Photo credit: Ian Wallman
Presentation
This project aims at developing a systematic framework to identify crucial physico-chemical parameters towards developing a digital twin for lithium-ion batteries. Such endeavor opens the door to battery optimization and monitoring/control, which has the potential to improve performance and lifetime of current batteries. First, the identifiability problem of electrochemical models for different battery chemistries (e.g. LFP and NCA) will be studied. The combination of different estimation techniques (data-only and dedicated tests) will be assessed in order to produce an efficient estimator with possibly optimal (minimum model-data error) guarantees. Secondly, battery degradation mechanisms will be incorporated into the model through datadriven and physics-based modelling frameworks. Finally, an arrangement of two cells will be considered, in order to start extending the proposed estimation approaches to multiple cells scenarios.
Promoters
- Michel Kinnaert, Department of Control Engineering and System Analysis, ULB
- David Howey, Department of Engineering Science, University of Oxford