基于拓扑分割的网络可观测性分析方法

Observability algorithm for power network based on topological segmentation

针对电力网络可观测性分析问题,对量测网络建模、网络拓扑可观测性分析理论、不可观测节点的影响范围等方面进行了研究,提出了一种基于拓扑分割的网络可观测性分析方法,在某42节点系统上对该方法进行了测试。该方法将带量测的电力网络建模为量测拓扑单线图,并在无注入量测节点处进行了拓扑分割,将网络分割成了多个连通子网。利用广度优先双向搜索算法对含线路潮流量测的各个子网进行了可观测性分析,在不可观测子网中分析了不可观测节点的影响范围,得到了所有可观测拓扑岛。再结合其结果对仅含注入型量测的子网进行分析,得到了剩余的可观测岛。最后将共有相同无注入量测节点的可观测岛进行了合并。算法在判断可观测性的同时,得到了全网所有的可观测拓扑岛。研究结果表明,该方法正确有效,保持了拓扑算法的快速性特点。

Aiming at power network observability problem, the measurement network modeling, network topological observability analysis theory, and the influence scope of unobservable nodes was analyzed to derive a new observability analysis algorithm, which was tested on a 42-node system, based on topological segmentation. The power network with measurement was modeled as a measurement one-line graph. Using breadth first bidirectional search, observability of sub-graphs with line flow measurement was analyzed after tearing the original one-line graph apart at node without injection measurement. The influence of unobservable part in unobservable sub-graph was then determined to obtain all observable islands. Considering known observable islands, observability of sub-graphs which only had injection measurement was analyzed and observable islands in them were got. Finally all observable islands sharing same nodes without injection measurement were combined. The algorithm can get all the observable islands while observability analysis. The results indicate that the algorithm is correct, effective and maintaining the rapidity of topological method.