Thesis supervisor: László Szathmáry
Location of studies (in Hungarian): University of Debrecen Faculty of Informatics Abbreviation of location of studies: DE IK
Description of the research topic:
Syllabus
Power supply companies are considered the backbone for any country. These companies face huge setbacks due to Non-Technical Loss (NTL) in distribution networks. For example, India loses 4.5 billion USD every year on account of NTL. This loss can range upto 50% of the total electricity produced in developing countries. The developed countries, including USA and UK, also suffer a loss of 1-6 billion annually. Any power supply company wants to minimize NTL by first detecting it and then addressing it properly.
Over the past decade, the research community has paid attention to detecting NTL with the collaboration of electric suppliers using machine learning classifiers. This includes using Support Vector Machine (SVM), Optimum Path Forest (OPF), random forest, multi-layer perceptron neural network (NN), K-Nearest Neighbors (KNN), Adaboost, naive bayes, decision trees and deep learning.
The goal of the thesis is to investigate, implement and compare various machine learning methods for NTL detection. Comparisons can be done using different performance evaluation metrics.
Bibliography
• Di Martino, M., Decia, F., Molinelli, J., Fernández, A.: Improving Electric Fraud Detection using Class Imbalance Strategies. In: ICPRAM (2). (2012) 135–141
• Glauner, P.O., Boechat, A., Dolberg, L., Meira, J.A., State, R., Bettinger, F., Rangoni, Y., Duarte, D.: The Challenge of Non-Technical Loss Detection using Artificial Intelligence: A Survey. CoRR abs/1606.00626 (2016)
• Coma-Puig, B., Carmona, J., Gavalda, R., Alcoverro, S., Martin, V.: Fraud detection in energy consumption: a supervised approach. In: Data Science and Advanced Analytics (DSAA), 2016 IEEE International Conference on, IEEE (2016) 120–129
• Bishop, C. M., Pattern Recognition and Machine Learning, Springer, 2006.
Deadline for application: 2021-11-15
2024. IV. 17. ODT ülés Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).
Login
