Stable Convergence State Estimation Method for Bad Leverage Measurements

Traditional state estimation methods which include bad data identification fail in bad leverage measurements, resulting the degradation of state estimation precision. In order to exclude the interference of bad leverage measurements automatically, while ensuring the numerical stability and convergence of state estimation calculation, this paper proposes a stable convergence state estimation method for bad leverage measurements. On the base of utilizing the principle of equivalent weight and unifying the scales of measurement residuals, this paper presents the exponential weight function by using the standardized residual and builds the model of robust state estimation. This model can correct the weight of measurement automatically according to the measurement residual, therefore, shows the strong ability to resist bad leverage measurements and effectively guarantee the numerical stability and convergence speed. The characteristics of the proposed method are demonstrated on the 3-bus and the 118-bus systems.