Tired of wrestling with unstable CFD simulations? We’ve got your back and your mesh! Recent work by now-former PhD student Mohammad Zandsalimy unveils a groundbreaking method for improving numerical stability using Dynamic Mode Decomposition (DMD).
We leverage DMD to understand how the numerical simulation evolves, allowing us to represent it with less complexity. By analyzing this simplified system, we pinpoint the key elements in the mesh that significantly impact numerical stability. We then strategically adjust these elements to create an optimized mesh that enhances numerical stability without any modifications to the flow solver.
We leverage DMD to understand how the numerical simulation evolves, allowing us to represent it with less complexity. By analyzing this simplified system, we pinpoint the key elements in the mesh that significantly impact numerical stability. We then strategically adjust these elements to create an optimized mesh that enhances numerical stability without any modifications to the flow solver.
This research pushes the boundaries of mesh optimization for stabilizing CFD simulations. To read more, go to https://doi.org/10.1016/j.jcp.2024.113195