A Robust Dynamic State Estimation Method for Power Systems Using Exponential Absolute Value-Based Estimator
Tengpeng Chen, He Ren, Po Li, G.A.J. Amaratunga
Abstract
Even though the noise model applied in power system dynamic state estimation is usually assumed to be Gaussian, this is not the case due to the unknown system inputs, influence from the communication channel noise and the outliers generated by phasor measurement units (PMUs). In this paper a robust power system dynamic state estimation (DSE) method combining a robust exponential absolute value based estimator and the unscented Kalman filter together is proposed under non-Gaussian noise. Based on the quadratic function and the exponential absolute value function, robust exponential absolute value based estimator is derived, further mitigating the effects of bad data or outliers. The influence function is utilized to calculate the state estimation error covariance of the proposed robust DSE method. The simulation results on the IEEE 39-bus system verify the robustness and effectiveness of the proposed dynamic state estimation method.