Electrical Fraud Detection Model Based on Machine Learning

• Geam Pfeiff UFPA
• Felipe Araújo UFPA
• Helder Oliveira UFPA http://orcid.org/0000-0003-3143-6280
• Denis Rosário UFPA http://orcid.org/0000-0003-1119-2450
• Eduardo Cerqueira UFPA http://orcid.org/0000-0003-2162-6523

DOI: https://doi.org/10.5753/sbcup.2020.11211

Abstract

Non-technical losses, in most cases caused by fraud, are the leading cause of the financial losses of electricity utilities. These losses dramatically decrease the quality of the electrical networks, increasing the chances of blackouts, short circuits, and equipment failures. Thus, it is strategic to develop models that can detect non-technical losses. This article presents an Electrical Fraud Detector Based on Machine Learning (EFDBML), which classifies users as fraudulent or honest based on electrical consumption patterns and stochastic features. EFDBML obtained a detection rate of 98.02% and a false positive rate of 2.47% after selecting the machine learning algorithm with the best performance.

Keywords: Non-technical losses, Machine Learning, Smart grids, Urban computing

References

ANEEL (2019). Perdas de energia elétrica na distribuiçõa. http://www.encurtador.com.br/imrIS, Acesso: 20/1/2020.

Buzau, M. M., Tejedor-Aguilera, J., Cruz-Romero, P., and Gómez-Expósito, A. (2018). Detection of non-technical losses using smart meter data and supervised learning. IEEE Transactions on Smart Grid, 10(3):2661–2670.

CER (2012). Irish Social Science Data Archive. http://www.ucd.ie/issda/data/commissionforenergyregulationcer/, Acesso: 7/8/2019.

Han, W. and Xiao, Y. (2017). Nfd: Non-technical loss fraud detection in smart grid. Computers & Security, 65:187–201.

Heaton, J. (2016). An empirical analysis of feature engineering for predictive modeling. In SoutheastCon 2016, pages 1–6. IEEE.

Jindal, A., Dua, A., Kaur, K., Singh, M., Kumar, N., and Mishra, S. (2016). Decision tree and svm-based data analytics for theft detection in smart grid. IEEE Transactions on Industrial Informatics, 12(3):1005–1016.

Jokar, P., Arianpoo, N., and Leung, V. C. (2015). Electricity theft detection in ami using customers’ consumption patterns. IEEE Transactions on Smart Grid, 7(1):216–226.

Messinis, G. M. and Hatziargyriou, N. D. (2018a). Review of non-technical loss detection methods. Electric Power Systems Research, 158:250–266.

Messinis, G. M. and Hatziargyriou, N. D. (2018b). Unsupervised classification for non-technical loss detection. In 2018 Power Systems Computation Conference (PSCC),pages 1–7. IEEE.

Messinis, G. M., Rigas, A. E., and Hatziargyriou, N. D. (2019). A hybrid method for non-technical loss detection in smart distribution grids. IEEE Transactions on Smart Grid, 10(6):6080–6091.

Punmiya, R. and Choe, S. (2019). Energy theft detection using gradient boosting theft detector with feature engineering-based preprocessing. IEEE Transactions on Smart Grid, 10(2):2326–2329.

Ramos, C. C., Rodrigues, D., de Souza, A. N., and Papa, J. P. (2018). On the study of commercial losses in brazil: a binary black hole algorithm for theft characterization. IEEE Transactions on Smart Grid, 9(2):676–683.

Zheng, Z., Yang, Y., Niu, X., Dai, H.-N., and Zhou, Y. (2017). Wide and deep convolutional neural networks for electricity-theft detection to secure smart grids. IEEETransactions on Industrial Informatics, 14(4):1606–1615.