Power system state estimation using contraction mapping and singular value decomposition

State estimation plays a key role in the operation of power systems. This role becomes more important considering the increasing demand of emerging power market. Many methods have been proposed for power system state estimation, mostly based on Weighted Least Squares (WLS) approach. However, it is well known that Least Absolute Value (LAV) estimators are more efficient in terms of robustness and accuracy. For these estimators there is no closed form solution and each LAV estimator has its own criteria in choosing desired measurements. In this research, two novel LAV estimators are introduced for power system state estimation. The first estimator employs contraction mapping concepts for rejecting redundant measurements. The second estimator is introduced for systems where sparsity and ill-conditioning occur in the system matrix. In the second estimator, Singular Value Decomposition (SVD) method is combined with contraction mapping technique to find the appropriate equations for the estimation. The application of the new estimator is studied on different IEEE power systems for verification. The estimator shows a robust performance in all the test systems, and the estimation error remains comparatively small even in the presence of significant number of bad data points.