Article: Prediction method of battery remaining life based on CEEMDAN-RF-SED-LSTM neural network Journal: International Journal of Information and Communication Technology (IJICT) 2024 Vol.25 No.9 pp.1 - 21 Abstract: In order to ensure the reliability and safety of electric vehicles, a CEEMDAN-RF-SED-LSTM method for lithium ion power battery system is proposed. Taking the time interval of equal voltage charging as the indirect health factor, taking into account the influence of external interference and capacity regeneration phenomenon, the degradation trend of battery is obtained by variational mode decomposition (VMD). The improved recurrent neural network model - short and long time series (LSTM) is used to obtain the residual life prediction. Finally, the established model is compared with FNN, CNN, LSTM and other neural networks, which gives full play to the characteristics of SCN such as strong autonomy, fast convergence speed and low network cost. The performance of this method is tested with NASA dataset as the research object. The experimental results show that CEEMDAN-RF-SED-LSTM model performs well in predicting RUL of batteries, and the prediction results have lower errors than that of a single model. Inderscience Publishers - linking academia, business and industry through research

Title: Prediction method of battery remaining life based on CEEMDAN-RF-SED-LSTM neural network

Authors: Hanping Pan; Wenjian Feng; Jiaqi Ge; Jijun Zhang

Addresses: College of Automotive and Information Engineering, Guangxi Eco-Engineering Vocational &Technical College, Liuzhou 545004, Guangxi, China ' College of Automotive and Information Engineering, Guangxi Eco-Engineering Vocational &Technical College, Liuzhou 545004, Guangxi, China ' College of Automotive and Information Engineering, Guangxi Eco-Engineering Vocational &Technical College, Liuzhou 545004, Guangxi, China ' College of Automotive and Information Engineering, Guangxi Eco-Engineering Vocational &Technical College, Liuzhou 545004, Guangxi, China

Abstract: In order to ensure the reliability and safety of electric vehicles, a CEEMDAN-RF-SED-LSTM method for lithium ion power battery system is proposed. Taking the time interval of equal voltage charging as the indirect health factor, taking into account the influence of external interference and capacity regeneration phenomenon, the degradation trend of battery is obtained by variational mode decomposition (VMD). The improved recurrent neural network model - short and long time series (LSTM) is used to obtain the residual life prediction. Finally, the established model is compared with FNN, CNN, LSTM and other neural networks, which gives full play to the characteristics of SCN such as strong autonomy, fast convergence speed and low network cost. The performance of this method is tested with NASA dataset as the research object. The experimental results show that CEEMDAN-RF-SED-LSTM model performs well in predicting RUL of batteries, and the prediction results have lower errors than that of a single model.

Keywords: lithium ion battery; life prediction; configure the network randomly; incremental learning.

DOI: 10.1504/IJICT.2024.142816

International Journal of Information and Communication Technology, 2024 Vol.25 No.9, pp.1 - 21

Received: 11 Dec 2023
Accepted: 11 May 2024
Published online: 22 Nov 2024 *

Title: Prediction method of battery remaining life based on CEEMDAN-RF-SED-LSTM neural network

Keep up-to-date