PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 3, 2026AI0 citationsOpen Access

Building-Level Energy Disaggregation Using AI-Based NILM Techniques in Heterogeneous Environments

View Full Paper
ARAna Rubio-BustosGCGloria Calleja-RodríguezJDJorge De-La-Torre-García

Key Points

  • This research aims to develop and evaluate AI methods for energy disaggregation in different building environments.
  • Developed multiple AI models, including ViT, VAE, and RF.
  • Implemented across residential, commercial, and industrial buildings.
  • Validated using three benchmark datasets: AMPds, COmBED, and IMDELD.
  • ViT-small model achieved R2 values exceeding 0.94 for industrial loads.
  • Random Forest showed R2 up to 0.97 for commercial HVAC systems.
  • Prophet-inspired model effectively captured seasonal patterns in residential appliances.

Abstract

Non-Intrusive Load Monitoring (NILM) represents a powerful approach for energy disaggregation, which enables detailed insights into energy consumption patterns without requiring extensive sensor deployment. While significant advances have been achieved in residential NILM applications, commercial and industrial buildings remain largely underexplored despite their substantial contribution to global energy consumption. This study addresses this gap by developing and evaluating multiple artificial intelligence approaches for energy disaggregation across residential, commercial, and industrial buildings under a unified experimental protocol. We implement and compare several AI-based models, including Vision Transformer (ViT), Variational Autoencoder (VAE), Random Forest (RF), and custom architectures inspired by TimeGPT and Prophet, alongside traditional baseline methods. The proposed framework is validated using three benchmark datasets representing residential (AMPds), commercial (COmBED), and industrial (IMDELD) environments. Experimental results demonstrate that architecture–load interactions, rather than model complexity alone, are the primary determinants of disaggregation accuracy: the ViT-small configuration achieves superior performance for complex industrial loads with R2 values exceeding 0.94, Random Forest proves most effective for finite-state commercial HVAC systems with R2 up to 0.97, and the Prophet-inspired model excels in capturing seasonal patterns in residential appliances. These findings provide evidence-based guidelines for selecting appropriate AI models based on load characteristics, signal-to-noise ratio, and building type, contributing to the practical deployment of NILM in heterogeneous building environments.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Rubio-Bustos et al. (2026) studied this question.

synapsesocial.com/papers/69cf5d345a333a821460acdbhttps://doi.org/10.3390/ai7040122
Ask AI
Helpful
Bookmark
Share
View Full Paper