New model for predicting battery swelling under mechanical constraints
Lithium-ion batteries powering electric vehicles do not operate in isolation. Packed tightly into modules, they are subject to constant mechanical pressure, a design feature that improves stability and performance, but also introduces a hidden risk: as cells charge and discharge, they swell and contract, generating fluctuating internal stresses that can crack electrodes, deform separators, and […]
Lithium-ion batteries powering electric vehicles do not operate in isolation. Packed tightly into modules, they are subject to constant mechanical pressure, a design feature that improves stability and performance, but also introduces a hidden risk: as cells charge and discharge, they swell and contract, generating fluctuating internal stresses that can crack electrodes, deform separators, and trigger dangerous side reactions, including internal short circuits and thermal runaway.
To address this challenge, researchers within the NEMO project from Graz University of Technology (TUG) and Vrije Universiteit Brussel (VUB) have developed a novel P2D-based computational model that integrates lithium-ion transport dynamics with pressure-dependent parameters to accurately predict cell thickness changes under mechanical constraints — and does so without requiring real-time sensors during operation.
The model was validated against experimental data for a 1C discharge cycle at 0.164 MPa of external pressure. It achieved a mean absolute percentage error (MAPE) of just 6.87% in predicting cell thickness change — a strong result that demonstrates the model’s ability to bridge the gap between laboratory measurements and real-world battery pack conditions.
Crucially, unlike existing approaches that rely on embedded physical sensors or ignore mechanical boundary conditions altogether, this model is entirely simulation-based. It can therefore be integrated into battery management systems (BMS) and digital twin frameworks without hardware modifications, reducing cost and complexity.
This work, developed within the NEMO project framework, directly supports the development of next-generation battery management systems capable of accounting for mechanical stress in real time, improving safety assessment, extending battery lifetimes, and ultimately reducing the total cost of ownership of electric vehicles and stationary energy storage systems.
The full paper is available open access on Zenodo: https://zenodo.org/records/19554839
