Distributed Energy Resources Topology Identification via Graphical Modeling
Distributed Energy Resources Topology Identification via Graphical Modeling
复制标题
DOI：
10.1109/tpwrs.2016.2628876
复制
发表时间：
2017-07-01
影响因子：
6.6
通讯作者：
Rajagopal, Ram
中科院分区：
文献类型：
--
作者：
Weng, Yang;Liao, Yizheng;Rajagopal, Ram
分布式能源（DER），如光伏，风能和燃气发电机，比以往任何时候都更多地连接到电网，这给配电网带来了巨大的变化。由于这些变化，重要的是要了解这些DER连接在哪里，以便可持续地运行配电网。但由于配电网频繁重构以及对新元件的认识不足，配电网的精确拓扑难以获得。在本文中，我们提出了一种方法，利用新的数据从传感器配备DER设备，以获得配电网拓扑结构。具体地说，一个图形模型来描述不同的电压测量之间的概率关系。在潮流分析的基础上，提出了一种基于互信息的树状和部分环网辨识算法。在IEEE标准配电网测试系统和电力科学研究院测试系统中的仿真结果表明，该方法具有较高的连通性识别精度。
Distributed energy resources (DERs), such as photovoltaic, wind, and gas generators, are connected to the grid more than ever before, which introduces tremendous changes in the distribution grid. Due to these changes, it is important to understand where these DERs are connected in order to sustainably operate the distribution grid. But the exact distribution system topology is difficult to obtain due to frequent distribution grid reconfigurations and insufficient knowledge about new components. In this paper, we propose a methodology that utilizes new data from sensor-equipped DER devices to obtain the distribution grid topology. Specifically, a graphical model is presented to describe the probabilistic relationship among different voltage measurements. With power flow analysis, a mutual information-based identification algorithm is proposed to deal with tree and partially meshed networks. Simulation results show highly accurate connectivity identification in the IEEE standard distribution test systems and Electric Power Research Institute test systems.