Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm

M. S. Hossain Lipu, Aini Hussain, Mohamad Hanif Md Saad, Afida Ayob, Hannan M A

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

7 Citations (Scopus)

Abstract

This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53% and 50% than BPNN based PSO model at 25°C.

Original languageEnglish
Title of host publicationISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages354-359
Number of pages6
ISBN (Electronic)9781538635278
DOIs
Publication statusPublished - 5 Jul 2018
Event2018 IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2018 - Penang Island, Malaysia
Duration: 28 Apr 201829 Apr 2018

Other

Other2018 IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2018
CountryMalaysia
CityPenang Island
Period28/4/1829/4/18

Fingerprint

Lithium-ion Battery
Recurrent neural networks
Recurrent Neural Networks
Neural Network Model
electric batteries
lithium
Charge
Back-propagation Neural Network
ions
Particle swarm optimization (PSO)
Mean square error
Particle Swarm Optimization
Neuron
root-mean-square errors
Roots
Backpropagation
neurons
Model Robustness
Feedback Delay
Neurons

Keywords

  • lithium-ion NMC battery
  • NARX neural network
  • Particle swarm optimization
  • State of charge
  • US06 drive cycle

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Computer Networks and Communications
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Control and Optimization
  • Instrumentation

Cite this

Hossain Lipu, M. S., Hussain, A., Md Saad, M. H., Ayob, A., & M A, H. (2018). Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm. In ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics (pp. 354-359). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISCAIE.2018.8405498

Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm. / Hossain Lipu, M. S.; Hussain, Aini; Md Saad, Mohamad Hanif; Ayob, Afida; M A, Hannan.

ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics. Institute of Electrical and Electronics Engineers Inc., 2018. p. 354-359.

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

Hossain Lipu, MS, Hussain, A, Md Saad, MH, Ayob, A & M A, H 2018, Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm. in ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics. Institute of Electrical and Electronics Engineers Inc., pp. 354-359, 2018 IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2018, Penang Island, Malaysia, 28/4/18. https://doi.org/10.1109/ISCAIE.2018.8405498
Hossain Lipu MS, Hussain A, Md Saad MH, Ayob A, M A H. Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm. In ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics. Institute of Electrical and Electronics Engineers Inc. 2018. p. 354-359 https://doi.org/10.1109/ISCAIE.2018.8405498
Hossain Lipu, M. S. ; Hussain, Aini ; Md Saad, Mohamad Hanif ; Ayob, Afida ; M A, Hannan. / Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm. ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 354-359
@inproceedings{adbdda77c0634f5ca9133ce2f71521e1,
title = "Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm",
abstract = "This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53{\%} and 50{\%} than BPNN based PSO model at 25°C.",
keywords = "lithium-ion NMC battery, NARX neural network, Particle swarm optimization, State of charge, US06 drive cycle",
author = "{Hossain Lipu}, {M. S.} and Aini Hussain and {Md Saad}, {Mohamad Hanif} and Afida Ayob and {M A}, Hannan",
year = "2018",
month = "7",
day = "5",
doi = "10.1109/ISCAIE.2018.8405498",
language = "English",
pages = "354--359",
booktitle = "ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Improved recurrent NARX neural network model for state of charge estimation of lithium-ion battery using pso algorithm

AU - Hossain Lipu, M. S.

AU - Hussain, Aini

AU - Md Saad, Mohamad Hanif

AU - Ayob, Afida

AU - M A, Hannan

PY - 2018/7/5

Y1 - 2018/7/5

N2 - This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53% and 50% than BPNN based PSO model at 25°C.

AB - This paper aims to develop an accurate estimation technique for computing state of charge (SOC) of a lithium-ion battery using recurrent neural network algorithm. Nonlinear autoregressive with exogenous input (NARX) model is a well-known subclass of the recurrent neural network which has proven to be very effective and computationally rich for controlling dynamic system and predicting time series. However, the accuracy of recurrent NARX neural network depends on the amount of input and output order as well as a number of neurons in a hidden layer. Therefore, this study presents an improved recurrent NARX neural network based SOC estimation with particle swarm optimization (PSO) algorithm for finding the best value of input delays, feedback delays and a number of neurons in a hidden layer. The proposed model uses three most significant factor such as current, voltage and temperature without considering battery model. The model robustness is checked at low temperature (0°C), medium temperature (25°C), and high temperature (45°C). The US06 drive cycle is selected for model training and testing. The effectiveness of the proposed approach is compared with the back-propagation neural network (BPNN) optimized by PSO based on the SOC error, root mean square error (RMSE) and mean absolute error (MAE) and average execution time (AET). The results prove that the proposed model has higher estimation speed and achieves higher accuracy in reducing RMSE and MAE by 53% and 50% than BPNN based PSO model at 25°C.

KW - lithium-ion NMC battery

KW - NARX neural network

KW - Particle swarm optimization

KW - State of charge

KW - US06 drive cycle

UR - http://www.scopus.com/inward/record.url?scp=85050666856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050666856&partnerID=8YFLogxK

U2 - 10.1109/ISCAIE.2018.8405498

DO - 10.1109/ISCAIE.2018.8405498

M3 - Conference contribution

AN - SCOPUS:85050666856

SP - 354

EP - 359

BT - ISCAIE 2018 - 2018 IEEE Symposium on Computer Applications and Industrial Electronics

PB - Institute of Electrical and Electronics Engineers Inc.

ER -