Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells

Muhammad Anjum; Moizzah Asif; Jonathan Williams

doi:10.1007/978-3-030-63916-7

Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells

Muhammad Anjum, Moizzah Asif, Jonathan Williams

Cyfadran Cyfrifiadureg, Peirianneg a Gwyddoniaeth

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Cyfraniad i bennod arall › adolygiad gan gymheiriaid

Crynodeb

This paper has identified a minimal configuration of a DNN architecture and hyperparameter settings to effectively estimate SOC of EV battery cells. The results from the experimental work has shown that a minimal configuration of hidden layers and neurons can reduce the computational cost and resources required without compromising the performance. This is further supported by the number of epochs taken to train the best DNN SOC estimations model. Hence, demonstrating that the risk of overfitting estimation models to training datasets can also be reduced. This is further supported by the generalisation capability of the best model demonstrated through the decrease in error metrics values from test phase to those in validation phase.

Iaith wreiddiol	Saesneg
Teitl	Energy and Sustainable Futures
Is-deitl	Proceedings of 2nd ICESF 2020
Golygyddion	Iosif Mporas, Pandelis Kourtessis, Amin Al-Habaibeh, Abhishek Asthana, Vladimir Vukovic, John Senior
Man cyhoeddi	Hertfordshire
Cyhoeddwr	Springer
Pennod	1
Tudalennau	11-18
Nifer y tudalennau	294
Argraffiad	1
ISBN (Electronig)	978-3-030-63916-7, 978-3-030-63918-1
ISBN (Argraffiad)	978-3-030-63915-0
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.1007/978-3-030-63916-7
Statws	Cyhoeddwyd - 4 Tach 2020
Digwyddiad	The International Conference on Energy and Sustainable Futures (ICESF) - Online, Hertfordshire, Y Deyrnas Unedig Hyd: 10 Medi 2020 → 11 Medi 2020

Cyfres gyhoeddiadau

Enw	Springer Proceedings in Energy
Cyhoeddwr	Springer International Publishing
ISSN (Argraffiad)	2352-2534
ISSN (Electronig)	2352-2534

Cynhadledd

Cynhadledd	The International Conference on Energy and Sustainable Futures (ICESF)
Teitl cryno	ICESF
Gwlad/Tiriogaeth	Y Deyrnas Unedig
Dinas	Hertfordshire
Cyfnod	10/09/20 → 11/09/20

NDC y CU

Mae’r allbwn hwn yn cyfrannu at y Nod(au) Datblygu Cynaliadwy canlynol

Mynediad at Ddogfen

10.1007/978-3-030-63916-7

Ffeiliau eraill a chysylltiadau

https://link.springer.com/book/10.1007/978-3-030-63916-7

Dyfynnu hyn

Anjum, M., Asif, M., & Williams, J. (2020). Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells. Yn I. Mporas, P. Kourtessis, A. Al-Habaibeh, A. Asthana, V. Vukovic, & J. Senior (Gol.), Energy and Sustainable Futures: Proceedings of 2nd ICESF 2020 (1 gol., tt. 11-18). erthygl 2 (Springer Proceedings in Energy). Springer. https://doi.org/10.1007/978-3-030-63916-7

Anjum, Muhammad ; Asif, Moizzah ; Williams, Jonathan. / Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells. Energy and Sustainable Futures: Proceedings of 2nd ICESF 2020. Gol. / Iosif Mporas ; Pandelis Kourtessis ; Amin Al-Habaibeh ; Abhishek Asthana ; Vladimir Vukovic ; John Senior. 1. gol. Hertfordshire : Springer, 2020. tt. 11-18 (Springer Proceedings in Energy).

@inbook{ac488462a0194da2abcb45774a7bd0f3,

title = "Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells",

abstract = "This paper has identified a minimal configuration of a DNN architecture and hyperparameter settings to effectively estimate SOC of EV battery cells. The results from the experimental work has shown that a minimal configuration of hidden layers and neurons can reduce the computational cost and resources required without compromising the performance. This is further supported by the number of epochs taken to train the best DNN SOC estimations model. Hence, demonstrating that the risk of overfitting estimation models to training datasets can also be reduced. This is further supported by the generalisation capability of the best model demonstrated through the decrease in error metrics values from test phase to those in validation phase.",

author = "Muhammad Anjum and Moizzah Asif and Jonathan Williams",

year = "2020",

month = nov,

day = "4",

doi = "10.1007/978-3-030-63916-7",

language = "English",

isbn = "978-3-030-63915-0",

series = "Springer Proceedings in Energy",

publisher = "Springer",

pages = "11--18",

editor = "Iosif Mporas and Kourtessis, {Pandelis } and Amin Al-Habaibeh and Abhishek Asthana and Vladimir Vukovic and John Senior",

booktitle = "Energy and Sustainable Futures",

address = "Germany",

edition = "1",

note = "The International Conference on Energy and Sustainable Futures (ICESF), ICESF ; Conference date: 10-09-2020 Through 11-09-2020",

