Abstract:Data-driven modeling methods have become one of the main technologies for predicting aerodynamic heat in hypersonic conditions. However, due to the limitations of wind tunnel experimental conditions, the spatial distribution of aerothermal wind tunnel experimental data is often sparse, and the sample size is relatively small. Furthermore, there is a lack of direct correlation in the aerodynamic heat distribution data among different shapes of vehicles, which poses challenges for constructing high-performance data-driven aerodynamic heat prediction models. To address these issues, this paper proposes a high-precision aerodynamic heat modeling and prediction method based on data augmentation and transfer learning. First, integrating the concept of data fusion, we propose to enhance the sparse aerothermal wind tunnel experimental data by using deep neural networks and introducing low-precision numerical computation results. Next, based on the close physical correlation between boundary layer outer edge information and wall surface aerodynamic heat, we construct the aerodynamic heat prediction model ED-ResNet using a double-series residual neural network. Finally, by fine-tuning the ED-ResNet model for transfer learning, high-precision predictions of aerothermal wind tunnel experimental results for different shaped vehicles are achieved under small sample conditions. Verification using hypersonic double-ellipsoid, blunt cone, and blunt bicone shows that after data augmentation, the prediction error of the aerodynamic heat prediction model is significantly reduced to 1/3 of that when data augmentation is not used. Moreover, through transfer learning, the model effectively leverages existing hypersonic double-ellipsoid aerothermal wind tunnel experimental data to achieve high-precision predictions of aerodynamic heat distribution for blunt cone and blunt double cone under different incoming flow conditions, with normalized root mean square error(NRMSE) maintained below 10 %.

Transfer Learning from Two-Dimensional Supercritical Airfoils to Three-Dimensional Transonic Swept Wings

Study of transfer learning from 2D supercritical airfoils to 3D transonic swept wings

Rapid aerodynamic prediction of swept wings via physics-embedded transfer learning

Transferable Machine Learning Model for the Aerodynamic Prediction of Swept Wings

Aerodynamic Optimization of a Transonic Fan Rotor by Blade Sweeping Using Adaptive Gaussian Process

Machine-Learning for Nonintrusive Model Order Reduction of the Parametric Inviscid Transonic Flow past an airfoil

Prediction of transonic flow over supercritical airfoils using geometric-encoding and deep-learning strategies

Aerodynamic optimization of high-lift devices using a 2D-to-3D optimization method based on deep reinforcement learning and transfer learning

Supercritical wing design based on airfoil optimization and 2.75D transformation

Steady-State Transonic Flowfield Prediction via Deep-Learning Framework

One neural network approach for the surrogate turbulence model in transonic flows

Rapid prediction of unsteady aerodynamic characteristics of flapping wing based on GRU

Unsteady aerodynamic prediction using limited samples based on transfer learning

Learning Active Flow Control Strategies of a Swept Wing by Intelligent Wind Tunnel

High precision aerodynamic heat prediction method based on data augmentation and transfer learning

An adaptive machine learning based optimization methodology in the aerodynamic analysis of a finite wing under various cruise conditions

Combined space–time reduced-order model with three-dimensional deep convolution for extrapolating fluid dynamics

Machine Learning-augmented Predictive Modeling of Turbulent Separated Flows over Airfoils

Deep Learning for Multi-Fidelity Aerodynamic Distribution Modeling from Experimental and Simulation Data

Deep learning-based predictive modelling of transonic flow over an aerofoil

Learning the aerodynamic design of supercritical airfoils through deep reinforcement learning