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

被引：0

作者：

Gremes, Matheus Felipe ^{[1
]}

Gomes, Renato Couto ^{[2
]}

Heberle, Andressa Ullmann Duarte ^{[2
]}

Bergmann, Matheus Alan ^{[2
]}

Ribeiro, Luisa Treptow ^{[2
]}

Adamski, Janice ^{[2
]}

dos Santos, Flavio Alves ^{[2
]}

Moreira, Andre Vinicius Rodrigues ^{[3
]}

Lameirao, Antonio Manoel Matta dos Santos ^{[3
]}

de Toledo, Roberto Farias ^{[3
]}

Oseas de Filho, Antonio C. ^{[4
]}

Andrade, Cid Marcos Goncalves ^{[1
]}

Lima, Oswaldo Curty da Motta ^{[1
]}

机构：

[1] State Univ Maringa UEM, Dept Chem Engn, BR-87020900 Maringa, PR, Brazil

[2] Pix Force Tecnol SA, BR-90240200 Porto Alegre, RS, Brazil

[3] Light Serv Eletricidade SA, BR-20211050 Rio De Janeiro, RJ, Brazil

[4] Fed Univ Piaui UFPI, Dept Elect Engn & Comp Sci, BR-64049550 Teresina, PI, Brazil

来源：

SENSORS | 2024年 / 24卷 / 15期

关键词：

orbital monitoring system; non-technical losses (NTLs); electricity distribution networks; Sentinel-2 satellite imagery; computer vision; urban areas segmentation; DISTRIBUTION NETWORKS; THEFT;

D O I：

10.3390/s24154924

中图分类号：

O65 [分析化学];

学科分类号：

070302 ; 081704 ;

摘要：

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.

引用

页数：23

共 35 条

[31] Machine learning-based retrieval of benthic reflectance and Posidonia oceanica seagrass extent using a semi-analytical inversion of Sentinel-2 satellite data
Traganos, Dimosthenis
Reinartz, Peter
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2018, 39 (24) : 9428 - 9452
[32] Agricultural burned area detection using an integrated approach utilizing multi spectral instrument based fire and vegetation indices from Sentinel-2 satellite
Deshpande, Monish Vijay
Pillai, Dhanyalekshmi
Jain, Meha
METHODSX, 2022, 9
[33] Comparative evaluation of performances of algae indices, pixel- and object-based machine learning algorithms in mapping floating algal blooms using Sentinel-2 imagery
Colkesen, Ismail
Ozturk, Muhammed Yusuf
Altuntas, Osman Yavuz
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT, 2024, 38 (04) : 1613 - 1634
[34] Comparative evaluation of performances of algae indices, pixel- and object-based machine learning algorithms in mapping floating algal blooms using Sentinel-2 imagery
Ismail Colkesen
Muhammed Yusuf Ozturk
Osman Yavuz Altuntas
Stochastic Environmental Research and Risk Assessment, 2024, 38 : 1613 - 1634
[35] Designing a European-Wide Crop Type Mapping Approach Based on Machine Learning Algorithms Using LUCAS Field Survey and Sentinel-2 Data
Ghassemi, Babak
Dujakovic, Aleksandar
Zoltak, Mateusz
Immitzer, Markus
Atzberger, Clement
Vuolo, Francesco
REMOTE SENSING, 2022, 14 (03)

← 1 2 3 4 →