Spatiotemporal wall pressure forecast of a rectangular cylinder with physics-aware DeepU-Fourier neural network

Abstract

The wall pressure is of great importance in understanding the forces and structural responses induced by fluid. Recent works have investigated the potential of deep learning techniques in predicting mean pressure coefficients and fluctuating pressure coefficients, but most of existing deep learning frameworks are limited to predicting a single snapshot using full spatial information. To forecast spatiotemporal wall pressure of flow past a rectangular cylinder, this study develops a physics-aware DeepU-Fourier neural Network (DeepUFNet) deep learning model. DeepUFNet comprises the UNet structure and the Fourier neural network, with physical high-frequency loss control embedded in the model training stage to optimize model performance. Wind tunnel testing was performed to collect wall pressures on two-dimensional rectangular cylinders using high-frequency pressure scanning, thereby constructing a database for DeepUFNet training and testing. The DeepUFNet model is found capable of forecasting spatiotemporal wall pressure information with high accuracy on the rectangular cylinder with side ratio 1.5. The comparison between forecast results and experimental data presents agreement in statistical information and physical interpretation. It is also found that embedding a physical high-frequency loss control coefficient b in the DeepUFNet model can significantly improve model performance in forecasting spatiotemporal wall pressure information, particularly, high-order frequency fluctuation and wall pressure variance. Furthermore, the DeepUFNet extrapolation capability is tested with sparse spatial information input, and the model presents a satisfactory extrapolation ability. Last, the DeepUFNet is tested for generalization in unseen cases, rectangular cylinders with side ratio 4 and 3.75, and the model presents satisfactory generalization ability.

References in corpus (21)

• Machine Learning for Fluid Mechanics
• Solving high-dimensional partial differential equations using deep learning
• Deep learning for universal linear embeddings of nonlinear dynamics
• Spectral proper orthogonal decomposition and its relationship to dynamic mode decomposition and resolvent analysis
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Spectral proper orthogonal decomposition
• Nonlinear mode decomposition with convolutional neural networks for fluid dynamics
• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• Sub-grid modelling for two-dimensional turbulence using neural networks
• Unsupervised deep learning for super-resolution reconstruction of turbulence
• Sparse reduced-order modeling : Sensor-based dynamics to full-state estimation
• Robust flow field reconstruction from limited measurements via sparse representation
• An efficient streaming algorithm for spectral proper orthogonal decomposition
• Multi-fidelity prediction of fluid flow and temperature field based on transfer learning using Fourier Neural Operator
• Spectral POD analysis of the turbulent wake of a disk at Re = 50, 000
• Multi-scale data reconstruction of turbulent rotating flows with Gappy POD, Extended POD and Generative Adversarial Networks
• Energy transfer mechanisms and resolvent analysis in the cylinder wake
• Influence of adversarial training on super-resolution turbulence reconstruction
• Structure of turbulence in the flow around a rectangular cylinder
• Low-Order Flow Reconstruction and Uncertainty Quantification in Disturbed Aerodynamics Using Sparse Pressure Measurements
• Model-Based Estimation of Vortex Shedding in Unsteady Cylinder Wakes