群算法进行求解，提高了收敛速度与寻优效率。为消除粒子寻优过程中产生的不可行解，提出了开关-环路矩阵与节点分层策略的不可行解判断方法，该方法运用潮流算法的分层思想，可以达到完全消除不可行解的目的。在IEEE33和PGE69节点系统上对不含DG和含四种类型DG的重构进行多种情况的仿真测试，并分析了PV节点类型DG的额定电压设置对于重构结果的影响。为使配电网性能达到整体最优，对可调度类DG进行配电网重构和DG注入功率的综合优化，进一步降低了网络的有功损耗，提高了全网电压水平。
[Abstract]:As a supplement of the traditional power supply mode, (Distributed Generation, DG) plays an important role in the power system because of its advantages of local absorption, flexible and efficient generation, reducing environmental pollution, etc. After the DG is connected to the distribution network, it plays an important role in the power system. The structure and operation mode of distribution network are changed, and the original network power flow, network loss and node voltage will be affected. Distribution network reconfiguration can effectively reduce the network loss and improve the voltage quality by switching on and off to change the network topology so as to ensure the economic operation of the distribution network. Therefore, it is of great theoretical and practical significance to plan the capacity and position of distributed power supply and to realize the economical and safe operation of distribution network with distributed power supply by network reconfiguration. In this paper, the optimal configuration of distributed power and the reconfiguration of distribution network with distributed power are studied. On the basis of analyzing the operation mode and control mechanism of four typical DG such as wind power generation, photovoltaic power generation, fuel cell and micro gas turbine, the DG is divided into four kinds of node types of PQ,PI,PV,PQ (V), and the corresponding power flow calculation model of distribution network is established. In this paper, the method of hierarchical forward power flow calculation with DG is presented, and the performance of power flow algorithm and power flow calculation with various types of DG are simulated and analyzed. The optimal configuration model of distributed power supply based on active power loss and static voltage stability is established and solved by quantum particle swarm optimization (QPSO). Considering the location and capacity limitation of DG, the idea of zone optimization is put forward, and the optimal configuration of DG with different node types is realized, and the simulation test is carried out on the IEEE33 node system. In addition, the optimal configuration of different types of DG region considering the number of DG and the optimal injection capacity of DG position are simulated on the PGE69 node system. The performance of single objective and double objective optimal configuration is compared and analyzed, and the correctness and effectiveness of the method are proved. A distribution network reconfiguration model with minimum active power loss is established. The mutation strategy of Quantum Particle Swarm Optimization (QPSO) is improved by using integer loop coding, and the improved integer coded QPSO algorithm is applied to solve the problem. The convergence speed and optimization efficiency are improved. In order to eliminate the infeasible solution in the process of particle optimization, a method for judging the infeasible solution of the switching loop matrix and the node delamination strategy is proposed. This method can eliminate the infeasible solution completely by using the hierarchical thinking of the power flow algorithm. The reconfiguration without DG and four types of DG is simulated and tested on IEEE33 and PGE69 node systems, and the effect of the rated voltage of PV node type DG on the reconstruction results is analyzed. In order to achieve the overall optimal performance of the distribution network, the reconfiguration of the distribution network and the comprehensive optimization of the injection power of the DG are carried out for the dispatchable DG, which further reduces the active power loss of the network and improves the voltage level of the whole network.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM73

【参考文献】