主动配电网多目标规划及控制分区研究

发布时间：2018-04-20 07:28

本文选题：主动配电网 + 分布式发电　；参考：《山东大学》2017年硕士论文

【摘要】：2008 年,国际大电网会议(International Council on Large Electric Systems,CIGRE)配电与分布式发电专委会(C6)C6.11项目组提出主动配电网(Active Distribution Network,ADN)的概念。AND可对接入配电网中的分布式发电(Distributed Generation,DG)系统进行控制,并采用灵活的网络拓扑对配电网潮流进行管理,实现电能的"主动"控制和"主动"分配。ADN的提出标志着配电网控制方式的转变,含有DG的ADN 一方面能应对高渗透率DG的接入,解决未来配电网中分布式能源的大规模接入问题;另一方面可以利用ADN的技术优势,在保证系统稳定运行的同时提高经济效益,配合并促进配电网的进一步发展。对ADN进行多目标规划有助于兼顾多方面指标,使配电网的运行达到最佳状态。从网络拓扑出发对其进行分区控制,一方面可缩小控制范围,提高系统可控性;另一方面可以应对不同种类的分布式能源接入,不需要建立过于复杂的系统模型,降低计算的复杂度,使系统保持良好的拓展性,对坚强电网的建设有着指导和借鉴意义。本文围绕ADN规划与控制分区研究,完成了以下工作:首先,本文系统阐述了 ADN的概念、结构、功能和管理模式,对其与被动配电网、有源配电网之间的联系与区别进行阐述;对DG给配电网在规划、运行及保护带来的影响进行分析,对含有DG的ADN建立系统论述,为下文的研究提供理论基础。其次,本文分析了现有的ADN规划的方法及存在的问题,建立统筹考虑可靠性与经济性的多目标分层规划模型,上层规划模型中以节点位置系数为衡量标准预选出分布式发电安装位置;下层规划模型中设定包含DG设备费用、网络损耗费用、年购电费用和政府补贴的目标函数,采用粒子群算法(Particle swarm optimization,PSO)对其进行计算,确定DG的安装位置以及安装容量,在保证系统供电可靠性的前提下对其进行经济优化,从而兼顾可靠性、经济性两方面指标。最后,本文采用复杂网络理论对ADN主动管理模式中的控制分区进行方法研究。定义电气距离并以此建立权重矩阵,为AND控制区域划分提供度量单位;以模块度为衡量标准对分区结果进行评估;采用PSO算法对其进行计算,针对粒子的编码与更新提出适应ADN网络拓扑特点的编码方式以及更新策略,从而加速收敛并节约储存空间。采用IEEE 33、69节点配电系统作为算例,验证本文中的多目标规划模型以及控制分区方法的有效性。ADN结构以及功能上的复杂性使得配电网的网络特点愈来愈贴近输电网,应用于输电网的分析方法也逐步应用于配电网的分析中。对含有DG的ADN进行规划研究及控制分区方法研究有助于提高配电网管理效率,为进一步制定ADN的管理策略提供参考。
[Abstract]:In 2008 years, the International Council on Large Electric Systems CIGRE( International Council on Large Electric Systems) Project Group C6C6.11 put forward the concept of active Distribution Network Adn. And can control the distributed generation generation DGs connected to the distribution network. The flexible network topology is used to manage the power flow of the distribution network, which realizes the "active" control of electric energy and the "active" distribution. It marks the change of the control mode of the distribution network. On the one hand, the ADN with DG can deal with the access of DG with high permeability. To solve the problem of large-scale access to distributed energy in the future distribution network, on the other hand, we can make use of the technical advantages of ADN to ensure the stable operation of the system and increase economic benefits, and cooperate with and promote the further development of the distribution network. Multi-objective programming for ADN is helpful to give consideration to many indexes and make the distribution network run in the best condition. On the one hand, it can reduce the scope of control and improve the controllability of the system; on the other hand, it can deal with different kinds of distributed energy access, and it does not need to build too complex system model. Reducing the complexity of computation, making the system keep good expansibility, has guidance and reference significance for the construction of strong power grid. This paper focuses on the study of ADN planning and control zoning, and completes the following work: firstly, the concept, structure, function and management mode of ADN are systematically expounded, and the relationship and difference between ADN and passive distribution network and active distribution network are expounded. The influence of DG on the planning, operation and protection of distribution network is analyzed, and the ADN with DG is discussed systematically, which provides the theoretical basis for the following research. Secondly, this paper analyzes the existing methods and problems of ADN programming, and establishes a multi-objective hierarchical programming model considering reliability and economy as a whole. In the upper planning model, the location of distributed generation is pre-selected according to the node location coefficient, and the objective function including the cost of DG equipment, the cost of network loss, the annual cost of electricity purchase and the government subsidy are set in the lower planning model. Particle Swarm Optimization (PSO) algorithm is used to calculate the DG, to determine the installation position and capacity of DG, and to optimize the system economically on the premise of ensuring the reliability of power supply, so that both reliability and economy can be taken into account. Finally, the method of control partition in ADN active management mode is studied by using complex network theory. The electrical distance is defined and the weight matrix is established to provide the unit of measurement for the division of the AND control area; the modular degree is taken as the criterion to evaluate the result of the partition; the PSO algorithm is used to calculate it. In order to accelerate the convergence and save storage space, the coding method and updating strategy adapted to the topological characteristics of ADN network are proposed for particle coding and updating. With the example of IEEE 33-bus 69-bus distribution system, the validity of the multi-objective programming model and the control zoning method are verified. The complexity of the structure and function makes the network characteristics of the distribution network more and more close to the transmission network. The analysis method applied to transmission network is also applied to distribution network analysis step by step. The study on planning and control zoning of ADN with DG is helpful to improve the efficiency of distribution network management and provide a reference for the further development of ADN management strategy.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM715

【相似文献】