Hidden Cyber-Physical Contingency Identification, Classification and Evaluation in Modern Power Systems

This paper introduces an advanced stochastic hybrid system modeling framework for modern power systems (MPS) to identify, classify, and evaluate hidden contingencies, which cannot be detected by normal observation sensors. The stochastic hybrid system (SHS) model is designed to capture the dynamics of the internal states of individual nodes, considering their structural properties, and coupling variables under various local and network-level contingencies. Hidden contingencies are identified using a probing approach that measures changes in the eigenvalues of the SHS model and detects deviations from normal operation. Next, contingencies are categorized into three distinct groups according to their impact on MPS: physical contingencies, control network contingencies, and sensing and measurement network contingencies. This classification enables a proactive evaluation of contingencies. The practicality and efficacy of the proposed methodology are validated through simulation experiments on the electrical network of two real-world systems. These simulations underscore the approach's capacity to enhance the resilience of power systems against a spectrum of hidden contingencies.
Comment: Under review in IEEE Transactions on Power Systems