Integration of machine learning methods in the electric power sector to reduce commercial losses: A review of practice in a grid organization
Abstract
The integration of technological advances into an intelligent grid optimizes the energy system. The dataset provided by smart metering devices allows us to capture and transmit half-hourly electricity consumption profiles, events of poor quality of electricity, tampering attempts, consumption parameters and service events. Unfortunately, the problem of unaccounted consumption has remained relevant even with the use of smart meters. Moreover, today non-technical losses, as a consequence of intentional fraud consumption, are one of the main and very real problems for electric power (EP) distribution companies. There can be no doubt that electricity theft is a problem that affects the efficiency and profitability of power companies. However, the use of technologies can solve such problems. To solve this problem, this article suggests several solutions for classifying time series in order to detect anomalies in electricity consumption. The paper tried to prove the hypothesis about the applicability of artificial intelligence technologies based on neural network training for analyzing data on electricity consumption by consumers. The goal of our research was to develop our own neural network model for enhancing electricity theft. As a result of our work, we concluded that a convolutional neural network (CNN) based on time series classification is an effective tool for finding and combating unaccounted consumption. These results highlight the potential of the proposed method for practical applications in the electricity market, as it can provide reliable and timely information for energy management. The theoretical significance of the study lies in the fact that the work attempts to use machine learning methods to improve theft-detection accuracy.
Acknowledgments
The work was supported by a grant from the Tatarstan Academy of Science, provided to higher educational institutions, scientific and other organizations for the development of human resources, specifically to encourage their academic and research staff to defend doctoral dissertations and carry out research projects.
Downloads
References
Putilova N.N., Proskuryakova M.M. (2015) Reduction of commercial losses of electric energy in electric grids. Business. Education. Law. Bulletin of Volgograd Business Institute, vol. 4(33), pp. 108–112 (in Russian).
TASS (2024) The UES said that Russia's energy consumption growth is 3.7%. Available at: https://tass.ru/ekonomika/20657729 (accessed 25 November 2025) (in Russian).
Fardiev I.Sh., Kurbangaliev T.R., Usmanov R.R., Yakhin S.R. (2023) The use of artificial intelligence in the form of neural networks for detection of fraud consumption. Experience of use in JSC "Grid Company". Electric power. Transmission and distribution, no. 6(81), pp. 28–31 (in Russian).
Lukovnikov D.V. (2017) Unaccounted and unmetered consumption of electricity. Proceedings of Bratsk State University. Series: Natural and Engineering Sciences, vol. 1, pp. 43–47 (in Russian).
Messinis G.M., Hatziargyriou N.D. (2018) Review of non-technical loss detection methods. Electric Power Systems Research, vol. 158, pp. 250–266. https://doi.org/10.1016/j.epsr.2018.01.005
Vinogradov A.V., Borodin M.V., Yurov D.Yu. (2012) Prospects for the development of electricity metering systems. News from higher educational institutions of the Black Earth Region (Vesti vysshikh uchebnykh zavedeniy Chernozem'ya), no. 2, pp. 10–15 (in Russian).
Ghasemi A.A., Gitizadeh M. (2018) Detection of illegal consumers using pattern classification approach combined with Levenberg-Marquardt method in smart grid. International Journal of Electrical Power & Energy Systems, vol. 99, pp. 363–375. https://doi.org/10.1016/j.ijepes.2018.01.036
Lepolesa L.J., Achari S., Cheng L. (2022) Electricity theft detection in smart grids based on deep neural network. IEEE Access, vol. 10, pp. 39638–39655. https://doi.org/10.1109/ACCESS.2022.3166146
Louw Q., Bokoro P. (2019) An alternative technique for the detection and mitigation of electricity theft in South Africa. SAIEE Africa Research Journal, vol. 110, no. 4, pp. 209–216. https://doi.org/10.23919/SAIEE.2019.8864147
Xia R., Gao Y., Zhu Y., et al. (2023) An attention-based wide and deep CNN with dilated convolutions for detecting electricity theft considering imbalanced data. Electric Power Systems Research, vol. 214, part A, article 108886. https://doi.org/10.1016/j.epsr.2022.108886
Tsao Y.-C., Rahmalia D., Lu J.-C. (2024) Machine-learning techniques for enhancing electricity theft detection considering transformer reliability and supply interruptions. Energy Reports, vol. 12, pp. 3048–3064. https://doi.org/10.1016/j.egyr.2024.08.068
Zidi S., Mihoub A., Qaisar S.M., et al. (2023) Theft detection dataset for benchmarking and machine learning based classification in a smart grid environment. Journal of King Saud University – Computer and Information Sciences, vol. 35, no. 1, pp. 13–25. https://doi.org/10.1016/j.jksuci.2022.05.007
Makshanov A.V., Zhuravlev A.E., Tyndykar L.N. (2024) Big data. St. Petersburg: Lan.
