eduzhai > Applied Sciences > Engineering >

PowerGAN Synthesizing Appliance Power Signatures Using Generative Adversarial Networks

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Non-intrusive load monitoring (NILM) allows users and energy providers togain insight into home appliance electricity consumption using only thebuilding s smart meter. Most current techniques for NILM are trained usingsignificant amounts of labeled appliances power data. The collection of suchdata is challenging, making data a major bottleneck in creating wellgeneralizing NILM solutions. To help mitigate the data limitations, we presentthe first truly synthetic appliance power signature generator. Our solution,PowerGAN, is based on conditional, progressively growing, 1-D Wassersteingenerative adversarial network (GAN). Using PowerGAN, we are able to synthesisetruly random and realistic appliance power data signatures. We evaluate thesamples generated by PowerGAN in a qualitative way as well as numerically byusing traditional GAN evaluation methods such as the Inception score.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×