09 Juin – Thesis defense - Matthieu Maures
10 h In videoconferencing
Performance and ageing model of parallel assemblies of Lithium-ion cells for the determination of the state of health and lifespan of batteries.
This thesis work fits into the energy transition problem. For a few years, Lithium-ion cells have been incresingly used to power electric vehicles and provide support to energy grid networks. To meet the power and energy requirements of those applications, cells are connected in series and parallel to make batteries. Multiple degradation mecanisms occur in the cells. They are affected by temperature, state of charge, or electrical sollicitations. When connected in series, cells run into other issues because of their different caracteristics and working points. To limit the degradation gap as well as provide safety measure, this type of assembly is connected to a battery management system, whose role is to ensure voltage balancing of the cells.
This method however is not used in parallel assemblies. This document aims at evaluating and modeling the current gaps caused by the parallel assembly, as well as determining if balancing is necessary. This study is both done experimentally and through simulation, on two cell technologies often used in power applications.
Firstly, a study of the inner working of an individual cell is provided. It allows for the creation of electrical, thermal, and ageing models, as well as their interactions. Given the specific needs of the targeted applications, the method used in this document extend empirical and semi-empirical models (often used for their simplicity) to larger domains of validity, most notably with regards to temperature, where the model accounts for the entire recommended working range of the cells.
Secondly, the consequences of connecting three cells in parallel are studied. The purpose of this part is to evaluate the causes (resistance, capacity, temperature) and effects of imbalances of the short-term repartition of currents. Using the different models proposed in the previous part, a simulation tool helps validating them compared to experimental data. It also provides a way to analyse unattainable experimental behaviors, in particular with regards to resistive effects.
Finally, long-term effects of the parallel connection are analysed. In this section, focus is placed onto thermal imbalances. Two parallel assemblies with different ageing protocols are studied to compare their effects. Similarly to the previous part, this study is performed experimentally as well as using simulation. Furthermore, a quick study of state of health diagnosis tools is provided, which hints at their inability to assess correct estimations when going from a cell to an assembly. This part, combined with the others, provides a conclusion regarding the risks and needs of a balancing system when cells are connected in parallel.