含分布式电源的配电网重构与故障恢复

发布时间：2018-06-09 11:29

本文选题：分布式电源 + 配电网重构　；参考：《湖南大学》2015年硕士论文

【摘要】：分布式电源(Distributed Generation,DG)作为集中式发电的有益补充,因具有降低网络损耗,改善电压水平、减少环境污染等优点而被广泛研究。DG接入配电网后,使得配电网的拓扑结构与潮流分布发生改变。作为优化配电网运行的重要手段之一,配电网重构可分为正常运行状态下的网络优化重构和故障情况下的恢复重构。配电网重构通过改变开关状态来改变网络拓扑结构,实现配电网的经济、可靠运行。因此,研究含分布式电源的配电网重构与故障恢复具有重要的理论意义和实际意义。配电网潮流计算是配电网运行分析的基础,大量DG接入配电网后使得潮流计算更为复杂。针对传统的潮流计算未考虑负荷的电压特性,建立了计及负荷静态电压特性的含DG配电网潮流计算模型。通过对不同运行控制方式下的DG接口类型进行分析,将DG分为PQ、PQ(V)、PI和PV四种节点类型处理。采用基于支路电流和节点电压的前推回代潮流方法对计及静态电压特性的含DG潮流计算模型进行求解。以IEEE33节点系统验证了含DG配电网潮流算法的稳定性和有效性,为下一步的配电网重构做好准备。建立了计及负荷电压特性的含DG配电网重构模型,以二进制编码的量子遗传算法对重构模型进行求解。改进了量子自适应动态旋转、量子交叉、量子变异等进化操作,同时采用环路矩阵分层编码策略有效地避免了不可行解,使量子遗传算法能够准确、快速搜索到最优解。以IEEE33节点系统和PGE69节点系统为仿真算例,对不含DG、含单一类型DG和多类型DG的配网重构进行了验证分析,结果表明了所提算法和模型的有效性和稳定性。针对故障状态下的配电网,提出了一种含DG孤岛划分策略的配电网故障恢复性重构算法。将配电网故障恢复分为孤岛划分和孤岛外重构两个子问题求解。建立考虑负荷优先级和等效电气距离的孤岛划分模型,以具有独立支撑能力的DG为中心对失电负荷进行孤岛划分。孤岛外的配电网系统恢复重构过程中考虑线路损耗和开关次数的综合模型,应用改进的量子遗传算法对孤岛外的配电网系统恢复重构模型进行求解,以IEEE33节点系统为算例进行测试,结果表明该方法有利于保证配电系统可靠、经济运行。
[Abstract]:Distributed Generation (DG), as a useful supplement to centralized generation, has been widely studied after it is connected to the distribution network because of its advantages of reducing network loss, improving voltage level and reducing environmental pollution. The topology and power flow distribution of distribution network are changed. As one of the important means to optimize distribution network operation, distribution network reconfiguration can be divided into network optimization reconfiguration under normal operation and restoration reconfiguration under fault condition. Distribution network reconfiguration changes the topology of distribution network by changing the switching state, and realizes the economic and reliable operation of distribution network. Therefore, it is of great theoretical and practical significance to study the reconfiguration and fault recovery of distribution network with distributed generation. Distribution network power flow calculation is the basis of distribution network operation analysis, a large number of DG connected to the distribution network makes power flow calculation more complex. In view of the traditional power flow calculation without considering the voltage characteristics of the load, a power flow calculation model with DG distribution network is established, which takes into account the static voltage characteristics of the load. Based on the analysis of the DG interface types under different operation control modes, DG is divided into four node types: PQQ, PQV, Pi and PV. The DG power flow calculation model with static voltage characteristics is solved by the forward pushback power flow method based on branch current and node voltage. The stability and effectiveness of the power flow algorithm with DG are verified by IEEE33 bus system, which is ready for the next reconfiguration of distribution network. A DG distribution network reconfiguration model with load voltage characteristics is established and solved by binary coded quantum genetic algorithm (QGA). Quantum adaptive dynamic rotation, quantum crossover, quantum mutation and other evolutionary operations are improved. At the same time, loop matrix layered coding strategy is used to effectively avoid the infeasible solution, so that the quantum genetic algorithm can find the optimal solution accurately and quickly. Taking IEEE 33-bus system and PGE69-bus system as simulation examples, the reconfiguration of distribution network without DG, single type DG and multi-type DG is verified and analyzed. The results show the validity and stability of the proposed algorithm and model. A fault recovery reconstruction algorithm for distribution network with DG islanding strategy is proposed. The fault recovery of distribution network is divided into two sub-problems: islanding and external reconfiguration. An island partition model considering load priority and equivalent electrical distance is established. The isolated island partition of power loss load is carried out with DG with independent support capacity as the center. An improved quantum genetic algorithm (QGA) is applied to solve the restoration and reconfiguration model of distribution network outside the island considering the line loss and switching times in the restoration and reconstruction process of the distribution network system outside the island. Taking IEEE 33 bus system as an example, the results show that this method is helpful to ensure the reliable and economical operation of the distribution system.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM711

【参考文献】