He K., Zhang X., Ren S., Sun J. (2016) Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 770–778. https://doi.org/10.1109/CVPR.2016.90
LeCun Y., Boser B., Denker J.S., et al. (1989) Backpropagation applied to handwritten zip code recognition. Neural Computation, vol. 1, no. 4, pp. 541–551. https://doi.org/10.1162/neco.1989.1.4.541
Krizhevsky A., Sutskever I., Hinton G.E. (2017) ImageNet classification with deep convolutional neural networks. Communications of the ACM, vol. 60, no. 6, pp. 84–90. https://doi.org/10.1145/3065386
Saqib S.M., Mazhar T., Iqbal M., et al. (2024) Deep learning-based electricity theft prediction in non-smart grid environments. Heliyon, vol. 10, no. 15, e35167. https://doi.org/10.1016/j.heliyon.2024.e35167
Adil M., Javaid N., Qasim U., et al. (2020) LSTM and bat-based RUSBoost approach for electricity theft detection. Applied Sciences, vol. 10, no. 12, article 4378. https://doi.org/10.3390/app10124378
Zheng Z., Yang Y., Niuet X., et al. (2017) Wide and deep convolutional neural networks for electricity-theft detection to secure smart grids. IEEE Transactions on Industrial Informatics, vol. 14, no. 4, pp. 1606–1615. https://doi.org/10.1109/TII.2017.2785963
Buzau M.-M., Tejedor-Aguilera J., Cruz-Romero P., Gómez-Expósito A. (2020) Hybrid deep neural networks for detection of non-technical losses in electricity smart meters. IEEE Transactions on Power Systems, vol. 35, no. 2, pp. 1254–1263. https://doi.org/10.1109/TPWRS.2019.2943115
Bjelić M., Brković B., Žarković M., Miljković T. (2024) Machine learning for power transformer SFRA based fault detection. International Journal of Electrical Power & Energy Systems, vol. 156, article 109779. https://doi.org/10.1016/j.ijepes.2023.109779
Jindal A., Dua A., Kaur K., et al. (2016) Decision tree and SVM-based data analytics for theft detection in smart grid. IEEE Transactions on Industrial Informatics, vol. 12, no. 3, pp. 1005–1016. https://doi.org/10.1109/TII.2016.2543145
Cai Q., Li P., Wang R. (2023) Electricity theft detection based on hybrid random forest and weighted support vector data description. International Journal of Electrical Power & Energy Systems, vol. 153, article 109283. https://doi.org/10.1016/j.ijepes.2023.109283
Wu R., Wang L., Hu T. (2018) AdaBoost-SVM for electrical theft detection and GRNN for stealing time periods identification. IECON 2018 – 44th Annual Conference of the IEEE Industrial Electronics Society, Washington, DC, USA, pp. 3073–3078. https://doi.org/10.1109/IECON.2018.8591459
Wu P., Liu J., Shen F. (2020) A deep one-class neural network for anomalous event detection in complex scenes. IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 7, pp. 2609–2622. https://doi.org/10.1109/TNNLS.2019.2933554
Telikani A., Gandomi A.H., Choo K.-K.R., Shen J. (2022) A cost-sensitive deep learning-based approach for network traffic classification. IEEE Transactions on Network and Service Management, vol. 19, no. 1, pp. 661–670. https://doi.org/10.1109/TNSM.2021.3112283
Aghabozorgi S., Shirkhorshidi A.S., Wah T.Y. (2015) Time-series clustering – A decade review. Information Systems, vol. 53, pp. 16–38. https://doi.org/10.1016/j.is.2015.04.007
Hasan M.N., Toma R.N., Nahid A.-A., et al. (2019) Electricity theft detection in smart grid systems: A CNN-LSTM based approach. Energies, vol. 12(17), article 3310. https://doi.org/10.3390/en12173310
