AUTHOR=Huang Qifeng , Fang Kaijie , Ding Zecheng , Cheng Hanmiao , Huang Yixuan , Geng Lulu , Wang Puyu , Xu Haibo 

TITLE=A Non-Intrusive Residential Electric Bicycle Load Monitoring Method Based on Hybrid Feature Bitmap and DeiT

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.896398

DOI=10.3389/fenrg.2022.896398

ISSN=2296-598X

ABSTRACT=Taking into account the energy management and fire safety, Electric bicycle is one of most significant household loads required real-time sensing for non-intrusive load monitoring. V-I trajectories, power quantities, harmonic characteristics are the basic selection in feature space for appliances identification. Based on the study of the charging mode of electric bicycle, this paper expands the V-I trajectory into V-△I trajectory for gaining the load signature with multi appliances working the simultaneously. We perform linearly interpolation and pixelation to obtain a bitmap of V-△I trajectory. Meanwhile, active and harmonic features are encoded and combined to form a hybrid feature bitmap, which is unique to compensate for the high harmonic feature loss caused by the pixelation of the V-△I trajectory. Further, we train the DeiT model on the self-built datasets, and performed two experiments under single and superposition working conditions for electric bicycles. Our case results indicate that the Diet model using hybrid feature bitmap offers better overall precision of prediction of electric bicycle, against other deep convolutional neural networks.