Automatic phase identification of smart meter measurement data

Abstract

This paper highlights the importance of the knowledge of the phase identification for the different measurement points inside a low-voltage distribution network. Besides considering existing solutions, we propose a novel method for identifying the phases of the measurement devices, based exclusively on voltage measurement correlation. It relies on graph theory and the notion of maximum spanning tree. It has been tested on a real Belgian LV network, first with simulated unbalanced voltage for which it managed to correctly identify the phases of all measurement points, second, on preliminary data from a real measurement campaign for which it shows encouraging results.

Phase Identification of Smart Meters by Clustering Voltage Measurements

“…This paper extends the work done in [11]. Its contributions are four-fold: (i) It proposes a novel algorithm with a global network, using the advantages of both graph theory and correlation to identify the measurements that should be linked together and cluster them.…”

“…Indeed, a tree is defined as an undirected graph in which any two nodes are connected by exactly one path, in other words without cycles. It makes sense to use a tree structure to cluster the measurements, as explained in [11], because the tree structure of the network creates an underlying structure between the voltage measurements. Indeed, the voltages at two points that are neighbours in the network tree are more correlated.…”