Modeling Battery Aging for Optimal Control

When optimizing the operation of stationary batteries using optimal control in a time-of-use application, aging costs can have a strong influence on the economic outcome. Thus, it is beneficial to consider aging costs in the controller design. First, we give a detailed explanation of how to include varying stress factors into a lab-based aging model, even with non-linear dependencies on time and full equivalent cycles (FEC). We state and motivate the discrete dynamics and formulate an optimal control problem for a time-of-use scenario. As a result, we find a control strategy for the battery that provides clear benefits. Unlike the aging-unaware controller and a controller with light aging-awareness, the fully aging-aware controller manages to recoup the battery investment. Compared to a base case without any batteries, the aging-unaware and aging-light controller lose 3657 € and 710 € in every year of their operation, while the aging-aware controller gains 627 € per year because it increases the battery lifespan by 345%.