PHYSICS-INFORMED DEEP LEARNING FOR ENHANCED LITHIUM-ION BATTERY PERFORMANCE

Mohammadamin Rezaei; Baqıtjan Suleymen; Amir Mosavi; Adrienn Dineva; Imre Felde; Stefan Gubo

doi:10.26577/jpcsit53202610

Авторлар

Mohammadamin Rezaei Fricke and Mallah Microwave Technology GmbH, Peine, Germany , John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary https://orcid.org/0009-0005-7763-2326 (расталмаған)
Baqıtjan Suleymen John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary https://orcid.org/0000-0002-9559-4758 (расталмаған)
Amir Mosavi John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary , Institute of the Information Society, Ludovika University of Public Service, Budapest, Hungary , Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia https://orcid.org/0000-0003-4842-0613 (расталмаған)
Adrienn Dineva John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary , Audi Hungaria Faculty of Automotive Engineering, Szechenyi Istvan University, Gyor, Hungary https://orcid.org/0000-0001-8043-7023 (расталмаған)
Imre Felde John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary https://orcid.org/0000-0003-4126-2480 (расталмаған)
Stefan Gubo Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia https://orcid.org/0009-0001-7386-3187 (расталмаған)

DOI:

https://doi.org/10.26577/jpcsit53202610

Түйін сөздер:

fast charging, pulsed charging, hybrid charging strategy, thermal stress reduction, polarization mitigation, artificial intelligence

Аңдатпа

Lithium-ion batteries are widely used in electric vehicles, portable electronics, and grid energy systems. The constant current–constant voltage (CC–CV) method shows clear limits at high state of charge when polarization increases, temperature rises, and charging time is extended. These effects become stronger in the high SOC region, where the charging rate decreases. Consequently, to address this challenge, in this study we presents a hybrid charging method that combines physics-informed deep learning with an auxiliary hardware circuit. The circuit is placed between the charger and the battery without changes to the charger structure. It enables real-time control of pulse frequency and duty cycle. A one-dimensional convolutional neural network is applied during short relaxation periods for SOC estimation. The proposed method reduces polarization and thermal stress and improves charging efficiency at high SOC. It also improves open-circuit voltage estimation accuracy. Experimental tests are conducted on lithium-ion cells under controlled temperature conditions. The results confirm improved performance in fast charging applications.

##plugins.themes.default.displayStats.downloads##

##plugins.themes.default.displayStats.noStats##

Автор өмірбаяндары

Mohammadamin Rezaei, Fricke and Mallah Microwave Technology GmbH, Peine, Germany, John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary

Mohammadamin Rezaei is a PhD student at the John von Neumann Faculty of Informatics, Obuda University, Budapest, Hungary. He is also affiliated with Fricke and Mallah Microwave Technology GmbH, Peine, Germany. His research interests include artificial intelligence, machine learning, and microwave technology.
Baqıtjan Suleymen, John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary

Baqıtjan Suleymen is a PhD student at the John von Neumann Faculty of Informatics, Obuda University, Budapest, Hungary. His research interests include artificial intelligence, machine learning, data science, and applied computational methods.
Amir Mosavi, John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary, Institute of the Information Society, Ludovika University of Public Service, Budapest, Hungary, Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia

Dr. Amir Mosavi is a Professor of Artificial Intelligence and Data Science. He joined Obuda University, Budapest, Hungary in 2014. He is affiliated with the John von Neumann Faculty of Informatics and the Institute of the Information Society at Ludovika University of Public Service. He is also a Visiting Professor at the Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia. His research interests include artificial intelligence, machine learning, data science, responsible AI, and sustainability.
Adrienn Dineva, John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary, Audi Hungaria Faculty of Automotive Engineering, Szechenyi Istvan University, Gyor, Hungary

Dr. Adrienn Dineva is with Obuda University, Budapest, Hungary. She is also affiliated with the Audi Hungaria Faculty of Automotive Engineering, Széchenyi Istvan University, Gyor, Hungary. Her research interests include artificial intelligence, engineering applications, and data-driven methods.
Imre Felde, John Von Neumann Faculty of Informatics, Obuda University, 1034 Budapest, Hungary

Imre Felde is a Professor at the John von Neumann Faculty of Informatics, Óbuda University, Budapest, Hungary. His research interests include computer science, artificial intelligence, data-driven methods, and applied computational technologies.
Stefan Gubo, Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia

Stefan Gubo, PhD. Dr. Stefan Gubo is with the Faculty of Economics and Informatics, J. Selye University, Komarno, Slovakia. His research interests include artificial intelligence, data science, computational methods, and interdisciplinary applications.