Electricity Theft Detection: Pareto Optimization & Scalable Ensemble

by Archynetys Economy Desk
  • Northeast Group. Electricity theft and non-technical losses: global markets, solutions, and vendors. (2021). https://northeast-group.com/2021/10/20/electricity-theft-non-technical-losses/

  • Hasan, M. N., Toma, R. N., Nahid, A. A., Islam, M. M. M. & Kim, J. M. Electricity theft detection in smart grid systems: a CNN–LSTM based approach. Energies 123310. (2019).

    Google Scholar

  • Kumar, P. et al. Smart grid metering networks: a survey on security, privacy and open research issues. IEEE Commun. Surv. Tutor. 212886–2927. (2019).

    Google Scholar

  • Do, V. et al. Spatiotemporal distribution of power outages with climate events and social vulnerability in the USA. Nat. Common. 1438084. (2023).

    Google Scholar

  • Boran, F., Mutlu, I. & Akbaş, A. A comprehensive review of advanced metering infrastructure and its applications in smart grids. Renew. Sustain. Energy Rev. 175113183. (2023).

    Google Scholar

  • Ding, J. et al. Cyber threats to smart grids: review, taxonomy, potential solutions, and future directions. Energies 156799. (2022).

    Google Scholar

  • Abdul, R. F. & Saravanan, A. A novel data transmission model using hybrid encryption scheme for preserving data integrity. Adv. Technol. Innov. 1015–28. (2025).

    Google Scholar

  • Lai, J. & Heng, S. H. Secure file storage on cloud using hybrid cryptography. J. Inf. Web Eng. 11–18. (2022).

    Google Scholar

  • Gržinić, M. Springer, Singapore,. Data encryption in fog computing using hybrid cryptography with integrity check. In Proc. Int. Conf. Comput. Sci. 627–638 (2023). https://doi.org/10.1007/978-981-19-6525-8_48

  • Iqbal, M. S. et al. A critical review of technical case studies for electricity theft detection in smart grids. Energy Convers. Manag X. 26100965. (2025).

    Google Scholar

  • Zulu, C. L. & Dzobo, O. Real-time power theft monitoring and detection system with double-connected data capture system. Int. J. Renew. Energy Res. 133065–3083. (2023).

    Google Scholar

  • Implementation of a sustainable. security architecture using radio frequency identification (RFID) technology for access control. Preprint at (2023). https://arxiv.org/abs/2304.04628

  • Kumar, A. et al. Cost-efficient and scalable electricity theft detection in smart grids: challenges, tools, and future directions. IEEE Access. 1180234–80256. (2023).

    Google Scholar

  • Arias-Marín, C., Barragán-Escandón, A., Toledo-Orozco, M. & Serrano-Guerrero, X. Methodological validation of machine learning models for non-technical loss detection in electric power systems. Appl. Sci. 153912. (2025).

    Google Scholar

  • Jiménez, D., Barrera, J. & Cancela, H. Communication network reliability under geographically correlated failures using probabilistic seismic hazard analysis. IEEE Access. 1131341–31354. (2023).

    Google Scholar

  • Whalen, J. K. A review on game theory with smart grid security. Preprint at (2023). https://doi.org/10.48550/arxiv.2304.11738

  • Kim, S. et al. Data-driven approaches for energy theft detection: a comprehensive review. Energies 173057. (2024).

    Google Scholar

  • Mu, T., Yu, Y., Feng, G., Luo, H. & Yang, H. Detecting anomalous electricity consumption with transformer and synthesized anomalies. PeerJ Comput. Sci. 9e1721. (2023).

    Google Scholar

  • Shabad, P. K., Alrashide, A. & Mohammed, O. Anomaly detection in smart grids using machine learning. In Proc. IEEE IECON 1–8 (IEEE, 2021). https://doi.org/10.1109/IECON48115.2021.9589851

  • Banik, S. et al. Anomaly detection techniques in smart grid systems: a review. In Proc. IEEE World AI IoT Congr. 331–337 (IEEE, 2023). https://doi.org/10.1109/AIIoT58121.2023.10174485

  • Sun, Y., Sun, X., Hu, T. & Liu, H. Smart grid theft detection based on hybrid multi-time scale neural network. Appl. Sci. 135710. (2023).

    Google Scholar

  • Wang, Y. & Jin, S. A convolution–non-convolution parallel deep network for electricity theft detection. Sustainability 1510127. (2023).

