Real-time estimation of battery SoC through neural networks trained with model-based datasets: Experimental implementation and performance comparison
Giovanni Chianese,
Luigi Iannucci,
Ottorino Veneri and
Clemente Capasso
Applied Energy, 2025, vol. 389, issue C, No S0306261925005136
Abstract:
Data-driven methods have been widely investigated to estimate battery SoC due to their great potential in solving regression problems. However, expensive experimental campaigns are generally required to collect large training datasets. To address this need, this paper demonstrates the advantages of using a validated battery simulation model to easily generate data for training neural networks (NNs) estimating SoC. Such a procedure drastically reduces the number of experiments, which are only required to calibrate/validate a physics-based battery model and to test the NNs in real driving operative conditions. A Li-NMC storage cell for automotive applications was considered as case study to verify the presented methodology. The analysis was performed in a wide range of operative conditions in terms of temperatures and load dynamics. Offline tests, based on data collected during experiments, showed that the trained NNs were able to predict the SoC with an accuracy comparable to NNs trained with standard experimental-based procedures. In the end, the trained NNs were implemented on a microcontroller to prove their real-time applicability in BMS boards.
Date: 2025
References: Add references at CitEc
Citations:
Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0306261925005136
Full text for ScienceDirect subscribers only
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:eee:appene:v:389:y:2025:i:c:s0306261925005136
Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/journaldescription.cws_home/405891/bibliographic
http://www.elsevier. ... 405891/bibliographic
DOI: 10.1016/j.apenergy.2025.125783
Access Statistics for this article
Applied Energy is currently edited by J. Yan
More articles in Applied Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().