期刊
2019 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (ICMA)
卷 -, 期 -, 页码 1067-1072出版社
IEEE
DOI: 10.1109/icma.2019.8816512
关键词
Battery face imaging; voltage classification; convolutional neural network; linear discriminant analysis; wavelet transform
资金
- China Scholarship Council
By using a three-dimensional (3D) radio-frequency based sensor, which is called Walabot, and machine learning (ML) algorithm, this paper presents a contactless way to generate lithium-ion battery face images for battery voltage classification. First, Walabot was applied to sampling images, which can reflect inside physic structure of lithium-ion batteries (LIBs). Second, these images were preprocessed by data enhancement or wavelet transform. Finally, these preprocessed images were set as inputs of a convolutional neural network (CNN). After images network training, the CNN can be applied to validating test images in different voltage values. Experiment results of five LIBs illustrate that the proposed contactless battery face imaging method provides a totally new way to conduct voltage classification for LIBs.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据