Key points are not available for this paper at this time.
Interconnected power systems are increasingly vulnerable to parameter deviations—such as mechanical wear, blade loss, inertia degradation, or cyber-physical attacks—in turbine–governors, generators, and transmission lines. These deviations compromise stability and may lead to severe disturbances if not detected and isolated promptly. Conventional observer-based fault detection methods can identify anomalies but often fail to pinpoint the exact parameter responsible. This paper proposes a robust Fault Detection and Isolation (FDI) framework capable of estimating and isolating key dynamic parameters, including turbine (Tt) and governor (Tg) time constants, inertia (H), damping (D), and tie-line synchronizing coefficients (Tij). The method integrates an H∞/H₂ observer with pole placement for disturbance attenuation and rapid residual generation, followed by an adaptive sliding mode estimator for parameter-specific isolation. This two-stage scheme enables precise differentiation between faults and noise, as well as between different types of parametric shifts. Simulation studies on a multi-area load frequency control (LFC) system validate the accuracy and robustness of the proposed approach under diverse fault scenarios. Unlike conventional FDI techniques, the framework not only detects faults but also isolates their root causes, thereby providing actionable insights for operators and enhancing the resilience of modern interconnected power networks.
Nohra et al. (Sat,) studied this question.