A new approach for AC state estimation based on a linear network model

Abstract

This paper presents a new and efficient formulation for the state estimation (SE) which can be solved in one single shot of computation. The proposed approach adopts line flows and the square of voltage magnitudes as the problem state variable and considers both active and reactive power quantities. The objective function is to minimize the weighted sum of the least square values of measurement residuals. The nonlinearity associated with line losses is left by assuming these values as slack variables. Numerical studies are conducted through two standard networks. In order to compare the performance of the proposed method, the conventional weighted least square (WLS) state estimation with Gauss-Newton AC power flow is also accommodated. Simulation results demonstrate that the proposed method outperforms the conventional one particularly from the speed while keeping the accuracy in a reasonable range.