    Google Scholar

  • El-Toukhy, A. T. et al. Electricity theft detection using deep reinforcement learning in smart power grids. IEEE Access. 1159558–59574. (2023).

    Google Scholar

  • Li, P. et al. Multi-model running latency optimization in an edge computing paradigm. Sensors 226097. (2022).

    Google Scholar

  • Jain, T. et al. Latency-memory optimized splitting of convolution neural networks for resource-constrained edge devices. In Proc. COMSNETS 531–539 (IEEE, 2022). https://doi.org/10.1109/COMSNETS53615.2022.9668356

  • Almalki, A. J. Unsupervised learning with hybrid models for detecting electricity theft in smart grids. IEEE Access. 12187027–187040. (2024).

    Google Scholar

  • Hussain, S. et al. A novel unsupervised feature-based approach for electricity theft detection using robust PCA. Int. Trans. Electr. Energy Syst. 30e12572. (2020).

    Google Scholar

  • Aslam, Z. et al. A new clustering-based semi-supervised method to restrict anomalous electricity consumption. Electr. Eng. 1066431–6448. (2024).

    Google Scholar

  • Gunturi, S. K. & Sarkar, D. Ensemble machine learning models for the detection of energy theft. Electr. Power Syst. Res. 192106904. (2021).

    Google Scholar

  • Fei, K., Li, Q. & Zhu, C. Non-technical losses detection using missing values’ pattern and neural architecture search. Int. J. Electr. Power Energy Syst. 134107410. (2022).

    Google Scholar

  • Elhassan, A. et al. ILA4: overcoming missing values in machine learning datasets. J. King Saud Univ. Comput. Inf. Sci. 344284–4295. (2022).

    Google Scholar

  • Tureczek, A., Nielsen, P. S. & Madsen, H. Electricity consumption clustering using smart meter data. Energies 11859. (2018).

    Google Scholar

  • Huang, M. W. et al. Data preprocessing issues for incomplete medical datasets. Expert Syst. 33432–438. (2016).

    Google Scholar

  • Patidar, S. et al. Missing data imputation for community energy demand modelling. In Proc. IBPSA-Scotland Conf. (2022).

  • Zhang, Y. et al. Missing value imputation in multivariate time series with end-to-end generative adversarial networks. Inf. Sci. 55167–82. (2021).

    Google Scholar

  • Saputra, M. D. et al. Handling missing values and unusual observations using Kalman filter. J. Phys. Conf. Ser. 1863012035. (2021).

    Google Scholar

  • Sovilj, D. et al. Extreme learning machine for missing data using multiple imputations. Neurocomputing 174220–231. (2016).

    Google Scholar

  • Jiang, Z. et al. Robust smart meter data analytics using smoothed ALS. Energies 111401. (2018).

    Google Scholar

  • Stekhoven, D. J. & Bühlmann, P. MissForest—non-parametric missing value imputation. Bioinformatics 28112–118. (2012).

    Google Scholar

  • Rahman, M. G. & Islam, M. Z. Missing value imputation using fuzzy clustering-based EM. Knowl. Inf. Syst. 46389–422. (2016).

    Google Scholar

  • Mantuano, C. et al. Data imputation methods for intermittent renewable energy sources. Energy Convers. Manag. 339119857. (2025).

    Google Scholar

  • Zhao, J., Li, Y. & Mo, H. Handling missing data using the XGBoost-based multiple imputation by chained equations regression method. Front. Artif. Intell. 81553220. (2025).

    Google Scholar

  • Li, L. et al. Estimation of missing values in heterogeneous traffic data. Knowl. Based Syst. 194105592. (2020).

    Google Scholar

  • Hernández, Á. et al. Detection of anomalies in daily activities using data from smart meters. Sensors 24515. (2024).

    Google Scholar

  • Saleh, O. A. & Cevik, M. Secure edge-based smart grid communication using lightweight authentication modeling. Discover Comput. 28110. (2025).

    Google Scholar

  • Molokomme, D. N. et al. Edge intelligence in smart grids: a survey. J. Sens. Actuator Netw. 1147. (2022).

    Google Scholar

  • Azzalini, D. et al. An empirical evaluation of deep autoencoders for anomaly detection. Energy Build. 327115069. (2025).

