Convolutional Sequence to Sequence Non-intrusive Load Monitoring
This is an incremental improvement for energy management systems, enhancing appliance-level electricity disaggregation.
The paper tackled non-intrusive load monitoring by proposing a convolutional sequence-to-sequence model using gated linear units and residual blocks, achieving satisfactory disaggregation performance for varied appliances on the REDD dataset.
A convolutional sequence to sequence non-intrusive load monitoring model is proposed in this paper. 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. We apply the proposed model to the REDD 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.