Otro
Detection and Identification of Abnormalities in Customer Consumptions in Power Distribution Systems
Registro en:
IEEE Transactions on Power Delivery. Piscataway: IEEE-Inst Electrical Electronics Engineers Inc, v. 26, n. 4, p. 2436-2442, 2011.
0885-8977
10.1109/TPWRD.2011.2161621
WOS:000298981800041
Autor
dos Angelos, Eduardo Werley S.
Saavedra, Osvaldo R.
Carmona Cortes, Omar A.
de Souza, Andre Nunes
Resumen
This paper proposes a computational technique for the classification of electricity consumption profiles. The methodology is comprised of two steps. In the first one, a C-means-based fuzzy clustering is performed in order to find consumers with similar consumption profiles. Afterwards, a fuzzy classification is performed using a fuzzy membership matrix and the Euclidean distance to the cluster centers. Then, the distance measures are normalized and ordered, yielding a unitary index score, where the potential fraudsters or users with irregular patterns of consumption have the highest scores. The approach was tested and validated on a real database, showing good performance in tasks of fraud and measurement defect detection. Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)