Nonintrusive Load Identification Considering Unknown Load Based on Bimodal Fusion and One-Class Classification

Nonintrusive load monitoring (NILM) enables the real-time monitoring and data analysis of energy consumption, providing more accurate, efficient, and intelligent data support for energy management. However, traditional NILM models can only identify known load (KL) presented in the training dataset; they cannot perform unknown load (UKL) detection, and they have problems with feature selection, including information loss and limitations in capturing load signal information, which make them less suitable for the user side. In this article, a new method based on bimodal fusion and one-class (OC) classification is proposed for NILM. The bimodal features include 2-D image features and multidimensional sequence features. To enrich the feature representation, the bimodal features are fused by long short-term memory (LSTM) network. To address UKL detection, a simple and lightweight OC model based on deep support vector data description (DSVDD) is innovatively proposed. The proposed model is also validated on multiple publicly available datasets, demonstrating its significant advantages and applicability both on KL identification and UKL detection.