Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM

The prediction of remaining useful life (RUL) of lithium–ion batteries is key to the reliability assessment of batteries and affects safe application of batteries. This article introduces a CEEMDAN-RF-MHA-ED-LSTM method. Using CEEMDAN, the battery capacity data were decomposed to obtain intrinsic mo...

Full description

Saved in:
Bibliographic Details
Main Authors: Bingzeng Song, Guangzhao Yue, Dong Guo, Hanming Wu, Yonghai Sun, Yuhua Li, Bin Zhou
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Batteries
Subjects:
Online Access:https://www.mdpi.com/2313-0105/11/3/86
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850204889970376704
author Bingzeng Song
Guangzhao Yue
Dong Guo
Hanming Wu
Yonghai Sun
Yuhua Li
Bin Zhou
author_facet Bingzeng Song
Guangzhao Yue
Dong Guo
Hanming Wu
Yonghai Sun
Yuhua Li
Bin Zhou
author_sort Bingzeng Song
collection DOAJ
description The prediction of remaining useful life (RUL) of lithium–ion batteries is key to the reliability assessment of batteries and affects safe application of batteries. This article introduces a CEEMDAN-RF-MHA-ED-LSTM method. Using CEEMDAN, the battery capacity data were decomposed to obtain intrinsic mode functions (IMFs), and the weight of each IMF was obtained via the random forest (RF) algorithm. The LSTM neural network was used, the encoder–decoder (ED) structure was introduced, the multi-head attention (MHA) mechanism was used to construct a network model, and the particle swarm optimization (PSO) algorithm was used to optimize the model parameters. Each IMF was input into the model, and the obtained forecast results were weighted and reconstructed to obtain the final forecast data. This method was validated on the battery dataset released by NASA. Compared with that of the single LSTM model, the mean absolute error of the proposed method decreases by 74%, 62%, 71%, and 55% on the No. 05, 06, 07, and 18th battery datasets, respectively. The root mean square error decreased by 72%, 59%, 70%, and 54%, and the mean absolute percent error decreased by 75%, 65%, 71%, and 58%, respectively. This method can accurately predict battery RUL.
format Article
id doaj-art-c5395ab15de14f9e9e052db2cbfa4de9
institution OA Journals
issn 2313-0105
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Batteries
spelling doaj-art-c5395ab15de14f9e9e052db2cbfa4de92025-08-20T02:11:12ZengMDPI AGBatteries2313-01052025-02-011138610.3390/batteries11030086Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTMBingzeng Song0Guangzhao Yue1Dong Guo2Hanming Wu3Yonghai Sun4Yuhua Li5Bin Zhou6School of Transportation and Vehicle Engineering, Shandong University of Technology, Zibo 255000, ChinaSchool of Transportation and Vehicle Engineering, Shandong University of Technology, Zibo 255000, ChinaSchool of Transportation and Vehicle Engineering, Shandong University of Technology, Zibo 255000, ChinaNational Engineering Laboratory for Mobile Source Emission Control Technology, Tianjin 300300, ChinaJinan Heating Group Co., Ltd., Jinan 250011, ChinaCollege of Mechanical and Electronic Engineering, Shandong Agricultural University, Taian 271017, ChinaState Key Lab of Intelligent Transportation System, Beijing 100088, ChinaThe prediction of remaining useful life (RUL) of lithium–ion batteries is key to the reliability assessment of batteries and affects safe application of batteries. This article introduces a CEEMDAN-RF-MHA-ED-LSTM method. Using CEEMDAN, the battery capacity data were decomposed to obtain intrinsic mode functions (IMFs), and the weight of each IMF was obtained via the random forest (RF) algorithm. The LSTM neural network was used, the encoder–decoder (ED) structure was introduced, the multi-head attention (MHA) mechanism was used to construct a network model, and the particle swarm optimization (PSO) algorithm was used to optimize the model parameters. Each IMF was input into the model, and the obtained forecast results were weighted and reconstructed to obtain the final forecast data. This method was validated on the battery dataset released by NASA. Compared with that of the single LSTM model, the mean absolute error of the proposed method decreases by 74%, 62%, 71%, and 55% on the No. 05, 06, 07, and 18th battery datasets, respectively. The root mean square error decreased by 72%, 59%, 70%, and 54%, and the mean absolute percent error decreased by 75%, 65%, 71%, and 58%, respectively. This method can accurately predict battery RUL.https://www.mdpi.com/2313-0105/11/3/86lithium–ion batterylife predictionmodal decompositionparticle swarm optimization algorithmmodal decompositionlong short-term memory network
spellingShingle Bingzeng Song
Guangzhao Yue
Dong Guo
Hanming Wu
Yonghai Sun
Yuhua Li
Bin Zhou
Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
Batteries
lithium–ion battery
life prediction
modal decomposition
particle swarm optimization algorithm
modal decomposition
long short-term memory network
title Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
title_full Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
title_fullStr Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
title_full_unstemmed Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
title_short Prediction of the Remaining Useful Life of Lithium–Ion Batteries Based on Mode Decomposition and ED-LSTM
title_sort prediction of the remaining useful life of lithium ion batteries based on mode decomposition and ed lstm
topic lithium–ion battery
life prediction
modal decomposition
particle swarm optimization algorithm
modal decomposition
long short-term memory network
url https://www.mdpi.com/2313-0105/11/3/86
work_keys_str_mv AT bingzengsong predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT guangzhaoyue predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT dongguo predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT hanmingwu predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT yonghaisun predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT yuhuali predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm
AT binzhou predictionoftheremainingusefullifeoflithiumionbatteriesbasedonmodedecompositionandedlstm