Page 105 - EER-2-1

Explora: Environment
and Resource Artificial neural networks

temperatures and states of charge. Notably, the IR energy storage system design and management across
exhibited increases at both low and high SoC levels, with various industries.
temperature significantly impacting resistance values. The
capacity tests demonstrated a clear relationship between Acknowledgments
discharge rates, temperature, and cell capacity, with This study was supported by Solithor BV, a company
higher temperatures generally yielding increased capacity. located in Belgium.
Lifetime characterization tests uncovered critical factors
affecting battery longevity. While normal operating Funding
conditions displayed consistent degradation patterns, None.
severe conditions, such as high C-rates and extreme
temperatures, accelerated capacity loss and resistance Conflict of interest
growth significantly. Interestingly, rest time appeared to
have minimal impact on cell lifespan, whereas dynamic The authors have no conflicts of interest to declare.
cycling profiles proved particularly detrimental to battery Author contributions
capacity.
Conceptualization: Joris Jaguemont
The coupled ANN electrothermal model, derived from Data curation: Joris Jaguemont
characterization data, offers several advantages for battery Formal analysis: Joris Jaguemont
performance estimation. A key benefit of this model is its Funding acquisition: Fanny Bardé
ability to predict both thermal and electrical performances Investigation: Joris Jaguemont
simultaneously at different stages of the cell’s lifetime. Methodology: Joris Jaguemont
The model demonstrates high accuracy in its predictions, Project administration: Joris Jaguemont
with voltage predictions showing a maximum error of Resources: Joris Jaguemont
<2%. Temperature estimations exhibit minor deviations, Software: Joris Jaguemont
particularly in the middle of the discharge phase, but Supervision: Ali Darwiche
maintain an overall error below 3%. These results confirm Validation: Joris Jaguemont
the model’s reliability in predicting battery performance Visualization: Joris Jaguemont
under various conditions. Writing–original draft: Joris Jaguemont
To highlight the novelty and advantages of our Writing–review & editing: Joris Jaguemont, Ali Darwiche
approach, a comparison of the work with related studies is
reported in Table 7. Ethics approval and consent to participate
This study contributes valuable data and modeling Not applicable.
techniques to the field of battery electrothermal lifetime Consent for publication
modeling, offering insights that can inform the design and
optimization of battery models and battery management Not applicable.
systems. The findings have significant implications
for improving battery performance prediction and Availability of data
management, particularly in EV applications. Future Data are available on request from the authors.
work should focus on enhancing the model’s capabilities
through several avenues. Integration of more sophisticated References
algorithms and expansion of the training dataset could 1. Armand M, Axmann P, Bresser D, et al. Lithium-
further improve prediction accuracy. In addition, ion batteries-current state of the art and anticipated
validation across a wider range of operational conditions, developments. J Power Sources. 2020;479:228708.
including extreme environments and diverse battery doi: 10.1016/j.jpowsour.2020.228708
chemistries, would provide valuable insights into the
model’s robustness and applicability in various scenarios. 2. Elalfy DA, Gouda E, Kotb MF, Bureš V, Sedhom BE.
Developing advanced in situ diagnostic techniques and Comprehensive review of energy storage systems
real-time health monitoring algorithms could enhance technologies, objectives, challenges, and future trends.
Energy Strateg Rev. 2024;54:101482.
the model’s practical utility. These future directions aim
to develop a more comprehensive and widely applicable doi: 10.1016/j.esr.2024.101482
battery modeling framework, contributing both to 3. Jindal P, Katiyar R. Evaluation of accuracy for Bernardi
academic knowledge and practical advancements in equation in estimating heat generation rate for continuous

Volume 2 Issue 1 (2025) 12 doi: 10.36922/eer.7228