CDS&E: Health-Aware Optimization of Battery Charging for Proactive Prevention of Lithium Plating

University Of Connecticut, Storrs CT

Investigators

Abstract

The research objective is to advance the state of the art in battery health management by furthering our understanding of lithium plating and creating an online charging optimization approach to prevent plating. Lithium-ion (Li-ion) batteries have been increasingly adopted as energy storage devices in electric vehicles, renewable energy storage, implantable medical devices, and many other applications. Lithium plating is a critical failure mode in Li-ion cells that prevents fast charging and causes safety concerns. The success of this project will result in major advances in enhancing the safety and extending the lifetime of Li-ion batteries. Advances in battery reliability and safety will promote the electrification of transportation and penetration of renewables on the grid, accelerating our nation’s transition towards a more environmentally friendly economy. The education and outreach plan aims to integrate the newly generated knowledge into education and outreach activities, emphasizing broadening the participation of students in battery safety science. This plan will focus on three activities: (1) developing an open online repository of educational materials; (2) engaging community college, undergraduate, and graduate students in research on battery safety; and (3) giving Lunch & Learn Talks and laboratory tours to middle- and high-school students. This project will create a physics-informed active safety platform that enables prediction and proactive prevention of lithium plating. It will lay the foundation for a paradigm shift away from a focus on reactive SOH monitoring toward one that actively assesses and mitigates the risk of lithium plating. The research plan consists of four thrusts: (1) validation of multiphysics plating models; (2) prediction of lithium plating probability; (3) optimization of charging protocols; and (4) platform validation using two real-world applications. A novelty of the proposed active safety platform lies in the integration of multiphysics modeling, degradation diagnostics, and plating prognostics within a charging optimization process. In other words, this platform combines physics-based modeling and machine learning to achieve control-enabled active safety. As a result, this project will produce important insights into the fundamental mechanisms by which a charging optimization platform harnesses physics and data to reduce the likelihood of lithium plating, thus providing the scientific knowledge necessary for exploring novel strategies for physics-informed learning and optimization and promoting wider adoption of this new class of fast charging enablers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.