Dağıtım Şebekelerinde Teknik Olmayan Kayıpların Makine Öğrenme Yöntemleriyle Tespiti

Year 2025, Volume: 4 Issue: 1, 192 - 205, 18.02.2025

Mahmut Türk , Cem Haydaroglu , Heybet Kılıç

https://doi.org/10.62520/fujece.1551601

Abstract

Bu çalışma, elektrik şebeke sistemlerinde enerji hırsızlığından kaynaklanan teknik olmayan kayıpların (NTL) oluşturduğu ciddi sürdürülebilirlik ve güvenilirlik sorununa odaklanmaktadır. Bu kayıpları azaltmak amacıyla, farklı kaçak türlerinin (gerilim kaçağı, akım kaçağı ve gerilim-akım kaçağı) tespitinde derin öğrenme mimarilerinden yararlanan yapay zeka tabanlı bir yaklaşım öneriyoruz. Literatürdeki çalışmalardan farklı olarak veri seti iki boyutlu matrislere dönüştürülerek, günümüzün popüler yaklaşımları olan Convolutional Neural Network (CNN) ve Long Short-Term Memory (LSTM) modelleri ile analiz edilmiş; CNN, %97,50 doğruluk oranı ile LSTM'nin %64,17 doğruluk oranını geride bırakmıştır. Ayrıca, klasik yöntemlerden, k-En Yakın Komşu (k-NN) yöntemi ile 67,5 doğruluk oranı ve Destek Vektör Makineleri (SVM) yöntemi ile 62,25 doğruluk oranı elde edilmiştir. Bu gibi geleneksel yöntemlerle yapılan karşılaştırmalar, CNN'in karmaşık kaçak desenlerini belirlemedeki üstünlüğünü ortaya koymuştur. Bulgular, CNN'in akıllı şebeke sistemlerine entegre edilerek gerçek zamanlı hırsızlık tespiti için güvenilir bir araç olarak kullanılma potansiyelini vurgulamaktadır. Gelecekteki araştırmalar, gerçek zamanlı verilerin entegrasyonunu ve hibrit model yaklaşımlarını inceleyerek bu çözümün ölçeklenebilirliğini ve etkinliğini daha da artırmayı hedefleyecektir.

Keywords

Teknik olmayan kayıplar (TOK), Kaçak elektrik tespiti, Derin öğrenme, Konvolüsyonel sinir ağı (CNN), Akıllı şebeke

Machine Learning-Based Detection of Non-Technical Losses in Power Distribution Networks

Abstract

This study focuses on the serious sustainability and reliability problem caused by non-technical losses (NTL) due to energy theft in electrical grid systems. In order to reduce these losses, we propose an artificial intelligence-based approach that utilizes deep learning architectures in the detection of different types of leakage (voltage leakage, current leakage and voltage-current leakage). Unlike the studies in the literature, the data set is converted into two-dimensional matrices and analyzed with today's popular approaches, Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) models; CNN surpassed LSTM's 64.17% accuracy rate with 97.50% accuracy rate. In addition, from the classical methods, 67.5 accuracy rate was obtained with the k-Nearest Neighbor (k-NN) method and 62.25 accuracy rate was obtained with the Support Vector Machines (SVM) method. Comparisons with such traditional methods have revealed the superiority of CNN in determining complex leakage patterns. The findings highlight the potential of CNN to be used as a reliable tool for real-time theft detection by integrating it into smart grid systems. Future research will aim to further increase the scalability and effectiveness of this solution by examining the integration of real-time data and hybrid model approaches.

Keywords

Non-technical losses (NTL), Electricity theft detection, Deep learning, Convolutional neural network(CNN), SmartGrid

Ethical Statement

There is no need to obtain ethics committee permission for the article prepared. "There is no conflict of interest with any person/institution in the article prepared.

Thanks

I would like to thank Diyarbakır Organized Industrial Zone Directorate for providing support during the conduct of this study.

