Adaptive Integration of Steady-State Change Features for Enhanced Multi-Label Energy Disaggregation

The primary objective of this project is to develop an efficient and accurate energy disaggregation framework titled “Adaptive Integration of Steady-State Change Features for Enhanced Multi-Label Energy Disaggregation.” The proposed system utilizes Adaptive Feature-Temporal NILM (AFT-NILM), Lightweight Hybrid Interpretable NILM (LHI-NILM), and Hybrid Ensemble NILM (HE-NILM) to identify and classify multiple appliance operating states from aggregate energy consumption data. The framework focuses on adaptively integrating steady-state change features to capture appliance behavior and improve multi-label disaggregation performance. AFT-NILM is employed to learn feature-temporal relationships, LHI-NILM provides efficient and interpretable appliance-level predictions, and HE-NILM combines prediction outcomes to enhance overall accuracy and robustness. The project aims to improve appliance identification, increase disaggregation accuracy, reduce classification errors, and support effective energy monitoring and management systems.