Knowledge-based segmentation to improve accuracy and explainability in non-technical losses detection

Utility companies have a great interest in identifying energy losses. Here, we focus on Non-Technical Losses (NTL), which refer to losses caused by utility theft or meter errors. Typically, utility companies resort to machine learning solutions to automate and optimise the identification of such los...

Descripción completa

Detalles Bibliográficos
Autores: Calvo Ibáñez, Albert|||0000-0002-3761-9478, Coma Puig, Bernat|||0000-0003-3944-797X, Carmona Vargas, Josep|||0000-0001-9656-254X, Arias Vicente, Marta|||0000-0001-7359-1815
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/346088
Acceso en línea:https://hdl.handle.net/2117/346088
https://dx.doi.org/10.3390/en13215674
Access Level:acceso abierto
Palabra clave:Machine learning
Energy consumption
Non-technical losses
Supervised learning
Ensemble learning
Explainability
Aprenentatge automàtic
Energia -- Consum
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial::Aprenentatge automàtic
Àrees temàtiques de la UPC::Energies::Gestió de l'energia
Descripción
Sumario:Utility companies have a great interest in identifying energy losses. Here, we focus on Non-Technical Losses (NTL), which refer to losses caused by utility theft or meter errors. Typically, utility companies resort to machine learning solutions to automate and optimise the identification of such losses. This paper extends an existing NTL-detection framework: by including knowledge-based NTL segmentation, we have detected some opportunities for improving the accuracy and the explanations provided to the utility company. Our improved models focus on specific types of NTL and therefore, the explanations provided are easier to interpret, allowing stakeholders to make more informed decisions. The improvements and results presented in the article may benefit other industrial frameworks.