Numerical Simulation of Tornado-Like Vortices: Swirl Ratio Optimization

This paper introduces an inverse design framework for tuning a digital twin of the Texas Tech VorTECH tornado simulator to reproduce tornado-like vortices. A three-dimensional RANS model using OpenFOAM (SST k–omega) is first verified against experimental tangential velocity profiles across two swirl ratios (S = 0.17 and 0.36). An adjoint‐based optimization (DAFoam) is applied to adjust the inlet turning vane angle, minimizing the mean squared error between simulated and experiment velocity profiles. For S = 0.17 and 0.36, the optimized vane angles reduce profile error by 19.2% and 30.9%, respectively, with consistent convergence from varied initial guesses. These results demonstrate the robustness and effectiveness of adjoint‐based inverse design for tornado simulators and lay the groundwork for expanding to multi‐parameter optimizations and transient simulations to replicate real‐world tornadic flow fields.