Decoupled Distributed State Estimator with Reduced Number of Measurements for Power System Applications

AbstractThis paper presents the formulation of a decoupled distribut-ed state estimator using a reduced number of measurements. This proposal aims to solve a distributed state estimation in which the estimation of each area of the system will be performed locally. The measurements made by the PMUs installed in different nodes defined as references are included in the solution of the state estimator. The estimator is based on the weighted least squares (WLS) method. Furthermore, we intend to use a reduced number of measurements during the estimation, so we use a minimum spanning tree to provide the minimum amount of measurements that guarantee the observability of each area to be estimated. Consequently, the proposed state estimator uses only the transferred power and phasor voltages aggregating the phase angles of the voltages in each area of the decomposed system. In order to validate the performance of the implemented distributed estimator, the states obtained for the IEEE-39 bus system case study are compared with the states of an implementation of the system in the PowerFactory electrical analysis tool. The results show an adequate performance of the estimator. Consequently, the feasibility of a distributed estimation that can be performed in a decoupled way and with a reduced number of measurements is verified.KeywordsDistributed state estimationDecoupled estimationMinimum spanning treePhasor measurement unitWeighted least squares