    Google Scholar

  • Jithish, J. et al. Distributed anomaly detection in smart grids using federated learning. IEEE Access. 117157–7179. (2023).

    Google Scholar

  • Javaid, N. et al. Adaptive synthesis to handle imbalanced big data. J. Parallel Distrib. Comput. 15344–52. (2021).

    Google Scholar

  • Marangu, D. K. et al. Comparative analysis of class imbalance handling techniques. Int. J. Artif. Intel. Appl. 1513–27. (2024).

    Google Scholar

  • Bunkhumpornpat, C., Sinapiromsaran, K. & Lursinsap, C. DBSMOTE: density-based synthetic minority over-sampling technique. Appl. Intel. 36664–684. (2012).

    Google Scholar

  • Koziarski, M. Radial-based undersampling for imbalanced data classification. Pattern Recognit. 102107262. (2020).

    Google Scholar

  • Buda, M., Maki, A. & Mazurowski, M. A. Systematic study of the class imbalance problem in CNNs. Neural Netw. 106249–259. (2018).

    Google Scholar

  • Sáez, J. A., Krawczyk, B. & Woźniak, M. Oversampling strategies in multi-class imbalanced datasets. Pattern Recognit. 57164–178. (2016).

    Google Scholar

  • Adil, M. et al. LSTM and bat-based RUSBoost approach for electricity theft detection. Appl. Sci. 104378. (2020).

    Google Scholar

  • Naeem, A. et al. Deep fractal network with light gradient boosting approach. Helion 9e18928. (2023).

    Google Scholar

  • Wu, J. C., Lu, S., Fuh, C. S. & Liu, T. L. One-class anomaly detection via novelty normalization. Comput. Vis. Image Underst. 210103226. (2021).

    Google Scholar

  • Buzau, M. M. et al. Detection of non-technical losses using smart meter data and supervised learning. IEEE Trans. Smart Grid. 102661–2670. (2019).

    Google Scholar

  • Lin, M. et al. Detection of ionospheric scintillation using XGBoost improved by SMOTE-ENN. Remote Sens. 132577. (2021).

    Google Scholar

  • Galar, M., Fernández, A., Barrenechea, E. & Herrera, F. EUSBoost: enhancing ensembles for highly imbalanced datasets. Pattern Recognit. 463460–3471. (2013).

    Google Scholar

  • Azeem, A. et al. Mitigating concept drift challenges in evolving smart grids. Energy Rep. 131369–1383. (2025).

    Google Scholar

  • Tripathi, A. K., Pandey, A. C. & Sharma, N. Hybrid adaptive sampling and pipeline machine learning for electricity theft detection. Multimed Tools Appl. 8354521–54544. (2024).

    Google Scholar

  • Liu, Z. et al. Electricity theft detection through contrastive learning. J. Wirel. Commun. Netw. 2023, 54 (2023). https://doi.org/10.1186/s13638-023-02258-z

  • Gong, X. & Ma, L. Smart meter outlier detection using PCA dimensionality reduction and K-means clustering. In Proc. 4th Int. Conf. Robotics, Automation and Intelligent Control (ICRAIC) 444–447 (IEEE, 2024). https://doi.org/10.1109/ICRAIC65937.2024.00084

  • Shamim, G. & Rihan, M. Exploratory data analytics and PCA-based dimensionality reduction for smart meter data clustering. IETE J. Res. 704159–4168. (2024).

    Google Scholar

  • Razavi, R. et al. Practical feature-engineering framework for electricity theft detection. Appl. Energy. 238481–494. (2019).

    Google Scholar

  • Dong, G. G. & Liu, H. Feature Engineering for Machine Learning and Data Analytics (CRC, 2018). https://doi.org/10.1201/9781315181080

  • Ibrahim, M. S., Dong, W. & Yang, Q. Machine learning driven smart electric power systems. Appl. Energy. 272115237. (2020).

    Google Scholar

  • Saeed, M. S. et al. Boosted C5.0 decision-tree-based classification for non-technical loss detection. Energies 133242. (2020).

    Google Scholar

  • Punmiya, R. & Choe, S. Energy theft detection using gradient boosting theft detector. IEEE Trans. Smart Grid. 102326–2329. (2019).

    Google Scholar

  • Lee, L. C. & Jemain, A. A. PCA application strategy in high dimensional forensic data. Microchem J. 169106608. (2021).

