Artificial intelligence model and balancing techniques for controlling non-technical losses in the electrical energy sector
DOI:
https://doi.org/10.15649/2346030X.5635Keywords:
energy, artificial intelligence, model, non-technical loss, userAbstract
Non-technical losses (NTLs) in electrical power distribution systems represent a critical problem due to their economic and operational impact, as they are primarily associated with fraud, tampering, and administrative errors. This article proposes the design and evaluation of a hybrid deep learning model for the automatic detection of NTLs, integrating convolutional neural networks (CNNs), long short-term memory networks (LSTMs), and dense neural networks (DNNs). To mitigate the class imbalance characteristic of this type of problem, synthetic resampling techniques were applied, specifically SMOTE, BorderlineSMOTE, and ADASYN, analyzing their effect on model performance. Validation was performed using a real dataset from a power distribution company in the municipality of Aguachica, Colombia, consisting of 44,231 users and 365 variables per record. The results show that the BorderlineSMOTE + CNN–LSTM–DNN combination achieves the best overall performance, reaching an accuracy of 74.47%, along with improvements in key metrics such as recall, F1 score, and AUC. These findings demonstrate that integrating deep architectures with advanced balancing techniques is an effective strategy for PNT detection in real-world electrical environments.
References
[1] H. Córdoba, J. Ramírez, and A. Méndez, “Identification of the characteristics involved in the detection of non-technical losses for two Colombian energy companies,” Energy Eng. Technol. J., vol. 23, no. 2, pp. 45–61, 2020. [Online]. Available: https://www.redalyc.org/journal/430/43052680008/html/.
[2] F. A. A. Paredes, R. C. de Souza, C. C. O. Ramos, and J. P. Papa, “Data-driven machine learning methods for non-technical losses of electrical energy detection: A state-of-the-art review,” Energies, vol. 16, no. 21, art. no. 7460, Nov. 2023, doi: 10.3390/en16217460.
[3] V. Orjales, M. J. Rodríguez, A. Gómez, and J. M. Esteban, “Using regression analysis to identify patterns of non-technical losses on power utilities,” in Proc. KES 2010 – Knowl.-Based Intell. Inf. Eng. Syst., Cardiff, U.K., 2010, pp. 410–419, doi: 10.1007/978-3-642-15387-7_45.
[4] A. H. T. Alfarra, B. A. Attia, and C. S. M. El Safty, “Non-technical loss detection for metered customers using support vector machines,” Res., Eng. Policy Q. J., vol. 16, no. 4, pp. 1–10, 2024, doi: 10.24084/repqj16.353.
[5] A. Hasan, R. Toma, A. A. Nahid, M. Islam, and J. M. Kim, “Electricity theft detection in smart grid systems: A CNN–LSTM based approach,” Energies, vol. 12, no. 14, art. no. 3310, Jul. 2019, doi: 10.3390/en12173310.
[6] T. Sadeghi, Z. Nasr, B. R. Tahour, H. Nowzari-Dalini, and M. Zolghadr, “Electricity theft detection in power grids with a hybrid convolutional neural network–support vector machine model,” in Proc. 5th Int. Conf. Future Netw. Distrib. Syst., Dubai, UAE, Oct. 2022, pp. 1–6, doi: 10.1145/3508072.3508078.
[7] D. Ali, M. S. Ali, M. Imamuddin, and M. E. E. Rizvi, “A novel framework for electricity theft detection in smart grids using principal component analysis and machine learning techniques,” Heliyon, vol. 10, no. 15, art. no. e35167, Aug. 2024, doi: 10.1016/j.heliyon.2024.e35167.
[8] V. S. Viegas, P. R. Esteves, and S. M. Vieira, “Clustering-based novelty detection for identification of non-technical losses,” Int. J. Electr. Power Energy Syst., vol. 101, pp. 301–311, Oct. 2018, doi: 10.1016/j.ijepes.2018.03.031.
[9] Pamir, N. Javaid, U. Qasim, A. S. Yahaya, E. H. Alkhammash and M. Hadjouni, "Non-Technical Losses Detection Using Autoencoder and Bidirectional Gated Recurrent Unit to Secure Smart Grids," in IEEE Access, vol. 10, pp. 56863-56875, 2022, doi: 10.1109/ACCESS.2022.3171229.
[10] M. Z. H. Jesmeen, G. R. Murthy, J. Hossen, J. Ganesan, A. A. Aziz, and K. Tawsif, “Detecting abnormal electricity usage using unsupervised learning model in unlabeled data,” Int. J. Adv. Appl. Sci., vol. 8, no. 9, pp. 102–111, Sep. 2021, doi: 10.21833/ijaas.2021.09.014.
[11] S. Chen, B. Mulgrew, and P. M. Grant, “A clustering technique for digital communications channel equalization using radial basis function networks,” IEEE Trans. Neural Netw., vol. 4, no. 4, pp. 570–578, Jul. 1993, doi: 10.1109/72.238311.
[12] J. Wang, “Fundamentals of erbium-doped fiber amplifiers,” IEEE J. Quantum Electron., submitted for publication.
[13] Y. Yorozu, M. Hirano, K. Oka, and Y. Tagawa, “Electron spectroscopy studies on magneto-optical media and plastic substrate interfaces,” IEEE Transl. J. Magn. Jpn., vol. 2, pp. 740–741, Aug. 1987.
[14] T. Duncombe, “Infrared navigation—Part I: An assessment of feasibility,” IEEE Trans. Electron Devices, vol. ED-11, pp. 34–39, Jan. 1959.
[15] S. Chen, B. Mulgrew, and P. M. Grant, “Clustering techniques for channel equalization using RBF networks,” IEEE Trans. Neural Netw., vol. 4, no. 4, pp. 570–578, 1993.
[16] M. Young, The Technical Writer’s Handbook. Mill Valley, CA, USA: Univ. Science, 1989.
[17] N. Javaid et al., “Unsupervised learning approaches for anomaly detection in smart grids,” IEEE Access, vol. 10, pp. 56863–56875, 2022.
[18] V. S. Viegas et al., “Novelty detection techniques for non-technical losses,” Int. J. Electr. Power Energy Syst., vol. 101, pp. 301–311, 2018.
[19] D. Ali et al., “Machine learning-based PCA framework for electricity theft detection,” Heliyon, vol. 10, art. no. e35167, 2024.
[20] T. Sadeghi et al., “Hybrid deep learning models for electricity theft detection,” in Proc. Int. Conf. Future Netw. Distrib. Syst., 2022.
[21] A. Hasan et al., “Deep learning models for smart grid fraud detection,” Energies, vol. 12, art. no. 3310, 2019.
[22] F. A. A. Paredes et al., “Machine learning techniques for non-technical loss detection: A comprehensive review,” Energies, vol. 16, art. no. 7460, 2023.
[23] N. M. Elshennawy, D. M. Ibrahim, y A. M. Gab Allah, “An efficient electricity theft detection based on deep learning,” Sci. Rep., vol. 15, art. no. 12866, 15 abr. 2025, doi: 10.1038/s41598-025-93140-z.
[24] M. Sleiman, C. Dagli, y R. Bo, “Electricity theft detection with an adaptive deep learning architecture,” Procedia Comput. Sci., vol. 268, pp. 392–401, Oct. 2025, doi: 10.1016/j.procs.2025.08.218.
Downloads
Published
How to Cite
Downloads
Issue
Section
License
Copyright (c) 2026 AiBi Journal of Research, Administration and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
The journal offers open access under a Creative Commons Attibution License
This work is under license Creative Commons Attribution (CC BY 4.0).