}

Anjum, M, Asif, M & Williams, J 2020, Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells. yn I Mporas, P Kourtessis, A Al-Habaibeh, A Asthana, V Vukovic & J Senior (gol.), Energy and Sustainable Futures: Proceedings of 2nd ICESF 2020. 1 gol., 2, Springer Proceedings in Energy, Springer, Hertfordshire, tt. 11-18, The International Conference on Energy and Sustainable Futures (ICESF), Hertfordshire, Y Deyrnas Unedig, 10/09/20. https://doi.org/10.1007/978-3-030-63916-7

Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells. / Anjum, Muhammad; Asif, Moizzah; Williams, Jonathan.
Energy and Sustainable Futures: Proceedings of 2nd ICESF 2020. gol. / Iosif Mporas; Pandelis Kourtessis; Amin Al-Habaibeh; Abhishek Asthana; Vladimir Vukovic; John Senior. 1. gol. Hertfordshire: Springer, 2020. t. 11-18 2 (Springer Proceedings in Energy).

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Cyfraniad i bennod arall › adolygiad gan gymheiriaid

TY - CHAP

T1 - Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells

AU - Anjum, Muhammad

AU - Asif, Moizzah

AU - Williams, Jonathan

PY - 2020/11/4

Y1 - 2020/11/4

N2 - This paper has identified a minimal configuration of a DNN architecture and hyperparameter settings to effectively estimate SOC of EV battery cells. The results from the experimental work has shown that a minimal configuration of hidden layers and neurons can reduce the computational cost and resources required without compromising the performance. This is further supported by the number of epochs taken to train the best DNN SOC estimations model. Hence, demonstrating that the risk of overfitting estimation models to training datasets can also be reduced. This is further supported by the generalisation capability of the best model demonstrated through the decrease in error metrics values from test phase to those in validation phase.

AB - This paper has identified a minimal configuration of a DNN architecture and hyperparameter settings to effectively estimate SOC of EV battery cells. The results from the experimental work has shown that a minimal configuration of hidden layers and neurons can reduce the computational cost and resources required without compromising the performance. This is further supported by the number of epochs taken to train the best DNN SOC estimations model. Hence, demonstrating that the risk of overfitting estimation models to training datasets can also be reduced. This is further supported by the generalisation capability of the best model demonstrated through the decrease in error metrics values from test phase to those in validation phase.

UR - https://link.springer.com/book/10.1007/978-3-030-63916-7

U2 - 10.1007/978-3-030-63916-7

DO - 10.1007/978-3-030-63916-7

M3 - Other chapter contribution

SN - 978-3-030-63915-0

T3 - Springer Proceedings in Energy

SP - 11

EP - 18

BT - Energy and Sustainable Futures

A2 - Mporas, Iosif

A2 - Kourtessis, Pandelis

A2 - Al-Habaibeh, Amin

A2 - Asthana, Abhishek

A2 - Vukovic, Vladimir

A2 - Senior, John

PB - Springer

CY - Hertfordshire

T2 - The International Conference on Energy and Sustainable Futures (ICESF)

Y2 - 10 September 2020 through 11 September 2020

ER -

Anjum M, Asif M, Williams J. Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells. Yn Mporas I, Kourtessis P, Al-Habaibeh A, Asthana A, Vukovic V, Senior J, golygyddion, Energy and Sustainable Futures: Proceedings of 2nd ICESF 2020. 1 gol. Hertfordshire: Springer. 2020. t. 11-18. 2. (Springer Proceedings in Energy). doi: 10.1007/978-3-030-63916-7

Towards an optimal deep neural network for SOC estimation of Electric-Vehicle Lithium-ion battery cells

Crynodeb

Cyfres gyhoeddiadau

Cynhadledd

NDC y CU

Mynediad at Ddogfen

Ffeiliau eraill a chysylltiadau

Ôl bys

Dyfynnu hyn