Lithium ion batteries are complex. Their performance depends upon a myriad of interactions between numerous materials under a broad range of current and voltage. As yet, battery performance cannot be predicted from analytical or computational characterization of the individual components; instead, progress occurs primarily through physical experimentation. To accelerate the process of battery research, Wildcat developed a unique high throughput platform for battery research. In addition to accelerating research, this platform ensures high quality, reproducible and cross-comparable data is collected. Importantly, this data includes both good and bad results. Such data is well suited to the creation of predictive models and, to further enhance our research efficiency, we are augmenting our system with AI. In this presentation, I will discuss Wildcat’s approach to battery research, our high throughput data system and learnings from deploying machine learning models into the system.

Data systems for high throughput lithium ion research

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.