Robust State Estimation Against Measurement and Network Parameter Errors

Abstract

Errors in network parameters can cause serious bias in state estimation solution and make good measurements identified and discarded as bad data. This paper addresses this issue by developing a state estimator that remains robust against parameter errors. This is accomplished by modifying the formulation of the well-known least absolute value state estimator in order to identify and reject not only gross measurement errors, but also parameter errors in the network model. The resulting state estimate will not only be free from the impact of erroneous parameters, but will also reliably estimate and correct the erroneous parameters at the same time. The proposed approach can be formulated as a linear programming (LP) problem, and solved in a computational efficient manner by existing LP solvers. Simulation results show that it is effective under different types of parameter errors, gross measurement errors, and Gaussian noise.