Deep Learning-Based Energy Disaggregation and On/Off Detection of Household Appliances

Energy disaggregation, a.k.a. Non-Intrusive Load Monitoring, aims to separate the energy consumption of individual appliances from readings a mains power meter measuring total of, e.g., whole house. can be useful in many applications, providing appliance-level feedback end users help them understand their and ultimately save energy. Recently, with availability large-scale datasets, various neural network models such as convolutional networks recurrent have been investigated solve disaggregation problem. Neural learn complex patterns large amounts data shown outperform traditional machine learning methods variants hidden Markov models. However, current for are either computational expensive or not capable handling long-term dependencies. In this article, we investigate application recently developed WaveNet task disaggregation. Based on real-world dataset collected 20 households over 2 years, show that outperforms state-of-the-art deep proposed literature terms both error measures cost. On basis then performance two deep-learning based frameworks on/off detection which at estimating whether an appliance is operation not. The first framework obtains states by binarising predictions regression model trained while second directly training binary classifier binarised serving target values. same dataset, framework, i.e., classifier, achieves better F1 score.