Optimizing Battery Design for Fast Charging through a Genetic Algorithm Based Multi-Objective Optimization Framework

Abstract

As the innovation of battery technology, batteries have higher power density than before and electric vehicles can travel longer distance per charge. However, the long charging time of electric vehicles is still one of its limitations. Electric car takes much longer time to fully charge its batteries. Fast charging and super charging are the available solutions but they have a potential of damaging battery with accelerated degradation. It is critical to reduce the charge time without compromise long-term performance. New materials, nanoscale structures, charging strategy have been studies to decrease the charging time. All of these approaches can further benefit from an optimal design of the batter structure for fast charging. We investigate the effect of design variables such as porosity, particle size, and thickness for both anode and cathode electrodes on battery charging and degradation by using a physics-based side-reaction coupled electrochemical model. We develop a computational optimization framework with the Elitist Non-Dominated Sorting Genetic Algorithm (NSGA-II) to find the optimal design variables to improve battery charging and long-term performance. The computational optimization work is computationally efficient of providing a guidance for battery design.