针对大规模新能源并网场景下电网关键输电断面频繁转移、难以快速准确辨识的问题，提出了一种基于改进规范切与安全风险指标的关键输电断面辨识方法。基于谱图论构建了网络拓扑赋权图，对支路赋权时考虑了节点电压和支路潮流因素的影响，以综合反映节点之间的电气联系紧密程度；采用改进规范切的谱聚类算法对电网进行分区，对于分区之间形成的输电断面，利用安全风险指标来筛选关键输电断面。IEEE 39节点系统算例和广东电网算例的结果表明，所提方法能准确有效地辨识系统的关键输电断面及其越限风险，并且在实际电网的多场景应用中能够筛选出未识别的关键输电断面。

Aiming at the problem that key transmission sections of the power grid are frequently transferred and difficult to identify quickly and accurately in large-scale renewable energy grid connection scenarios, a key transmission section identification method based on improved normalized cut and security risk index is proposed. A weighted graph of network topology is constructed based on spectral graph theory, considering the influences of node voltage and branch power flow factors when assigning weights to branches. This aims to comprehensively reflect the electrical connectivity between nodes. The network is partitioned using an improved spectral clustering algorithm based on normalized cut, and for the transmission section formed between partitions, key transmission sections are selected using security risk index. The results from case studies on the IEEE 39-bus system and Guangdong Power Grid indicate that the proposed method can accurately and effectively identify key transmission sections within the system and assess their risk of exceeding limits. Moreover, it can identify some unrecognized key transmission sections in multi-scenario applications in the actual power grid.