Dataset for the paper "Steady-State Transonic Flowfield Prediction via Deep-Learning Framework"

This repository provides the Computational Fluid Dynamics (CFD) reference datasets used in:G. Immordino, A. Da Ronch, M. Righi. "Steady-State Transonic Flowfield Prediction via Deep-Learning Framework." AIAA Journal, 62(5), 1915–1931, 2024.https://doi.org/10.2514/1.J063545 The datasets contain steady Reynolds-averaged Navier–Stokes (RANS) simulations of transonic aerodynamic flows used for training and testing a fully connected neural network (FCNN) surrogate model. Three benchmark configurations of increasing geometric and physical complexity are included: Benchmark Supercritical Wing (BSCW) Semi-span wing with rectangular planform and supercritical airfoil. Grid: structured type with130k surface points. Outputs: surface pressure coefficient (CP), skin friction coefficient components (CFx,CFy,CFz). ONERA M6 Wing Swept-wing configuration with λ-shaped shock waves. Grid: structured type with149k surface points. Outputs: surface pressure coefficient (CP), skin friction coefficient components (CFx,CFy,CFz). NASA Common Research Model (CRM) Full wing–body geometry representative of a wide-body aircraft. Grid: unstructured type with 80k surface points. Outputs: surface pressure coefficient (CP), skin friction coefficient components (CFx,CFy,CFz). Generation Details Solver: SU2 v7.2.1, RANS formulation with Spalart–Allmaras turbulence model. Convergence criterion: Cauchy residual method on CL (tolerance 10−7). Discretisation: JST central scheme with artificial dissipation; gradients via Green–Gauss method. Sampling: 70 design points generated via Latin Hypercube Sampling (LHS). Flight envelope: Mach number = [0.70 , 0.84] , angle of attack AoA = [0 , 5] deg. PurposeThese datasets were used to train, validate, and benchmark deep-learning model for predicting steady transonic flowfields. Applications include aerodynamic loads estimation, aeroelastic simulations, and uncertainty quantification. Contents CFD snapshots of surface pressure and skin-friction fields at each Mach-AoA condition for three different test cases. KeywordsTransonic aerodynamics; reduced-order modelling; deep learning; CFD; Benchmark Supercritical Wing; ONERA M6; NASA CRM; machine learning; flowfield prediction.