A Cost-Effective NILM Solution With Three-Point Labelling and Non-Causal Convolution Technique

Abstract

Although deep learning is increasingly promising in the field of Non-Intrusive Load Monitoring (NILM) these days, the high costs of data recording and labelling represent a significant challenge for the training of supervised models. To address this, a cost-effective sequence-to-points NILM solution is proposed, integrating three-point labelling with non-causal convolution techniques. The approach introduces a semi-automatic labelling framework for obtaining NILM three-point data, which provides a low-cost data collection and labelling solution for large-scale applications. Then, a novel loss function combining coordinate loss and confidence loss is developed to address the positional misalignment and negative sample confusion in sequence-to-points scenario in NILM. Furthermore, an advanced neural network architecture based on multi-scale non-causal temporal convolution techniques is designed to capture unique features and operational modes of different appliances. Experimental results on the UK-DALE dataset show that the proposed mixed loss function has an advantage over plain Mean Absolute Error (MAE) on the sequence-to-points occasion, and the novel network outperforms on all of the appliances, demonstrating its potential for practical NILM applications.