Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm

M. S.Hossain Lipu, M. A. Hannan, Aini Hussain, M. H.M. Saad, A. Ayob, K. M. Muttaqi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents an enhanced machine learning based state of charge (SOC) estimation method for a lithium-ion battery using a deep recurrent neural network (DRNN) algorithm. DRNN is suitable for SOC evaluation due to strong computation intelligence and self-learning capabilities. Nevertheless, the performance of DRNN is constrained due to the training accuracy and duration which entirely depends on the appropriate selection of hyper-parameters including hidden layer and hidden neurons. Therefore, firefly algorithm (FA) is employed to find the optimal number for hyper-parameters of DRNN networks. The optimized DRNN based FA algorithm for SOC estimation does not require extensive knowledge about battery chemistry, electrochemical battery model and added filter, rather only needs battery test bench to measure current and voltage. The developed model is tested using two different types of lithium-ion batteries namely lithium nickel manganese cobalt oxide (LiNiMnCoO2) and lithium nickel cobalt aluminum oxide (LiNiCoAlO2). The proposed model is validated by two experimental tests; one with static discharge test and other with pulse discharge test at room temperature. The experimental results indicate the superiority of the DRNN based FA method in comparison with the back-propagation neural network (BPNN) and radial basis function neural network (RBFNN).

Original languageEnglish
Title of host publication2019 IEEE Industry Applications Society Annual Meeting, IAS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538645390
DOIs
Publication statusPublished - Sep 2019
Event2019 IEEE Industry Applications Society Annual Meeting, IAS 2019 - Baltimore, United States
Duration: 29 Sep 20193 Oct 2019

Publication series

Name2019 IEEE Industry Applications Society Annual Meeting, IAS 2019

Conference

Conference2019 IEEE Industry Applications Society Annual Meeting, IAS 2019
CountryUnited States
CityBaltimore
Period29/9/193/10/19

Fingerprint

Recurrent neural networks
neural network
Nickel
Cobalt
Lithium
Neural networks
Oxides
Aluminum Oxide
Manganese
Backpropagation
test battery
Neurons
Learning systems
Lithium-ion batteries
learning
Aluminum
intelligence
chemistry
Electric potential
evaluation

Keywords

  • Deep recurrent neural network
  • Levenberg-Marquardt algorithm Firefly algorithm
  • Lithium-ion battery
  • State of charge

ASJC Scopus subject areas

  • Filtration and Separation
  • Fluid Flow and Transfer Processes
  • Process Chemistry and Technology
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering
  • Industrial and Manufacturing Engineering
  • Transportation

Cite this

Lipu, M. S. H., Hannan, M. A., Hussain, A., Saad, M. H. M., Ayob, A., & Muttaqi, K. M. (2019). Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm. In 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019 [8912322] (2019 IEEE Industry Applications Society Annual Meeting, IAS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IAS.2019.8912322

Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm. / Lipu, M. S.Hossain; Hannan, M. A.; Hussain, Aini; Saad, M. H.M.; Ayob, A.; Muttaqi, K. M.

2019 IEEE Industry Applications Society Annual Meeting, IAS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 8912322 (2019 IEEE Industry Applications Society Annual Meeting, IAS 2019).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lipu, MSH, Hannan, MA, Hussain, A, Saad, MHM, Ayob, A & Muttaqi, KM 2019, Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm. in 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019., 8912322, 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019, Institute of Electrical and Electronics Engineers Inc., 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019, Baltimore, United States, 29/9/19. https://doi.org/10.1109/IAS.2019.8912322
Lipu MSH, Hannan MA, Hussain A, Saad MHM, Ayob A, Muttaqi KM. Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm. In 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 8912322. (2019 IEEE Industry Applications Society Annual Meeting, IAS 2019). https://doi.org/10.1109/IAS.2019.8912322
Lipu, M. S.Hossain ; Hannan, M. A. ; Hussain, Aini ; Saad, M. H.M. ; Ayob, A. ; Muttaqi, K. M. / Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm. 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. (2019 IEEE Industry Applications Society Annual Meeting, IAS 2019).
@inproceedings{22a695018453449abe1354a416a5d81f,
title = "Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm",
abstract = "This paper presents an enhanced machine learning based state of charge (SOC) estimation method for a lithium-ion battery using a deep recurrent neural network (DRNN) algorithm. DRNN is suitable for SOC evaluation due to strong computation intelligence and self-learning capabilities. Nevertheless, the performance of DRNN is constrained due to the training accuracy and duration which entirely depends on the appropriate selection of hyper-parameters including hidden layer and hidden neurons. Therefore, firefly algorithm (FA) is employed to find the optimal number for hyper-parameters of DRNN networks. The optimized DRNN based FA algorithm for SOC estimation does not require extensive knowledge about battery chemistry, electrochemical battery model and added filter, rather only needs battery test bench to measure current and voltage. The developed model is tested using two different types of lithium-ion batteries namely lithium nickel manganese cobalt oxide (LiNiMnCoO2) and lithium nickel cobalt aluminum oxide (LiNiCoAlO2). The proposed model is validated by two experimental tests; one with static discharge test and other with pulse discharge test at room temperature. The experimental results indicate the superiority of the DRNN based FA method in comparison with the back-propagation neural network (BPNN) and radial basis function neural network (RBFNN).",
keywords = "Deep recurrent neural network, Levenberg-Marquardt algorithm Firefly algorithm, Lithium-ion battery, State of charge",
author = "Lipu, {M. S.Hossain} and Hannan, {M. A.} and Aini Hussain and Saad, {M. H.M.} and A. Ayob and Muttaqi, {K. M.}",
year = "2019",
month = "9",
doi = "10.1109/IAS.2019.8912322",
language = "English",
series = "2019 IEEE Industry Applications Society Annual Meeting, IAS 2019",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2019 IEEE Industry Applications Society Annual Meeting, IAS 2019",

