Derivative-Free Kalman Filtering Based Approaches to Dynamic State Estimation for Power Systems With Unknown Inputs

This paper proposes a decentralized derivative-free dynamic state estimation method in the context of a power system with unknown inputs, to address cases when system linearization is cumbersome or impossible. The suggested algorithm tackles situations when several inputs, such as the excitation voltage, are characterized by uncertainty in terms of their status. The technique engages one generation unit only and its associated measurements, and it remains totally independent of other system wide measurements and parameters, facilitating in this way the applicability of this process on a decentralized basis. The robustness of the method is validated against different contingencies. The impact of parameter errors, process, and measurement noise on the unknown input estimation performance is discussed. This understanding is further supported through detailed studies in a realistic power system model.