References

• L. J. Lepolesa, S. Achari, and L. Cheng, "Electricity theft detection in smart grids based on deep neural network," IEEE Access, vol. 10, pp. 39638–39655, 2022.
• T. Sharma, K. K. Pandey, D. K. Punia, and J. Rao, "Of pilferers and poachers: Combating electricity theft in India," Energy Res. Soc. Sci., vol. 11, pp. 40–52, 2016.
• S. Sahoo, D. Nikovski, T. Muso, and K. Tsuru, "Electricity theft detection using smart meter data," in Proceedings of the 2015 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), Washington, DC, USA, 18–20 February 2015, pp. 1–5.
• E. Villar-Rodriguez, J. Del Ser, I. Oregi, M. N. Bilbao, and S. Gil-Lopez, "Detection of non-technical losses in smart meter data based on load curve profiling and time series analysis," Energy, vol. 137, pp. 118–128, 2017.
• H. O. Henriques, R. L. S. Corrêa, M. Z. Fortes, B. S. M. C. Borba, and V. H. Ferreira, "Monitoring technical losses to improve non-technical losses estimation and detection in LV distribution systems," Measurement, vol. 161, p. 107840, 2020.
• E. S. Ibrahim, "Management of loss reduction projects for power distribution systems," Elect. Power Syst. Res., vol. 55, pp. 49–56, 2000.
• K. M. Ghori, M. Imran, A. Nawaz, R. A. Abbasi, A. Ullah, and L. Szathmary, "Performance analysis of machine learning classifiers for non-technical loss detection," J. Ambient. Intell. Humaniz. Comput., pp. 1–16, 2023.
• M. Çelikpençe, "Elektrik dağıtım şebekelerinde teknik olmayan kayıp kaçakların makine öğrenmesi ile tespiti," 2023.
• K. M. Ghori, R. A. Abbasi, M. Awais, M. Imran, A. Ullah, and L. Szathmary, "Performance analysis of different types of machine learning classifiers for non-technical loss detection," IEEE Access, vol. 8, pp. 16033–16048, 2019.
• G. Figueroa, Y. S. Chen, N. Avila, and C. C. Chu, "Improved practices in machine learning algorithms for NTL detection with imbalanced data," in Proceedings of the 2017 IEEE Power & Energy Society General Meeting, pp. 1–5, 2017.
• M. B. Capeletti, B. K. Hammerschmitt, R. G. Negri, F. G. K. Guarda, L. R. Prade, N. Knak Neto, and A. D. R. Abaide, "Identification of nontechnical losses in distribution systems adding exogenous data and artificial intelligence,"Energies, vol. 15, no. 23, p. 8794, 2022.
• M. Žarković and G. Dobrić, "Artificial Intelligence for Energy Theft Detection in Distribution Networks," Energies, vol. 17, no. 7, p. 1580, 2024.
• A. B. Pengwah, R. Razzaghi, and L. L. Andrew, "Model-less non-technical loss detection using smart meter data," IEEE Trans. Power Deliv., vol. 38, no. 5, pp. 3469–3479, 2023.
• M. Jené-Vinuesa, M. Aragüés-Peñalba, and A. Sumper, "Comprehensive Data-Driven Framework for Detecting and Classifying Non-Technical Distribution Losses," IEEE Access, 2024.
• M. A. Souza, H. T. Gouveia, A. A. Ferreira, R. M. de Lima Neta, O. Nóbrega Neto, M. M. da Silva Lira, and R. R. de Aquino, "Detection of Non-Technical Losses on a Smart Distribution Grid Based on Artificial Intelligence Models," Energies, vol. 17, no. 7, p. 1729, 2024.
• Y. Kara and A. Aksu, "Tüketicilerde Kaçak Elektrik Kullanımının Akıllı Sayaç Verisi Üzerinden Gradyan Artırmalı Karar Ağacı Tabanlı Makine Öğrenmesi Yöntemleriyle Tespiti,"J. Investig. Eng. Technol., vol. 6, no. 1, pp. 1–12, 2023.
• M. Salman Saeed, M. W. Mustafa, U. U. Sheikh, T. A. Jumani, I. Khan, S. Atawneh, and N. N. Hamadneh, "An efficient boosted C5.0 decision-tree-based classification approach for detecting non-technical losses in power utilities," Energies, vol. 13, no. 12, p. 3242, 2020.
