EPE Association: EPE 2025 - LS3b: Energy Storage and Management Systems

EPE 2025 - LS3b: Energy Storage and Management Systems
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2025 - Conference > EPE 2025 - Topic 03: Energy Storage Systems > EPE 2025 - LS3b: Energy Storage and Management Systems

   [return to parent folder]

   A K-nearest Neighbours Inspired Direct MPC for SOC Balancing in Smart Batteries
By Francesco SIMONETTI, Roberta DI FONSO, Remus TEODORESCU [View]
[Download]

Abstract: Lithium-ion batteries dominate energy storage systems. Working circumstances and parameter variations in single cells can result in state of charge imbalances that reduce the battery's lifetime. Modular batteries have been proven to be capable of actively reducing the imbalance among the cells by connecting or bypassing the individual batteries, sharing the work load in real-time. This paper presents a machine learning-inspired direct model predictive control for reconfigurable batteries to balance the cells that solves the underlying optimization problem in polynomial time.

   A Versatile Prognostics Approach for Batteries and Hydrogen Storage Systems Health Management
By Théo LENOIR, Daniela CHRENKO, Robin ROCHE, Samir JEMEI, Mickaël HILAIRET [View]
[Download]

Abstract: The degradation of energy storage systems, including batteries, proton exchange membrane fuel cells, and electrolyzers, presents a critical challenge to their long-term reliability and efficiency. This study addresses the prediction of Remaining Useful Life by selecting a robust state-of-health indicator tailored to each system. A novel data-driven prognostic approach is proposed, leveraging Long Short-Term Memory neural networks to capture temporal dependencies and nonlinear degradation trends. The recurrent neural network model demonstrates versatility, adapting to both electrochemical storage and hydrogen-based systems by effectively learning from diverse datasets. Results highlight the method's capability to generalize across technologies, enabling accurate degradation predictions and offering significant insights into performance management.

   Embedded Spectroscopy: Potentialities and Constraints for Onboard Battery Diagnostics
By Charles BECHARA, Guy FRIEDRICH, Christophe FORGEZ, Samuel CREGUT [View]
[Download]

Abstract: This paper introduces an on-board Electrochemical Impedance Spectroscopy (EIS) technique for battery diagnostics, utilizing numerical simulations and experimental results. EIS, commonly used in laboratories to assess charge transfer and diffusion in electrochemical cells, is ideal for monitoring battery performance, state of charge, and health. However, traditional EIS equipment is too large and expensive for automotive use. The proposed system offers a compact, cost-effective solution for electric vehicles. We discuss the challenges and trade-offs for accurate on-board measurements, based on simulations. Prototype results are then compared with laboratory EIS measurements on 260 A.h Li-NMC EV pouch cells, demonstrating that embedded spectroscopy achieves precise results, even for low-impedance cells. The study highlights the potential of this technique as a reliable and effective method for battery diagnostics in automotive applications.

EPE 2025 - LS3b: Energy Storage and Management Systems
You are here: EPE Documents > 01 - EPE & EPE ECCE Conference Proceedings > EPE 2025 - Conference > EPE 2025 - Topic 03: Energy Storage Systems > EPE 2025 - LS3b: Energy Storage and Management Systems
	[return to parent folder]

	A K-nearest Neighbours Inspired Direct MPC for SOC Balancing in Smart Batteries By Francesco SIMONETTI, Roberta DI FONSO, Remus TEODORESCU	[View] [Download]
Abstract: Lithium-ion batteries dominate energy storage systems. Working circumstances and parameter variations in single cells can result in state of charge imbalances that reduce the battery's lifetime. Modular batteries have been proven to be capable of actively reducing the imbalance among the cells by connecting or bypassing the individual batteries, sharing the work load in real-time. This paper presents a machine learning-inspired direct model predictive control for reconfigurable batteries to balance the cells that solves the underlying optimization problem in polynomial time.

	A Versatile Prognostics Approach for Batteries and Hydrogen Storage Systems Health Management By Théo LENOIR, Daniela CHRENKO, Robin ROCHE, Samir JEMEI, Mickaël HILAIRET	[View] [Download]
Abstract: The degradation of energy storage systems, including batteries, proton exchange membrane fuel cells, and electrolyzers, presents a critical challenge to their long-term reliability and efficiency. This study addresses the prediction of Remaining Useful Life by selecting a robust state-of-health indicator tailored to each system. A novel data-driven prognostic approach is proposed, leveraging Long Short-Term Memory neural networks to capture temporal dependencies and nonlinear degradation trends. The recurrent neural network model demonstrates versatility, adapting to both electrochemical storage and hydrogen-based systems by effectively learning from diverse datasets. Results highlight the method's capability to generalize across technologies, enabling accurate degradation predictions and offering significant insights into performance management.

	Embedded Spectroscopy: Potentialities and Constraints for Onboard Battery Diagnostics By Charles BECHARA, Guy FRIEDRICH, Christophe FORGEZ, Samuel CREGUT	[View] [Download]
Abstract: This paper introduces an on-board Electrochemical Impedance Spectroscopy (EIS) technique for battery diagnostics, utilizing numerical simulations and experimental results. EIS, commonly used in laboratories to assess charge transfer and diffusion in electrochemical cells, is ideal for monitoring battery performance, state of charge, and health. However, traditional EIS equipment is too large and expensive for automotive use. The proposed system offers a compact, cost-effective solution for electric vehicles. We discuss the challenges and trade-offs for accurate on-board measurements, based on simulations. Prototype results are then compared with laboratory EIS measurements on 260 A.h Li-NMC EV pouch cells, demonstrating that embedded spectroscopy achieves precise results, even for low-impedance cells. The study highlights the potential of this technique as a reliable and effective method for battery diagnostics in automotive applications.