基于故障链聚类算法的电网关键线路辨识

研究连锁故障发展机理以及辨识故障演化路径中的关键线路,对揭示电力系统薄弱环节、降低连锁故障风险具有重要意义。为此,文中提出一种基于故障链聚类算法的电网关键线路辨识方法。首先,建立改进的基于直流潮流法的电力系统解列模拟器(DCSS)连锁故障仿真模型,配合随机化学(RC)法高效生成含详细时序信息的故障链集合;然后,以编辑距离为相似性指标实现对故障链的层次聚类,分类后的故障链集合不仅可以降低后续数据挖掘的难度,而且能够更加全面地识别出不同连锁故障演化模式中的薄弱环节;最后,以Matpower 2383节点系统为例,根据关键线路扩容后系统风险水平的下降程度评估各类算法所辨识线路的重要性。结果表明文中方法能更好降低系统连锁故障风险水平,证明了所提模型及算法的有效性。

Critical line identification of power grid based on fault chain clustering algorithm

Research on the development mechanism of cascading failure and identification the critical line of evolution path is of great significance for exposing weak links in the power system and reducing the risk of cascading failures. For this reason, an identification method for critical lines of power grid based on fault chain clustering algorithm is proposed. Firstly, an improved direct current power flow simulator of power system separation (DCSS) cascading failure simulation model, which combines with the random chemistry (RC) method to efficiently generate fault chain set with detailed timing information is established. Then, the hierarchical clustering of the failure chain set is realized by using edit distance as the similarity index, and the classified fault chain set can not only reduce the difficulty of subsequent data mining, but also more comprehensively identify the weak links in different cascading failure evolution modes. Finally, taking the Matpower 2 383 node system as an example, the importance of the lines identified by various algorithms is evaluated through the reduction of the system risk level after the expansion of critical lines. The results show that the proposed method can better reduce the risk level of cascading failure, which proves the effectiveness of the proposed model and algorithm.