Robust State Estimator Based on Maximum Exponential Absolute Value

Abstract

In this paper, a robust state estimator based on maximum exponential absolute value (MEAV) is proposed by using the maximum correntropy criterion with Laplace kernel function as the objective function of state estimation. Since the objective function of MEAV model is continuous but nondifferentiable, its equivalent model is given, which is then solved based on the primal-dual interior point method. To further improve the computational efficiency, a new correction equation with much lower order is presented and the number of fill-ins in the solving process is reduced by a systematic method. Simulations based on the IEEE benchmark systems and two real grids of China demonstrate that the proposed MEAV estimator is very robust with high efficiency.