The National Science Foundation awarded a $670,636 project grant under the Engineering federal grant program (CFDA 47.041) to the California Institute of Technology. The grant will fund the development of a robust wall model for large-eddy simulations through reinforcement learning methods. Present wall models rely on assumptions that limit applicability, often incorrectly predicting separation and transition crucial to external aerodynamics simulations. This project aims to develop novel wall models that can adapt to various flow configurations based on instantaneous input, casting the wall modeling problem as a control problem. The discovered model will be optimized to accurately reproduce quantities of interest by automating exploration of relevant physics. Advancing wall model simulation capabilities could provide a means for cheaper, more reliable simulations of complex external aerodynamic flows such as aircraft. The grant period is from June 15, 2022 to May 31, 2025.
Generated 1/6/24, 12:35 PM