Inferring Appliance Energy Usage from Smart Meters using Fully Convolutional Encoder Decoder Networks

Energy management presents one of the principal sustainability challenges within urban centers given that they account for 75% of the energy consumption worldwide. In the context of a smart city framework, the use of intelligent urban systems provides a key opportunity in addressing the energy sustainability issue as an informatics problem where the goal is to deliver energy usage feedback to the users as a means of enabling behavioral change towards energy sustainability. In this paper we present a method to provide appliance energy usage feedback from smart meters using energy disaggregation. We put energy disaggregation in the context of a source separation and signal reconstruction problem in which we train a fully convolutional encoder decoder network to separate appliance energy usage from aggregate whole house electricity consumption data. The results show that the proposed fully convolutional encoder decoder model can achieve competitive accuracy compared with several state-of-the-art methods.

[1]  Jack Kelly,et al.  Neural NILM: Deep Neural Networks Applied to Energy Disaggregation , 2015, BuildSys@SenSys.

[2]  Andrew Y. Ng,et al.  Energy Disaggregation via Discriminative Sparse Coding , 2010, NIPS.

[3]  Abhay Gupta,et al.  Is disaggregation the holy grail of energy efficiency? The case of electricity , 2013 .

[4]  Roberto Cipolla,et al.  SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Jürgen Schmidhuber,et al.  Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction , 2011, ICANN.

[6]  John Salvatier,et al.  Theano: A Python framework for fast computation of mathematical expressions , 2016, ArXiv.

[7]  Marc'Aurelio Ranzato,et al.  Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[8]  R. N. Elliott,et al.  American Council for an Energy-Efficient Economy , 2002 .

[9]  Shwetak N. Patel,et al.  ElectriSense: single-point sensing using EMI for electrical event detection and classification in the home , 2010, UbiComp.

[10]  G. W. Hart,et al.  Nonintrusive appliance load monitoring , 1992, Proc. IEEE.

[11]  Yu-Bin Yang,et al.  Image Restoration Using Very Deep Convolutional Encoder-Decoder Networks with Symmetric Skip Connections , 2016, NIPS.

[12]  J. Zico Kolter,et al.  REDD : A Public Data Set for Energy Disaggregation Research , 2011 .

[13]  Jack Kelly,et al.  Does disaggregated electricity feedback reduce domestic electricity consumption? A systematic review of the literature , 2016, ArXiv.