Power system state estimation accuracy and observability evaluation with optimal PMU placement

Abstract

This paper introduces a phasor measurement units (PMUs) placement technique where redundant measurements can be removed and only critical measurements that will guarantee system observability are kept. The criterion for the measurement reduction algorithm is the minimum condition number of the measurement matrix. In this paper, the solution of the state estimation problem is achieved by two methods, the weighted least square (WLS) state estimation and the least absolute value (LAV) state estimation. In the WLS methodology, the Singular Value Decomposition (SVD) technique is used. The WLS would fail in the presence of bad data measurements, and a post WLS process will be required to identify and eliminate the effect of bad data. The LAV estimator, on the other hand, has the advantage of automatic bad data rejection if the measurements do not contain any leverage measurements. The simulations were done using the IEEE 9 and 14 bus system.