A State Monitoring and Multi-Level Safety Pre-Warning Method for Electric Vehicle Charging Process

被引：0

作者：

Gao D. ^{[1
]}

Zheng X. ^{[1
]}

Wang Y. ^{[1
]}

Yang Q. ^{[2
]}

机构：

[1] School of Automation and Electronic Engineering, Qingdao University of Science & Technology, Qingdao

[2] School of Information Science and Technology, Qingdao University of Science & Technology, Qingdao

来源：

Diangong Jishu Xuebao/Transactions of China Electrotechnical Society | 2022年 / 37卷 / 09期

关键词：

Bi-directional long-short term memory; Charging process; Convolutional neural networks; Electric vehicle; Multi-level safety pre-warning;

D O I：

10.19595/j.cnki.1000-6753.tces.211846

中图分类号：

学科分类号：

摘要：

The frequent occurrence of electric vehicle burning accidents during the charging process has become a key issue restricting the development of electric vehicles. Aiming at the problem of charging safety, this paper proposes a new method of electric vehicle charging status monitoring and multi-level safe pre-warning. The method is based on convolutional neural networks (CNN) and bi-directional long-short memory (BiLSTM), uses the charging history data of electric vehicles to construct a CNN-BiLSTM multi-level safety pre-warning model; designed the charging status monitoring and multi-level safe pre-warning implementation process of the model; compared with other models, verified the prediction accuracy of the model; through the sliding window method, the pre-warning threshold of the model is determined. Experiments have shown that this method can monitor the charging process of electric vehicles in real time, find faults in time and send out pre-warning signals to ensure the safety of electric vehicle charging. © 2022, Electrical Technology Press Co. Ltd. All right reserved.

引用

页码：2252 / 2262

页数：10