Lifetime Prediction of Lithium-Ion Battery Using Machine Learning For E-Vehicles

A life cycle of battery with long testing time and without contact measurement devices will be applicable for industrial applications. To this problem, the solution for potential will be provided by the combined technique of supervised learning and infrared thermography. The focus of the research wi...

Full description

Saved in:
Bibliographic Details
Published in:Journal of physics. Conference series 2021-05, Vol.1916 (1), p.12200
Main Authors: Venugopal, D, Pranav, V Sharan, Lakshmi, B Swathi, Vishnu, K
Format: Article
Language:English
Subjects:
Artificial neural networks
Battery cycles
Charging
Discharge
Error analysis
Industrial applications
Infrared imagery
Infrared imaging
Learning theory
Life cycle assessment
Lithium
Lithium-ion batteries
Machine learning
Measuring instruments
Physics
polymer batteries
Rechargeable batteries
Support vector machines
Temperature profiles
Testing time
Thermography
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A life cycle of battery with long testing time and without contact measurement devices will be applicable for industrial applications. To this problem, the solution for potential will be provided by the combined technique of supervised learning and infrared thermography. The focus of the research will be on machine learning techniques. Artificial neural networks (ANNs) and support vector machines (SVMs) are used in conjunction with thermography to estimate the life cycle of lithium ion polymer batteries. The capturing of infrared images at 1 frame per minute and charging of 70 minutes followed by the discharging of 60 minutes for 410 cycles. For ANN and SVM models the input nodes of charging or discharging use the surface temperature profiles. The input will be in thermal profile for the result. Under 10 minutes of testing time, we can estimate the current life cycle of a studied cell with an error of less than 10%. In SVM the accuracy will be similar while comparing with ANN but the testing time will be longer.
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1916/1/012200