    Google Scholar

  • Gul, H. et al. Detection of non-technical losses using SOSTLink and bidirectional GRU. Appl. Sci. 103151. (2020).

    Google Scholar

  • Davila Delgado, J. M. & Oyedele, L. Deep learning with small datasets using autoencoders. Appl. Soft Comput. 112107836. (2021).

    Google Scholar

  • Khan, Z. A. et al. Electricity theft detection using supervised learning techniques. Sustainability 128023. (2020).

    Google Scholar

  • Munawar, S. et al. Electricity theft detection using a hybrid BiGRU–BiLSTM model. Sensors 227818. (2022).

    Google Scholar

  • Zheng, K. et al. A novel combined data-driven approach for electricity theft detection. IEEE Trans. Ind. Inf. 151809–1819. (2019).

    Google Scholar

  • Elshennawy, N. M., Ibrahim, D. M. & Gab Allah, A. M. Efficient electricity theft detection based on deep learning. Sci. Rep. 1512866. (2025).

    Google Scholar

  • Mbey, C. F. et al. Electricity theft detection in a smart grid using hybrid deep learning-based data analysis technique. J. Electr. Comput. Eng. 2024 (6225510). (2024).

  • Iftikhar, H. et al. Electricity theft detection in smart grid using machine learning. Front. Energy Res. 121383090. (2024).

    Google Scholar

  • He, Y. et al. Electricity theft detection based on minimal gated memory network. Sustain. Energy Grids Netw. 39101415. (2024).

    Google Scholar

  • Abro, S. A. et al. Machine learning-based electricity theft detection using support vector machines. Int. J. Electr. Comput. Eng. 141240–1250. (2024).

    Google Scholar

  • Cai, Q., Li, P. & Wang, R. Electricity theft detection based on hybrid random forest and weighted SVDD. Int. J. Electr. Power Energy Syst. 153109283. (2023).

    Google Scholar

  • Yang, X. et al. TLEL: a two-layer ensemble learning approach. Inf. Softw. Technol. 87206–220. (2017).

    Google Scholar

  • Mohammed, A. R. & Idris, R. M. Enhancing diabetes mellitus onset prediction through ensemble learning. J. Stat. Model. Anal. 611–28. (2024).

    Google Scholar

  • Al-Mayyahi, S. et al. Machine learning techniques for solar power output predicting. Int. J. Smart Grid. 898–107. (2024).

    Google Scholar

  • Hussain, S. et al. Electric theft detection in advanced metering infrastructure using Jaya-optimized boosting classifier. IET Gener Transm Distrib. 161257–1275. (2022).

    Google Scholar

  • Zhang, Y., Liu, J. & Shen, W. Review of ensemble learning algorithms in remote sensing. Appl. Sci. 128654. (2022).

    Google Scholar

  • Satyapal, K. S. & Patil, A. Enhancing electricity theft detection using XGBoost with Optuna optimization. In Proc. IEEE ISGT Asia 1–6 (IEEE, 2024). https://doi.org/10.1109/ISGTAsia61245.2024.10876295

  • Rahaman, M. A. & Idris, R. M. Enhancing energy fraud detection with ensemble learning techniques. In Proc. IEEE APEE 159–164 (IEEE, 2024). https://doi.org/10.1109/APEE60256.2024.10790893

  • Hussain, S. et al. Feature-engineered CatBoost-based supervised machine learning framework. Energy Rep. 74425–4436. (2021).

    Google Scholar

  • Ullah, A. et al. Hybrid deep neural network for electricity theft detection. Wirel. Commun. Mob. Comput. 2021 (9933111). (2021).

  • Hussain, S. et al. Feature-engineered NGBoost machine-learning framework for fraud detection. Sensors 218423. (2021).

    Google Scholar

  • Garciarena, U. & Santana, R. Interaction between missing data types, imputation methods, and supervised classifiers. Expert Syst. Appl. 8952–65. (2017).

    Google Scholar

  • Alwateer, M. et al. Missing data imputation: a comprehensive review. J. Comput. Commun. 1253–75. (2024).

    Google Scholar

  • Blank, J. & Deb, K. Pymoo: multi-objective optimization in python. IEEE Access. 889497–89509. (2020).

    Google Scholar

  • National Institute of Diabetes and Digestive and Kidney Diseases. PIMA Indians diabetes database. Kaggle (2016).

  • Related Posts

    Leave a Comment