Esmael A.A., da Silva H.H., Ji T., da Silva Torres R. (2021) Non-technical loss detection in power grid using information retrieval approaches: A comparative study. IEEE Access, vol. 9, pp. 40635–40648. https://doi.org/10.1109/ACCESS.2021.3064858
Viegas J.L., Esteves P.R., Vieira S.M. (2018) Clustering-based novelty detection for identification of non-technical losses. International Journal of Electrical Power & Energy Systems, vol. 101, pp. 301–310. https://doi.org/10.1016/j.ijepes.2018.03.031
Gubayeva L. (2023) Smart meters, big data and new competencies: How the digital transformation is going in the "Network Company". Realnoe Vremya (Real Time). Available at: https://realnoevremya.ru/articles/294427-kak-idet-cifrovaya-transformaciya-v-setevoy-kompanii?erid=Kra23mmud (accessed 25 November 2025) (in Russian).
Depuru S.S.S.R., Wang L., Devabhaktuni V., Nelapati P. (2011) A hybrid neural network model and encoding technique for enhanced classification of energy consumption data. 2011 IEEE Power and Energy Society General Meeting, Detroit, MI, USA, pp. 1–8. https://doi.org/10.1109/PES.2011.6039050
Yakhin Sh. (2025) An electric power consumption dataset for ML-training. GitHub. Available at: https://github.com/tlegenda/consumption_dataset (accessed 25 November 2025).
Wang D., Gan J., Mao J., et al. (2023) Forecasting power demand in China with a CNN-LSTM model including multimodal information. Energy, vol. 263, article 126012. https://doi.org/10.1016/j.energy.2022.126012
Zhang J., Khayatnezhad M., Ghadimi N. (2022) Optimal model evaluation of the proton-exchange membrane fuel cells based on deep learning and modified African Vulture Optimization Algorithm. Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, vol. 44(1), pp. 287–305. https://doi.org/10.1080/15567036.2022.2043956
Han E., Ghadimi N. (2022) Model identification of proton-exchange membrane fuel cells based on a hybrid convolutional neural network and extreme learning machine optimized by improved honey badger algorithm. Sustainable Energy Technologies and Assessments, vol. 52, article 102005. https://doi.org/10.1016/j.seta.2022.102005
Guo H., Gu W., Khayatnezhad M., Ghadimi N. (2022) Parameter extraction of the SOFC mathematical model based on fractional order version of dragonfly algorithm. International Journal of Hydrogen Energy, vol. 47(57), pp. 24059–24068. https://doi.org/10.1016/j.ijhydene.2022.05.190
Papernot N., McDaniel P. (2018) Deep k-nearest neighbors: Towards confident, interpretable and robust deep learning. arXiv:1803.04765. https://doi.org/10.48550/arXiv.1803.04765
Fawaz I.H., Forestier G., Weber J., et al. (2019) Deep learning for time series classification: A review. Data Mining and Knowledge Discovery, vol. 33, pp. 917–963. https://doi.org/10.1007/s10618-019-00619-1
Khattak A., Bukhsh R., Aslam S., et al. (2022) A hybrid deep learning-based model for detection of electricity losses using big data in power systems. Sustainability, vol. 14(20), article 13627. https://doi.org/10.3390/su142013627
Feng X., Hui H., Liang Z., et al. (2020) A novel electricity theft detection scheme based on text convolutional neural networks. Energies, vol. 13(21), article 5758. https://doi.org/10.3390/en13215758
Copyright (c) 2025 National Research University Higher School of Economics
This work is licensed under a Creative Commons Attribution 4.0 International License.
