SAMNet: Toward Latency-Free Non-Intrusive Load Monitoring via Multi-Task Deep Learning

• Published in: IEEE transactions on smart grid 2022-05, Vol.13 (3), p.2412-2424
• Main Authors: Liu, Yinyan, Qiu, Jing, Ma, Jin
• Format: Article
• Language: English
• Subjects: Aggregates; Convolution; Electrical loads; Energy consumption; Energy Disaggregation; Energy management; Fault diagnosis; Load monitoring; Monitoring; Multi-Task learning; Multitasking; Neural networks; Non-Intrusive Load Monitoring; Self-Attention Mechanism; Smart grid; Task analysis; Training
• ISSN: 1949-3053; 1949-3061
• DOI: 10.1109/TSG.2021.3139395

Non-intrusive load monitoring (NILM), including state detection and energy disaggregation, aims to identify the on/off state and energy consumption from the aggregate load of a building. By monitoring the electrical behavior of consumers, smart grid applications such as demand response and recommendation services can also be realized for saving energy bills, environmental effectiveness, assisted living, and fault diagnosis. To achieve latency-free monitoring, this paper presents a S cale- and A ttention-experts based M ulti-task neural network (SAMNet) with a large enough context of the view to make full use of the correlation between the tasks of the NILM. A shared expert learner is designed to learn a good summary of the features. Self-attention mechanism is creatively adopted to realize the weighted fusion of different experts. To address the problem of manually setting different threshold values for different appliances to decide the on/off states, we designed the network automatically labeling the on/off states. Extensive experimental results with the public datasets demonstrate the effectiveness and superiority of our proposed model.