}

TY - GEN

T1 - Lithium-ion Battery State of Charge Estimation Method Using Optimized Deep Recurrent Neural Network Algorithm

AU - Lipu, M. S.Hossain

AU - Hannan, M. A.

AU - Hussain, Aini

AU - Saad, M. H.M.

AU - Ayob, A.

AU - Muttaqi, K. M.

PY - 2019/9

Y1 - 2019/9

N2 - This paper presents an enhanced machine learning based state of charge (SOC) estimation method for a lithium-ion battery using a deep recurrent neural network (DRNN) algorithm. DRNN is suitable for SOC evaluation due to strong computation intelligence and self-learning capabilities. Nevertheless, the performance of DRNN is constrained due to the training accuracy and duration which entirely depends on the appropriate selection of hyper-parameters including hidden layer and hidden neurons. Therefore, firefly algorithm (FA) is employed to find the optimal number for hyper-parameters of DRNN networks. The optimized DRNN based FA algorithm for SOC estimation does not require extensive knowledge about battery chemistry, electrochemical battery model and added filter, rather only needs battery test bench to measure current and voltage. The developed model is tested using two different types of lithium-ion batteries namely lithium nickel manganese cobalt oxide (LiNiMnCoO2) and lithium nickel cobalt aluminum oxide (LiNiCoAlO2). The proposed model is validated by two experimental tests; one with static discharge test and other with pulse discharge test at room temperature. The experimental results indicate the superiority of the DRNN based FA method in comparison with the back-propagation neural network (BPNN) and radial basis function neural network (RBFNN).

AB - This paper presents an enhanced machine learning based state of charge (SOC) estimation method for a lithium-ion battery using a deep recurrent neural network (DRNN) algorithm. DRNN is suitable for SOC evaluation due to strong computation intelligence and self-learning capabilities. Nevertheless, the performance of DRNN is constrained due to the training accuracy and duration which entirely depends on the appropriate selection of hyper-parameters including hidden layer and hidden neurons. Therefore, firefly algorithm (FA) is employed to find the optimal number for hyper-parameters of DRNN networks. The optimized DRNN based FA algorithm for SOC estimation does not require extensive knowledge about battery chemistry, electrochemical battery model and added filter, rather only needs battery test bench to measure current and voltage. The developed model is tested using two different types of lithium-ion batteries namely lithium nickel manganese cobalt oxide (LiNiMnCoO2) and lithium nickel cobalt aluminum oxide (LiNiCoAlO2). The proposed model is validated by two experimental tests; one with static discharge test and other with pulse discharge test at room temperature. The experimental results indicate the superiority of the DRNN based FA method in comparison with the back-propagation neural network (BPNN) and radial basis function neural network (RBFNN).

KW - Deep recurrent neural network

KW - Levenberg-Marquardt algorithm Firefly algorithm

KW - Lithium-ion battery

KW - State of charge

UR - http://www.scopus.com/inward/record.url?scp=85076785379&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076785379&partnerID=8YFLogxK

U2 - 10.1109/IAS.2019.8912322

DO - 10.1109/IAS.2019.8912322

M3 - Conference contribution

AN - SCOPUS:85076785379

T3 - 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019

BT - 2019 IEEE Industry Applications Society Annual Meeting, IAS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

ER -