Abstract: Accurate identification of key lines is vital to ensuring grid stability and forestalling large-scale cascading failures, yet existing methodologies remain inadequate. Previous research has predominantly employed singular value decomposition of the load rate matrix to discern these critical lines, a strategy that does not fully encapsulate the comprehensive operational state of the grid. Moreover, reactive power—chiefly determined by line voltage differences—exerts a significant influence on system stability. To address this shortcoming, this paper introduces a novel key line identification index that integrates both load rate and voltage drop. This approach is designed to extract more complete and nuanced information from temporal data, thereby enhancing the accuracy and practical applicability of key line recognition in power systems.