含分布式电源的配电网重构及故障恢复研究

发布时间：2018-03-28 04:38

本文选题：分布式电源　切入点：潮流计算　出处：《上海电力学院》2016年硕士论文

【摘要】：配电网是连接输电网与电力用户的中间桥梁,配电网稳定、可靠、经济运行对整个电力系统的稳定性具有重要的影响。配电网重构是配电网供电路径自我优化和自我治愈能力的体现,是正常运行时降低网络损耗、提高电能质量以及故障后快速恢复供电,保持供电连续性的重要保证。分布式电源的接入改变了传统配电网潮流模式,将改变配电网的网损和节点电压。因此,如何通过配电网的网络重构优化含有DG的配电网的供电路径,提高系统运行经济性、稳定性以及解决故障后的供电恢复问题具有重要的现实意义。本文针对含有DG的配电网重构优化和故障恢复策略两个方面进行研究,主要研究工作有:建立各类型DG的发电功率模型和节点类型,根据潮流计算中处理一些节点的困难之处,用无功补偿方法和无功转换的方式将其他类型节点进行转换,针对潮流计算过程中不确定性因素的影响,提出了基于两点估计的概率计算,提高潮流计算的效率和准确度。在供电路径优化重构时,根据分布式电源出力变化分为静态重构和动态重构策略,通过对初始复杂的配电网进行拓扑简化处理,简化计算过程,引入无重复生成树策略,从简化网络的基本树出发,通过树支与连支的交换生成不重复的辐射型网络结构,无需在寻优中进行重复性的辐射型约束校验,减少解的冗余度,提高供电路径寻优的效率;并与新型群智能动态优化算法-改进“教与学”优化算法相结合,提高动态优化的全局寻优能力。在动态寻优时,由于分布式电源出力随时间发生不确定性变化,在构造的重构概率场景模型基础上,提出了基于时序曲线划分多场景的分析法来分割分布式电源出力,该方法能够很好的适应出力具有间歇性的风机、光伏对配电系统影响,符合工程实际的运用,为当今智能配电网的供电路径实时寻优提供一定理论指导。在配电网故障恢复的过程中,考虑多目标、多约束的情况,与配电网整体运行特性相适应,具有全面性,针对孤岛划分问题,采用两阶段方法进行处理,CSP模型可以灵活的运用于较为复杂的多点电源接入的情况,并应用到孤岛划分第一阶段寻找可行解,第二阶段根据电压指标评价模型对初始方案进行评优修正来求最优解;最后对孤岛外的剩余网络模型进行开关组合寻优,求得最终的含多类型DG接入的配电网故障恢复方案。采用IEEE33节点系统和69节点系统对上述所提方法进行算例验证,结果表明所提方法的有效性、可行性。
[Abstract]:Distribution network is a bridge between transmission network and power users. Distribution network is stable and reliable. Economic operation has an important effect on the stability of the whole power system. The reconfiguration of distribution network is the embodiment of self-optimization and self-healing ability of power supply path of distribution network and the reduction of network loss during normal operation. It is important to improve the power quality and restore the power supply quickly after the fault and maintain the continuity of the power supply. The access of the distributed power supply changes the traditional power flow mode of the distribution network and will change the network loss and node voltage of the distribution network. How to optimize the power supply path of the distribution network with DG through the network reconfiguration of the distribution network, so as to improve the operation economy of the system, It is of great practical significance to solve the problem of power supply recovery after failure. In this paper, the reconfiguration optimization of distribution network with DG and the strategy of fault recovery are studied. The main research work is as follows: the generation power model and node type of DG are established. According to the difficulties of dealing with some nodes in power flow calculation, the reactive power compensation method and reactive power conversion are used to convert other types of nodes. In order to improve the efficiency and accuracy of power flow calculation, a probability calculation based on two-point estimation is proposed for the influence of uncertain factors in the process of power flow calculation. According to the dynamic and static reconfiguration strategies of distributed power generation, the topology simplification of the initial complex distribution network is carried out to simplify the calculation process, and the no-repetition spanning tree strategy is introduced to simplify the basic tree of the network. Through the exchange of tree branches and connected branches, the non-repeated radiation type network structure is generated, which does not need to carry out the repetitive radiation type constraint check in the optimization, reduces the redundancy of the solution, and improves the efficiency of power supply path optimization. Combined with the new swarm intelligence dynamic optimization algorithm, the improved "teaching and learning" optimization algorithm is used to improve the global optimization ability of dynamic optimization. In dynamic optimization, the output force of distributed power supply changes with the uncertainty of time. On the basis of the reconstruction probability scenario model, an analysis method based on time series curve partitioning multi-scene is proposed to divide the distributed power generation. This method can adapt to the intermittent fan and the influence of photovoltaic on the distribution system. In accordance with the practical application of engineering, this paper provides some theoretical guidance for real-time optimization of power supply path of smart distribution network. In the process of fault recovery of distribution network, the situation of multi-objective and multi-constraint is considered, which adapts to the overall operating characteristics of distribution network. Because of its comprehensiveness, the two-stage method can be used to deal with the problem of island partition, which can be flexibly applied to the complicated multi-point power access, and to the first stage of islanding to find a feasible solution. In the second stage, according to the voltage index evaluation model, the initial scheme is evaluated and modified to find the optimal solution. Finally, the residual network model outside the island is optimized by switching combination. Finally, the final fault recovery scheme of distribution network with multi-type DG access is obtained. The IEEE33 node system and 69 bus system are used to verify the proposed method. The results show that the proposed method is effective and feasible.
【学位授予单位】：上海电力学院
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM711

【相似文献】