NTL-Unet: A Satellite-Based Approach for Non-Technical Loss Detection in Electricity Distribution Using Sentinel-2 Imagery and Machine Learning-Reference-Cited by-同舟云学术

NTL-Unet: A Satellite-Based Approach for Non-Technical Loss Detection in Electricity Distribution Using Sentinel-2 Imagery and Machine Learning

Published:2024-07-30 Issue:15 Volume:24 Page:4924
ISSN:1424-8220
Container-title:Sensors
language:en
Short-container-title:Sensors

Author:

Gremes Matheus Felipe¹^ORCID,Gomes Renato Couto²,Heberle Andressa Ullmann Duarte²,Bergmann Matheus Alan²,Ribeiro Luísa Treptow²,Adamski Janice²^ORCID,dos Santos Flávio Alves²,Moreira André Vinicius Rodrigues³,Lameirão Antonio Manoel Matta dos Santos³^ORCID,de Toledo Roberto Farias³^ORCID,de C. Filho Antonio Oseas⁴^ORCID,Andrade Cid Marcos Gonçalves¹,Lima Oswaldo Curty da Motta¹

Affiliation:

1. Department of Chemical Engineering, State University of Maringá (UEM), Maringá 87020-900, PR, Brazil

2. Pix Force Tecnologia S.A, Porto Alegre 90240-200, RS, Brazil

3. Light Serviços de Eletricidade S.A, Rio de Janeiro 20211-050, RJ, Brazil

4. Department of Electrical Engineering and Computer Science, Federal University of Piauí—UFPI, Teresina 64049-550, PI, Brazil

Abstract

This study introduces an orbital monitoring system designed to quantify non-technical losses (NTLs) within electricity distribution networks. Leveraging Sentinel-2 satellite imagery alongside advanced techniques in computer vision and machine learning, this system focuses on accurately segmenting urban areas, facilitating the removal of clouds, and utilizing OpenStreetMap masks for pre-annotation. Through testing on two datasets, the method attained a Jaccard index (IoU) of 0.9210 on the training set, derived from the region of France, and 0.88 on the test set, obtained from the region of Brazil, underscoring its efficacy and resilience. The precise segmentation of urban zones enables the identification of areas beyond the electric distribution company’s coverage, thereby highlighting potential irregularities with heightened reliability. This approach holds promise for mitigating NTL, particularly through its ability to pinpoint potential irregular areas.

Funder

PDI ANEEL program

Publisher

MDPI AG

Link

https://www.mdpi.com/1424-8220/24/15/4924/pdf

Reference39 articles.

1. Agência Nacional de Energia Elétrica (ANEEL) (2021). Electric Power Losses in Distribution. Technical Report Edition 1/2021, Obtained through the Market Information Monitoring System—SAMP Balance.

2. Hybrid deep neural networks for detection of non-technical losses in electricity smart meters;Buzau;IEEE Trans. Power Syst.,2019

3. Non-technical losses: A systematic contemporary article review;Siluk;Renew. Sustain. Energy Rev.,2021

4. Agência Nacional de Energia Elétrica (ANEEL) (2008). Technical Note No.342/2008. Methodology for Regulatory Treatment of Non-Technical Losses of Electrical Energy—Second Cycle of Periodic Tariff Revision of Electricity Distribution Concessionaires, Technical Report.

5. Chaves, A.C., Tavares, A., Ferreira, D., Tommaso, F., Dantas, G., de Barros Alvares, J., Takeuchi, J., Câmara, L., Mendes, P., and Maestrini, M. (2020). As Perdas Não Técnicas no Setor de Distribuição Brasileiro: Uma Abordagem Regulatória (Non-technical Losses in the Brazilian Distribution Sector: A Regulatory Approach), Grupo de Estudos do Setor Elétrico (GESEL).