What is it about?
The article is devoted to the development of a machine learning model for detecting non-technical losses of electricity (NTL). The proposed model uses an ensemble method (stacking), combining the Random Forest, LightGBM, and artificial neural networks algorithms. The high accuracy of the model (ROC-AUC = 0.88) has been experimentally confirmed, which allows for the effective prediction of the probability of NTL for each consumer based on daily consumption readings. This approach can serve as a basis for decision support systems aimed at improving the effectiveness of combating NTL and the sustainable development of the energy industry.
Featured Image
Photo by Vadim Bogulov on Unsplash
Why is it important?
The article is unique because it integrates multiple advanced machine learning algorithms into an ensemble model to accurately detect non-technical losses (NTL) in electricity consumption, a critical issue for energy providers. Its innovative stacking approach enhances prediction performance compared to individual models, addressing a significant challenge in energy management. The relevance of this research lies in its potential to reduce financial losses and improve the efficiency of power distribution systems. Promoting this article through Research Engagement Activities (REA) such as webinars, social media campaigns, and collaborations with industry experts can significantly broaden its readership and impact by reaching both academic and professional audiences interested in smart energy solutions.
Perspectives
Writing this article gave me great pleasure because I have co-authors with whom I have collaborated for a long time and whom I deeply respect. Thanks to this article, energy companies have approached us, expressing interest in our development. Currently, the proposed ensemble model is undergoing pilot testing and validation.
Irbek Morgoev
North-Caucasian Mining and Metallurgical Institute
Read the Original
This page is a summary of: Methodology for Detecting Non-Technical Energy Losses Using an Ensemble of Machine Learning Algorithms, Computer Modeling in Engineering & Sciences, January 2025, Tsinghua University Press,
DOI: 10.32604/cmes.2025.064502.
You can read the full text:
Contributors
The following have contributed to this page
