Distributed state estimation of interconnected power systems with time-varying disturbances and random communication link failures

Abstract

State estimation of multi-area interconnected power systems is crucial for reflecting system operations and guiding actuator responses. However, sensor faults, communication link failures, and external random disturbances can inevitably lead to power system failures. Therefore, designing a highly effective state estimation method capable of timely and accurate detection of sensor faults in the power grid to mitigate losses is of significant practical importance. This paper addresses the fault estimation problem in multi-area power systems with parameter uncertainties and proposes a fault-tolerant state estimator that accounts for communication link failures between network layers in each area. These communication link failures are modelled as Bernoulli-distributed variables. State estimation is achieved using information from adjacent power system areas. Sufficient conditions for the error system's $H_{\infty }$ performance are provided using Lyapunov stability theory and linear matrix inequality methods. Finally, simulations on a three-area interconnected power system validate the proposed method's effectiveness in mitigating the effects of communication link failures and random disturbances. The method accurately and rapidly estimates sensor faults and the system state, ensuring stable operation and enhancing grid reliability.