Robust Dynamic State Estimation for Power System Based on Adaptive Cubature Kalman Filter With Generalized Correntropy Loss

Abstract

Due to the unfavorable interference of non-Gaussian noise, abnormal system states, and rough measurement errors, dynamic state estimation (DSE) plays an important role in the safe operation of power system. A novel DSE method based on an adaptive cubature Kalman filter (CKF) with generalized correntropy loss (GCL) criterion, termed AGCLCKF, is developed to deal with the complex non-Gaussian distribution noises of power system in this paper. First, a nonlinear regression model is derived to simultaneously incorporate the state and noise errors into the GCL cost function, and a fixed-point iteration is exploited to recursively update the posterior state estimate. Then, considering that the filtering performance of the estimator is largely determined by the kernel bandwidth in correntropy, an adaptive factor is established to adjust the kernel bandwidth of kernel function in real-time, which can improve the flexibility and accuracy of dynamic state estimation in the existence of bad measurement information. Finally, extensive simulation results carried out on the IEEE 39-bus test system demonstrate that the proposed method can achieve much accuracy and robustness under various situations.