Open AccessConvolutional sequence to sequence non‐intrusive load monitoring
Author(s) -
Chen Kunjin,
Wang Qin,
He Ziyu,
Chen Kunlong,
Hu Jun,
He Jinliang
Publication year - 2018
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2018.8352
Subject(s) - sequence (biology) , convolutional neural network , residual , computer science , set (abstract data type) , aggregate (composite) , algorithm , time sequence , pattern recognition (psychology) , artificial intelligence , genetics , materials science , composite material , biology , programming language
A convolutional sequence to sequence non‐intrusive load monitoring model is proposed in this study. Gated linear unit convolutional layers are used to extract information from the sequences of aggregate electricity consumption. Residual blocks are also introduced to refine the output of the neural network. The partially overlapped output sequences of the network are averaged to produce the final output of the model. The authors apply the proposed model to the reference energy disaggregation data set dataset and compare it with the convolutional sequence to point model in the literature. Results show that the proposed model is able to give satisfactory disaggregation performance for appliances with varied characteristics.