Early Detection of Voltage Instability: A Transparent, Rule-Based Method for Cyber-Resilient Grids

This paper proposes a rule-based algorithm for fast prediction of long-term voltage stability status immediately after a disturbance, eliminating the need for post-disturbance measurements. Unlike traditional Phasor Measurement Unit (PMU) dependent methods vulnerable to cyber threats like data spoofing, the proposed approach uses PMU data only for initial system updates. It first validates measurements against the last trusted system state using consistency checks, corrects any discrepancies, and then bases stability decisions on inherent characteristics—power flow convergence and generator reactive power margins—ensuring cyber-resilient operation. The method’s rule-based logic ensures intrinsic immunity to cybersecurity risks while maintaining compatibility with both large and small disturbances triggering long-term voltage instability. Optimized for real-time operation, the simulation-based algorithm meets stringent computational speed requirements for online stability assessment. Comprehensive simulations on IEEE test systems under N-1/N-2 contingencies and load disturbances demonstrate three key advantages: (1) Early and accurate voltage stability prediction (96.8% detection accuracy), (2) Reliable identification of critical generators/loads contributing to instability, and (3) Cyber-resilient operation verified through adversarial test cases. The results establish the method as a transparent, infrastructure-independent solution for grid stability monitoring, particularly valuable for cybersecurity-sensitive environments.