A data driven approach to distribution network topology identification

摘要

This paper proposes a methodology for estimating radial distribution grid topologies with the use of smart meter data. Power and voltage measurements from smart meters installed in consumer point of common coupling to the distribution grid can be used for calculating the voltage sensitivity matrix. The authors argue that this matrix can be utilized for estimating the distribution grid topology. The distribution grid sensitivity approach followed is first presented. Next the authors describe a simple method for identifying the voltage sensitivity matrix given a set of appropriate smart meter power and voltage measurements and without full knowledge of the distribution grid topology. After identifying the voltage sensitivity matrix an algorithm is presented for the estimation of the distribution grid topology. Since GIS is not yet supported for MV and especially LV networks, this methodology can enhance the optimal distribution grid planning and operation.