Physics-Based Modeling for Quantitative Assessment of Rechargeable Batteries for Electric Vehicles

Abstract

Modeling of rechargeable batteries for electric vehicles is pursued to assess new materials [[1],[2]] and to design and optimize battery systems for improved performance, life [[3],[4],[5]] and cost as well as maximize the safety. Simulation enables detailed spatial-temporal analyses of responses of interest which may be impossible or difficult to be determined by experiments. This is also usually achieved at a fraction of time and cost compared to experiments. Knowledge from detailed physics-based models is also used to develop reduced order models for battery controls. In this talk, recent modeling activities pursued in the Energy Storage, Materials and Strategy Department at Ford Motor Company will be discussed.A physics-based model [[6]] was used to simulate the galvanostatic charge [[7]] and discharge performance [[8]] of a dual lithium-ion insertion cell sandwich at various current densities. Customers of plug-in hybrid and all-electric vehicles may place a premium on the ability to quickly recharge the battery and therefore bottlenecks to fast charging at cell level have been studied in detail [7], [[9]]. Modeling results were validated with experimental data as shown in Figure 1 during discharge at various constant-current densities [8,[10]] and the accuracy of the model input parameters [[11]] were analyzed. The limitations to cell performance at high charge and discharge rates were quantified in terms of their contributions to cell overpotential as shown in Figure 2, as an example, during discharge at 3C rate, and further analyzed. Engineering solutions for averting bottlenecks to fast charge acceptance and high rate discharge performance and the corresponding improvements in performance were developed from the simulations.Other in-house efforts to understand [[12]] and measure [[13]] the input parameters for modeling supplier cell chemistries [[14]] and Li-Si negative electrodes [13] will be discussed. Ongoing collaborations with Michigan State University to investigate the performance of highly ordered hierarchical electrodes [[15]], [[16]], and room temperature ionic liquid electrolytes with Georgia Tech [[17],[18]] will be mentioned. Finally, other battery modeling projects (extended to Li-Si [[19]], solid polymer electrolytes, etc.) solved in preparation for serving as an invited industrial mentor at the Math Modeling in Industry workshop [[20]] organized by the Institute for Mathematics and its Applications at the University of Minnesota will be presented. Acknowledgements: Andy Drews and Ted Miller from Ford Motor Company are acknowledged for the support and all the collaborators as cited in the references are acknowledged for their contributions.