Lightweight and Scalable Transfer Learning Framework for Load Disaggregation

TL;DR

This paper introduces RefQuery, a scalable transfer learning framework for load disaggregation that adapts to new homes efficiently by learning appliance-specific embeddings, enabling real-time operation on resource-limited devices.

Contribution

RefQuery is a novel NILM framework that uses compact appliance fingerprints and a frozen pretrained network to adapt to new environments with minimal training.

Findings

Achieves high accuracy with low computational cost.

Outperforms existing methods on public datasets.

Supports real-time NILM on edge devices.

Abstract

Non-Intrusive Load Monitoring (NILM) aims to estimate appliance-level consumption from aggregate electrical signals recorded at a single measurement point. In recent years, the field has increasingly adopted deep learning approaches; however, cross-domain generalization remains a persistent challenge due to variations in appliance characteristics, usage patterns, and background loads across homes. Transfer learning provides a practical paradigm to adapt models with limited target data. However, existing methods often assume a fixed appliance set, lack flexibility for evolving real-world deployments, remain unsuitable for edge devices, or scale poorly for real-time operation. This paper proposes RefQuery, a scalable multi-appliance, multi-task NILM framework that conditions disaggregation on compact appliance fingerprints, allowing one shared model to serve many appliances without a fixed output set. RefQuery keeps a pretrained disaggregation network fully frozen and adapts to a target home by learning only a per-appliance embedding during a lightweight backpropagation stage. Experiments on three public datasets demonstrate that RefQuery delivers a strong accuracy-efficiency trade-off against single-appliance and multi-appliance baselines, including modern Transformer-based methods. These results support RefQuery as a practical path toward scalable, real-time NILM on resource-constrained edge devices.