Abstract
The application of Artificial Intelligence techniques in industry equips companies with new essential tools to improve their principal processes. This is especially true for energy companies, as they have the opportunity, thanks to the modernization of their installations, to exploit a large amount of data with smart algorithms. In this work we explore the possibilities that exist in the implementation of Machine-Learning techniques for the detection of Non-Technical Losses in customers. The analysis is based on the work done in collaboration with an international energy distribution company. We report on how the success in detecting Non-Technical Losses can help the company to better control the energy provided to their customers, avoiding a misuse and hence improving the sustainability of the service that the company provides.
Highlights
Companies distributing energy are confronted with the challenge of supplying energy to their customers in the most efficient way
Non-Technical Losses (NTL) practices represent a prejudice with respect to the rest of consumers, since the energy losses influence the computation of reference tariffs, which in reality implies that non-fraudulent customers are paying for fraudulent customers
We provide the benefits for the company in terms of knowledge, modernization and especially the improvement of detection of NTL cases increasing the amount of energy recovered by the company
Summary
Companies distributing energy are confronted with the challenge of supplying energy to their customers in the most efficient way This requires reducing as much as possible the energy loss that arises from the distribution network malfunctions, the irregularities in the supply points, or the fraudulent behavior of their customers. We explain a global perspective of our project, to provide the reader a vision of the possibilities of implementing machine-learning techniques in the detection of NTL. We consider that this work is useful for those that want to implement machine-learning techniques in industrial processes, especially for those that want to build a system to detect NTL in utility companies, since the challenges and information used are similar.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have