Title: New energy vehicle lithium battery life prediction method based on improved deep learning

Authors: Zhiwen An

Addresses: College of Mechanical and Electrical Engineering, Guangdong University of Science and Technology, Dongguan, 523083, China

Abstract: The traditional methods of life prediction of lithium battery in new energy vehicles have the problems of large error and low efficiency. The paper puts forward a new energy vehicle lithium battery life prediction method. The capacity, internal resistance, terminal voltage and charge discharge cycle parameters of lithium battery for new energy vehicles are extracted to determine the key parameters affecting the life of lithium battery. The gradient descent method is used to improve the deep learning algorithm, and the improved deep learning prediction model is constructed. The key parameters affecting the lithium battery life are taken as the input of the model, and the optimal value is found to predict the lithium battery life of new energy vehicles. The results show that the capacity estimated by the proposed method is basically consistent with the actual capacity, and the life prediction time is always less than 2.2 s.

Keywords: improving deep learning; lithium batteries; voltage curve; key parameters; life prediction.

DOI: 10.1504/IJVD.2022.128013

International Journal of Vehicle Design, 2022 Vol.89 No.1/2, pp.69 - 83

Received: 02 Feb 2021
Accepted: 06 Jul 2021
Published online: 04 Jan 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article