Loading...
Identifying thermo-kinetic parameters in lithium-sulfur battery models with optimization algorithms and the effect on model predictive capability
Korff, Daniel
Korff, Daniel
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2023
Date Submitted
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
The work presented here explores the impact of reaction mechanism complexity on
model-predictive capability for physics-based lithium-sulfur (Li-S) battery models. The
conversion nature of Li-S cathodes includes intermediate species, called polysulfides, that
are produced which are soluble in liquid electrolyte. These soluble polysulfides (PSs) lead
to many challenges for commercializing Li-S batteries. Physics-based models provide a
route to improving understanding of Li-S batteries; however, there are many questions
remaining with regards to reaction mechanism complexity, thermo-kinetic parameters for
species and reactions, and the impact these things have on model veracity. This work
demonstrates that assumptions regarding the solvation structure of species being modeled
and the reaction mechanism used has a significant impact on results at multiple levels. Due
to the complexity of Li-S batteries, cell voltage during charge/discharge cycles is only one
quantitative element to phenomena that are occurring during use. The evolution of species
concentrations throughout charge/discharge is also crucial to guide design and operation
decisions for Li-S batteries to avoid rapid degradation in battery performance. Increasing
reaction mechanism complexity increases the interplay between species in the electrolyte.
However, the increase in mechanism complexity increases the number of thermo-kinetic
parameters, for which there is little or no good knowledge of physically meaningful values.
In order to address this gap, the global optimization method differential evolution (DE) is
used to explore parameter spaces in a much more thorough way than could be done by
hand. The results demonstrate that improvement in goodness of fit with experimental data
is significant and the impact on model predictive capability further shows the importance
of the reaction mechanism used.
Associated Publications
Rights
Copyright of the original work is retained by the author.