Improving electricity non technical losses detection including neighborhood information

Pablo Massaferro; Henry Marichal; Matias Di Martino; Fernando Santomauro; Juan Pablo Kosut; Alicia Fernandez

doi:10.1109/PESGM.2018.8586146

Improving electricity non technical losses detection including neighborhood information

Pablo Massaferro
, Henry Marichal
, Matias Di Martino
, Fernando Santomauro
, Juan Pablo Kosut
, Alicia Fernandez

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Scopus citations

Abstract

Non technical losses (NTL) cause significant damage to power supply companies' economies. Detecting abnormal clients behavior is an important and difficult task. In this paper we analyze the impact of considering customers geo-localization information, in automatic NTL detection. A methodology to find optimal grid sizes to compute a set of local features with a random search procedure is proposed. The number and size of the grids, and other classification algorithm parameters are adjusted to maximize the area under receiver operating characteristic curve (AUC), showing performance improvements in a data set of 6 thousand of Uruguayan residential customers. Comparative analysis with different sub-sets of characteristics, that include the monthly consumption, contractual information and the new local features are presented. In addition, we probe that raw customers' geographical location used as an input feature, gives competitive results as well. In addition we evaluate a entire new database of 6 thousand Uruguayan customers, whom were inspected in-site by UTE experts between 2015 and 2017.

Original language	English
Title of host publication	2018 IEEE Power and Energy Society General Meeting, PESGM 2018
Publisher	IEEE Computer Society
ISBN (Electronic)	9781538677032
DOIs	https://doi.org/10.1109/PESGM.2018.8586146
State	Published - 21 Dec 2018
Event	2018 IEEE Power and Energy Society General Meeting, PESGM 2018 - Portland, United States Duration: 5 Aug 2018 → 10 Aug 2018

Publication series

Name	IEEE Power and Energy Society General Meeting
Volume	2018-August
ISSN (Print)	1944-9925
ISSN (Electronic)	1944-9933

Conference

Conference	2018 IEEE Power and Energy Society General Meeting, PESGM 2018
Country/Territory	United States
City	Portland
Period	5/08/18 → 10/08/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Access to Document

10.1109/PESGM.2018.8586146

Cite this

Massaferro, P., Marichal, H., Martino, M. D., Santomauro, F., Kosut, J. P., & Fernandez, A. (2018). Improving electricity non technical losses detection including neighborhood information. In 2018 IEEE Power and Energy Society General Meeting, PESGM 2018 Article 8586146 (IEEE Power and Energy Society General Meeting; Vol. 2018-August). IEEE Computer Society. https://doi.org/10.1109/PESGM.2018.8586146

@inproceedings{9dc8e48fc2134b9395c5a64babbb82b6,

title = "Improving electricity non technical losses detection including neighborhood information",

abstract = "Non technical losses (NTL) cause significant damage to power supply companies' economies. Detecting abnormal clients behavior is an important and difficult task. In this paper we analyze the impact of considering customers geo-localization information, in automatic NTL detection. A methodology to find optimal grid sizes to compute a set of local features with a random search procedure is proposed. The number and size of the grids, and other classification algorithm parameters are adjusted to maximize the area under receiver operating characteristic curve (AUC), showing performance improvements in a data set of 6 thousand of Uruguayan residential customers. Comparative analysis with different sub-sets of characteristics, that include the monthly consumption, contractual information and the new local features are presented. In addition, we probe that raw customers' geographical location used as an input feature, gives competitive results as well. In addition we evaluate a entire new database of 6 thousand Uruguayan customers, whom were inspected in-site by UTE experts between 2015 and 2017.",

author = "Pablo Massaferro and Henry Marichal and Martino, \{Matias Di\} and Fernando Santomauro and Kosut, \{Juan Pablo\} and Alicia Fernandez",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Power and Energy Society General Meeting, PESGM 2018 ; Conference date: 05-08-2018 Through 10-08-2018",

year = "2018",

month = dec,

day = "21",

doi = "10.1109/PESGM.2018.8586146",

language = "Ingl{\'e}s",

series = "IEEE Power and Energy Society General Meeting",

publisher = "IEEE Computer Society",

booktitle = "2018 IEEE Power and Energy Society General Meeting, PESGM 2018",

}

Massaferro, P, Marichal, H, Martino, MD, Santomauro, F, Kosut, JP & Fernandez, A 2018, Improving electricity non technical losses detection including neighborhood information. in 2018 IEEE Power and Energy Society General Meeting, PESGM 2018., 8586146, IEEE Power and Energy Society General Meeting, vol. 2018-August, IEEE Computer Society, 2018 IEEE Power and Energy Society General Meeting, PESGM 2018, Portland, United States, 5/08/18. https://doi.org/10.1109/PESGM.2018.8586146

Improving electricity non technical losses detection including neighborhood information. / Massaferro, Pablo; Marichal, Henry; Martino, Matias Di et al.
2018 IEEE Power and Energy Society General Meeting, PESGM 2018. IEEE Computer Society, 2018. 8586146 (IEEE Power and Energy Society General Meeting; Vol. 2018-August).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Improving electricity non technical losses detection including neighborhood information

AU - Massaferro, Pablo

AU - Marichal, Henry

AU - Martino, Matias Di

AU - Santomauro, Fernando

AU - Kosut, Juan Pablo

AU - Fernandez, Alicia

PY - 2018/12/21

Y1 - 2018/12/21

N2 - Non technical losses (NTL) cause significant damage to power supply companies' economies. Detecting abnormal clients behavior is an important and difficult task. In this paper we analyze the impact of considering customers geo-localization information, in automatic NTL detection. A methodology to find optimal grid sizes to compute a set of local features with a random search procedure is proposed. The number and size of the grids, and other classification algorithm parameters are adjusted to maximize the area under receiver operating characteristic curve (AUC), showing performance improvements in a data set of 6 thousand of Uruguayan residential customers. Comparative analysis with different sub-sets of characteristics, that include the monthly consumption, contractual information and the new local features are presented. In addition, we probe that raw customers' geographical location used as an input feature, gives competitive results as well. In addition we evaluate a entire new database of 6 thousand Uruguayan customers, whom were inspected in-site by UTE experts between 2015 and 2017.

AB - Non technical losses (NTL) cause significant damage to power supply companies' economies. Detecting abnormal clients behavior is an important and difficult task. In this paper we analyze the impact of considering customers geo-localization information, in automatic NTL detection. A methodology to find optimal grid sizes to compute a set of local features with a random search procedure is proposed. The number and size of the grids, and other classification algorithm parameters are adjusted to maximize the area under receiver operating characteristic curve (AUC), showing performance improvements in a data set of 6 thousand of Uruguayan residential customers. Comparative analysis with different sub-sets of characteristics, that include the monthly consumption, contractual information and the new local features are presented. In addition, we probe that raw customers' geographical location used as an input feature, gives competitive results as well. In addition we evaluate a entire new database of 6 thousand Uruguayan customers, whom were inspected in-site by UTE experts between 2015 and 2017.

UR - https://www.scopus.com/pages/publications/85060799814

U2 - 10.1109/PESGM.2018.8586146

DO - 10.1109/PESGM.2018.8586146

M3 - Contribución a la conferencia

AN - SCOPUS:85060799814

T3 - IEEE Power and Energy Society General Meeting

BT - 2018 IEEE Power and Energy Society General Meeting, PESGM 2018

PB - IEEE Computer Society

T2 - 2018 IEEE Power and Energy Society General Meeting, PESGM 2018

Y2 - 5 August 2018 through 10 August 2018

ER -

Improving electricity non technical losses detection including neighborhood information

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this