Identification of non-technical losses in distribution systems via state estimation and geometric tests

Abstract

The incidence of non-technical losses in distribution networks is a well-known detrimental factor to the whole economic chain related to the electric energy distribution process. It is thus important to equip distribution system operators with adequate tools to localize and quantify such losses, in order to subsequently take appropriate measures to curb undue consumer practices. This paper proposes a methodology based on state estimation and geometrically-based hypothesis testing capable of not only identifying the occurrence of non-technical losses, but also to estimate their magnitude. For that purpose, a three-phase state estimator is employed which processes available real-time measurements and load forecast values. Once the state estimates are available, hypotheses made by the operator about the load at selected nodes of the network are either confirmed or rejected by the accessory geometric procedure. In the latter case, the load mismatch corresponds to non-technical losses at that node and can be readily estimated. Two IEEE test feeders are used to illustrate the applicability and to assess the performance of the proposed approach.