DualNILM: Energy Injection Identification Enabled Disaggregation with Deep Multi-Task Learning

2025-08-20

AI Summary

Traditional Non-Intrusive Load Monitoring (NILM) methods rely solely on aggregate electricity meter data; however, their performance degrades significantly in residential settings with widespread integration of distributed energy resources (DERs)—such as photovoltaics and battery storage—whose injected power obscures appliance-level signatures. To address this, we propose DualNILM, the first framework that jointly models appliance state detection and DER injection detection as a cooperative multi-task learning problem. Methodologically, DualNILM introduces a Transformer-based dual-path architecture that simultaneously performs sequence-to-point (state classification) and sequence-to-sequence (power disaggregation) modeling, explicitly capturing multi-scale temporal dependencies. Evaluated on both a newly constructed real-world dataset and synthetic benchmarks, DualNILM consistently outperforms state-of-the-art NILM approaches across all key metrics for both tasks. This advancement enhances the robustness and interpretability of load monitoring in source-integrated distribution systems.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: State Estimation
• Computer Vision: Multi-modal Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Non-Intrusive Load Monitoring (NILM) offers a cost-effective method to obtain fine-grained appliance-level energy consumption in smart homes and building applications. However, the increasing adoption of behind-the-meter energy sources, such as solar panels and battery storage, poses new challenges for conventional NILM methods that rely solely on at-the-meter data. The injected energy from the behind-the-meter sources can obscure the power signatures of individual appliances, leading to a significant decline in NILM performance. To address this challenge, we present DualNILM, a deep multi-task learning framework designed for the dual tasks of appliance state recognition and injected energy identification in NILM. By integrating sequence-to-point and sequence-to-sequence strategies within a Transformer-based architecture, DualNILM can effectively capture multi-scale temporal dependencies in the aggregate power consumption patterns, allowing for accurate appliance state recognition and energy injection identification. We conduct validation of DualNILM using both self-collected and synthesized open NILM datasets that include both appliance-level energy consumption and energy injection. Extensive experimental results demonstrate that DualNILM maintains an excellent performance for the dual tasks in NILM, much outperforming conventional methods.

Problem

• Identifies appliance states and injected energy in NILM
• Addresses performance decline from behind-the-meter energy sources
• Solves multi-task learning for energy disaggregation challenges

Innovation

• Deep multi-task learning framework
• Transformer-based architecture integration
• Sequence-to-point and sequence-to-sequence strategies