Abstract

A reliable and timely thermal runaway alarming method is useful to avoid potential fires happening to electric vehicles (EVs). In this paper, fault diagnosis for the battery pack in EVs using thresholds for multiple safety indicators. Firstly, a deep neural network is used to predict temperature and voltage in the future. Then both voltage-related and temperature-related safety indicators are extracted, including extremum entropy, variance entropy, cumulative center distance, and temperature rise rate. Their corresponding thresholds are defined using a boxplot. Verification using real-world fired EV data shows that the proposed method could realize 10–13 min ahead thermal runaway alarming.

Details

Title
Fault diagnosis for battery pack in electric vehicles using thresholds for multiple safety indicators
Author
Wang, Yi 1 ; Luo, Shaohua 1 ; Yang, Ou 1 ; Bai, Qin 1 ; Zhang, Wendong 1 

 China Automotive Engineering Research Institute Co., Ltd. , Chongqing 401122 , China 
First page
012034
Publication year
2023
Publication date
Aug 2023
Publisher
IOP Publishing
ISSN
17426588
e-ISSN
17426596
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2848452796
Copyright
Published under licence by IOP Publishing Ltd. This work is published under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.