Published: Mar 30, 2022 by Daning Huang
Together with Dr. Herschel Pangborn from Mechanical Engineering, we obtained a seed grant from the ICDS of Penn State. The APUS lab will contribute by developing a novel geometric deep learning (GDL) methodology for the hierarchical modeling and control of multi-physics energy systems.
Specifically, the methodology will be applied to improve the efficiency, performance, and safety of electro-thermal systems used for electrified transportation. Based on the GDL model, a hierarchical control strategy will be developed to optimize electrified vehicle energy management by coordinating real-time decision-making across multiple domains and dynamic timescales. If successful, the methods will introduce a paradigm shift in distributed control, from expensive and time-consuming manual iterations to an automated and efficient procedure, and thus impacting systems integration of platforms for electrified mobility (e.g., eVTOL aircraft and naval ships).