Uncertainty analysis of aggregated smart meter data for state estimation

In distribution networks, the data redundancy is usually assumed to be an inevitable bottleneck of traditional grid control and operation. Recently, the availability of smart meter data in distribution systems has provided an opportunity to improve the observability. As for the medium-voltage (MV) distribution system, there is an increasing interest to use the spatially aggregated smart meter data from low-voltage (LV) feeders in the state estimation, instead of inaccurate pseudo-measurements. However, good performance of state estimators requires good knowledge of the available measurements, in terms of both the mean value and the associated uncertainty. Hence, this paper intends to pave a new way of utilizing and aggregating smart meter data for the purpose of state estimation in the MV distribution system, in a concrete and reliable manner. The feasibility of the proposed method is verified on the IEEE European Low Voltage Test Feeder with a set of real-world smart meter data. Simulation results show that the utilization of aggregated smart meter data is able to improve the accuracy of load modelling of three-phase transformers.