Studying the Impact of Smart Meter Placement on Low-Voltage Grid State Estimation

This paper comprehensively investigates the influence of smart meter allocation methods, considering sensors at grid buses and branches, on the performance of distribution system state estimation algorithms. Three algorithms are used, namely the Weighted Least Squares, the Extended Kalman Filter, and the Schweppe-type GM-estimator with the Huber psi-function. These are tested on a real low-voltage distribution grid with radial structure for multiple scenarios characterized by different penetration levels and types of measurements. Based on Monte Carlo simulations, different locations of sensors at buses and branches are considered for each scenario. An empirical study is carried out to assess the correlation of the placement of meters with the state estimation error. The results suggest that bus meters are most profitable at customers with the highest energy consumption. In addition, well distributed sensors at the grid branches based on a newly proposed `path search' method appear to be the most effective.