Estimation of Power System Frequency Dynamics Using Cubature and Ensemble Kalman Filters

Maintaining a stable power grid ensures a reliable electricity supply. To achieve this, advanced monitoring and control of the frequency dynamics of the microgrid are essential. Accurate estimation of frequency dynamics enables fast frequency support to the grid during abnormalities, thus enhancing grid stability and resilience. Various state- and parameter-estimation methods for frequency dynamics have been widely discussed in the literature. Among these methods, the Kalman filter stands out as one of the most well-known and extensively utilized techniques. This paper compares the accuracy and computational time of two prominent Kalman filter (KF) variants: the Cubature Kalman Filter (CKF) and the Ensemble Kalman Filter (EnKF). Both were designed to estimate the states and parameters for frequency dynamics of the power system by introducing Gaussian and non-Gaussian noise, which is highly important to emulate the real-world measurements. EnKF showed better accuracy than other Kalman filters for state estimation but was computationally expensive. This paper provides insights on selecting a suitable filter for practical implications.