A fully distributed state estimation using matrix splitting methods

Abstract

In this paper, we propose a fully distributed state estimation algorithm for the electric power grid using matrix splitting methods. Our method incorporates both traditional SCADA measurements as well as new PMU data. The proposed distributed scheme relies only on local information and a minimal amount of information from neighboring areas which reduces communication burdens and increases the robustness of state estimation. The distributed estimate is guaranteed to exponentially converge to the centralized optimal estimate. The effectiveness of the method is demonstrated in various numerical experiments.