Novel meter placement algorithm for enhanced accuracy of distribution system state estimation

Abstract

As distribution networks are turning into active systems, enhanced observability and continuous monitoring becomes essential for effective management and control. Distribution networks require further instrumentation with sensors and meters to monitor the network in the control room close to real time. The metered data will be fed into the state estimation tool which will compute all the states of the network. The cost of this meter placement is a major issue as the network operator will have to engage in a massive investment. In this paper we propose a stochastic meter placement algorithm to reduce the voltage magnitude estimation errors, along with improving the robustness of the state estimation problem, via minimum measurement placements. The algorithm exploits the scaling property of the singular values obtained from the singular value decomposition of a matrix for reduced computation.