Degradation and thermal performance of Li-ion batteries: implications for electric vehicles

Thesis

Worldwide adoption of transport electrification creates a demand for optimised energy storage and control systems. This technology, despite being under strong development, has not yet reached its maturity and the electric vehicle (EV) performance envelope still needs improvement. This work investigates key challenges in battery management systems (BMS) and battery modelling for EVs, with a focus on ageing diagnostics, its effective inclusion within the BMS, and an accurate internal state estimation during high current applications i.e. fast charging.

Entropy profiling was investigated as a new approach to tackle the challenge of battery degradation diagnosis. This method leverages the interpretation of temperature and concentration dependencies of cell voltage to provide insight into the morphological changes experienced during battery life. The study finds that entropy profiling can successfully track ageing markers in a complementary way to differential voltage analysis, making it a useful battery diagnostics tool.

Even if degradation diagnosis is performed successfully, the inclusion of ageing information into a BMS is problematic. As an alternative, a periodic model parameter update is proposed here. The impact of this work was two-fold. Firstly, it highlights how the single particle model can accurately simulate both pristine and aged cell voltage responses with appropriate parameter updates. Secondly, it provides qualitative insight into the impact of ageing on model parameters, informing safety issues such as increased heat generation.

The prediction of heat generation during fast charging is a significant concern when considering the safety and performance of batteries. To address this issue, a pseudo-3D thermal-continuum model was proposed and tested up to 10C. The results showed that the fast diffusion encountered in high-power cells allows for substantial model simplifications without compromising prediction accuracy.

Authors: Wojtala, ME

• Alternative title: Modelling the degradation and thermal performance of Li-ion batteries: implications for electric vehicles
• DOI: 10.5287/ora-vj4g0a82q
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Li-ion; battery; multiphysics modelling; high C-rate; DFN; fast charging; thermal; entropy; EV; SPM; degradation
• Subjects: Entropy profiling; Battery degradation; Battery modelling; Fast charge modelling