• S. Chatterjee, V. Archana, K. Suresh, R. Saha, R. Gupta, and F. Doshi, "Detection of non-technical losses using advanced metering infrastructure and deep recurrent neural networks," in 2017 IEEE International Conference on Environment and Electrical Engineering and 2017 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe), pp. 1–6, 2017.
• H. M. Khan, F. Jabeen, A. Khan, S. A. Badawi, C. Maple, and G. Jeon, "Hybrid non-technical-loss detection in fog-enabled smart grids," Sust. Energy Technol. Assess., vol. 65, p. 103775, 2024.
• Y. Xing, L. Guo, Z. Xie, L. Cui, L. Gao, and S. Yu, "Non-technical losses detection in smart grids: An ensemble data-driven approach," in 2020 IEEE 26th International Conference on Parallel and Distributed Systems (ICPADS), pp. 563–568, Dec. 2020.
• R. M. R. Barros, E. G. da Costa, and J. F. Araujo, "Maximizing the financial return of non-technical loss management in power distribution systems," IEEE Trans. Power Syst., vol. 37, no. 2, pp. 1634–1641, 2021.
• M. S. Saeed, M. W. Mustafa, U. U. Sheikh, T. A. Jumani, and N. H. Mirjat, "Ensemble bagged tree based classification for reducing non-technical losses in Multan Electric Power Company of Pakistan," Electronics, vol. 8, no. 8, p. 860, 2019.
• S. C. Huang, Y. L. Lo, and C. N. Lu, "Non-technical loss detection using state estimation and analysis of variance," IEEE Trans. Power Syst., vol. 28, no. 3, pp. 2959–2966, 2013.
• Ç. Yurtseven, "The causes of electricity theft: An econometric analysis of the case of Turkey," Utilities Policy, vol. 37, pp. 70–78, 2015.
• S. C. Yip, K. Wong, W. P. Hew, M. T. Gan, R. C. W. Phan, and S. W. Tan, "Detection of energy theft and defective smart meters in smart grids using linear regression," Int. J. Electr. Power Energy Syst., vol. 91, pp. 230–240, 2017.
• E. Villar-Rodriguez, J. Del Ser, I. Oregi, M. N. Bilbao, and S. Gil-Lopez, "Detection of non-technical losses in smart meter data based on load curve profiling and time series analysis," Energy, vol. 137, pp. 118–128, 2017.
• Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436–444, 2015.
• A. H. Mirza, "Low complexity efficient online learning algorithms using LSTM networks," M.S. thesis, Bilkent Univ., Ankara, Turkey, 2018.
• Y. Yu, X. Si, C. Hu, and J. Zhang, "A review of recurrent neural networks: LSTM cells and network architectures," Neural Comput., vol. 31, no. 7, pp. 1235–1270, 2019.
• H. Sak, A. W. Senior, and F. Beaufays, "Long short-term memory recurrent neural network architectures for large scale acoustic modeling," in Proc. ICASSP, 2014.
• P. K. Syriopoulos, N. G. Kalampalikis, S. B. Kotsiantis, and M. N. Vrahatis, "k NN Classification: a review," Ann. Math. Artif. Intell., pp. 1–33, 2023.
• Ö. Tomak and O. Ö. Mengi, "K-Nearest Neighbor Classification of Harmonics Using Akaike Information Criterion," Karadeniz Fen Bilim. Derg., vol. 7, no. 1, pp. 1–8, 2017.
• J. Ringelberg and E. Van Gool, "On the combined analysis of proximate and ultimate aspects in diel vertical migration (DVM) research," Hydrobiologia, vol. 491, pp. 85–90, 2003.
• K. Blachowiak-Samolyk, S. Kwasniewski, K. Richardson, K. Dmoch, E. Hansen, H. Hop, et al., "Arctic zooplankton do not perform diel vertical migration (DVM) during periods of midnight sun," Mar. Ecol. Prog. Ser., vol. 308, pp. 101–116, 2006.
• Ö. Türk, "Classification of electroencephalogram records related to cursor movements with a hybrid method based on deep learning," Int. J. Imaging Syst. Technol., vol. 31, no. 4, pp. 2322–2333, 2021.