Recent advances and achievements in reinforcement learning (RL) have opened the door for its applications to a broad range of power system problems, including distributed energy resource (DER) coordination. This talk will cover deep RL methods for DER coordination, with the loss of life of battery energy storage system (BESS) explicitly modeled. The modeling of BESS loss of life accounts for both calendrical and cyclical aging effects, explicitly modeling the impacts of ambient temperature on BESS lifespan. The talk will also present how a customized policy design can be developed, utilizing the underlying properties of the problem to improve the exploration capability and learning efficiency of existing deep RL approaches. Additionally, the talk will showcase comprehensive case studies and comparative studies to demonstrate solution accuracy and the effects of incorporating degradation models into control design. 

Biography:

AVIJIT DAS (Member, IEEE) received the Ph.D. degree in Electrical Engineering from Florida Atlantic University, Boca Raton, FL, in 2021, the M.S. degree in Electrical Engineering from South Dakota State University, SD, in 2017, and the B.S. degree in Electrical and Electronic Engineering from American International University-Bangladesh, Bangladesh, in 2013. He is currently an Energy System Optimization and Control Engineer at Pacific Northwest National Laboratory. His research interests include optimization in smart grid systems, machine learning applications in power systems, grid operation with distributed energy resources, electric vehicles, and power system resilience and reliability. He is also actively involved in professional community engagements, including serving as a member of the IEEE CIS ADPRL technical committee, various IEEE PES subcommittees, and as a reviewer for many reputed journals.