基于PageRank改进算法的电力系统关键节点识别

Identification Method of Key Nodes in Power System Based on Improved PageRank Algorithm

为了能够在电力系统发生故障之初准确识别出系统中关键节点,提出一种基于PageRank改进算法的系统关键节点识别方法。首先,通过雅可比矩阵获取系统的电压无功灵敏度矩阵和相角有功灵敏度矩阵,定义系统不同指标下的链接矩阵。其次,考虑电力通信系统对电网的影响,基于节点收缩原理定义系统衍生网络,并对链接矩阵进行修正得到网络拓展矩阵。最后,基于改进的PageRank算法得到系统节点重要性权重,并按照权重值对节点排序。通过IEEE 39节点系统的仿真验证了该方法的正确性。

In order to identify the key nodes at the beginning of the power system fault, a recognition method of key nodes based on the improved PageRank algorithm is proposed. First, the sensitivity matrix of voltage reactive power and the sensitivity matrix of phase angle active power are obtained based on Jacobi matrix to define the link matrices under different indices of system. Secondly, considering the influence of power communication system on power network, the derived networks are defined based on the node contraction principle, and the link matrices are modified to get the network expansion matrices. Finally, the weight of system nodes are obtained based on the improved PageRank algorithm and the system nodes are sorted according to the weight values. The correctness of the method is verified on the simulation of IEEE39 bus system.