Sep 2026
CFD-validated deep active subspace learning for reliable offline transonic airfoil optimization and shock-control analysis
A data-driven deep active subspace method (DASM) that jointly learns nonlinear active coordinates and a neural response for direct reduced-space optimization and converts surrogate convergence into reliable CFD improvement without adaptive enrichment is developed.
Bo Pang, Yang Zhang, Jun-Lin Li et al.
· The Physics of Fluids · 0 citations