Robust State Estimation for Power Systems Using Quasi-Newton Method

This article addresses the state estimation problem in power system under non-Gaussian noise environment. This estimator is designed based on maximum correntropy criterion (MCC), which exhibits the robustness with reference to non-Gaussian noise. To avoid computing the inverse matrix of Hessian matrix, the quasi-Newton Broyden-Fletcher-Goldfarb-Shanno (BFGS) method is adopted to propose a centralized power system state estimation algorithm. For the reason of security, privacy and further reducing the computation burden, adapt-then-combine (ATC) strategy is applied to design a distributed state estimation algorithm based on the centralized algorithm. The effectiveness of the proposed method is verified by the numerical results of IEEE 14-bus and IEEE 118-bus systems under uniformly distributed noise, t-distribution noise and impulsive noise.