Least Squares Estimation and Kalman Filter Based Dynamic State and Parameter Estimation

In this paper, two types of generator model state and parameter estimation methods via Phasor Measurement Unit (PMU) data are described. The first type is the least square errors estimation (LSE) for parameter estimation and the second type is Kalman filter based estimation for both parameters and states. For LSE-based method, with parameters estimated, states can be estimated via event playback. LSE-based estimation employs a window of time-series data, while Kalman filtering method conducts estimation at every time step. LSE, extended Kalman filter (EKF) and unscented Kalman filter (UKF)-based estimation approaches will be demonstrated through case studies.