Data-driven prediction of unsteady flow over a circular cylinder using deep learning
From MaRDI portal
Abstract: Unsteady flow fields over a circular cylinder are trained and predicted using four different deep learning networks: convolutional neural networks with and without consideration of conservation laws, generative adversarial networks with and without consideration of conservation laws. Flow fields at future occasions are predicted based on information of flow fields at previous occasions. Deep learning networks are trained first using flow fields at Reynolds numbers of 100, 200, 300, and 400, while flow fields at Reynolds numbers of 500 and 3000 are predicted using the trained deep learning networks. Physical loss functions are proposed to explicitly impose information of conservation of mass and momentum to deep learning networks. An adversarial training is applied to extract features of flow dynamics in an unsupervised manner. Effects of the proposed physical loss functions, adversarial training, and network sizes on the prediction accuracy are analyzed. Predicted flow fields using deep learning networks are in favorable agreement with flow fields computed by numerical simulations.
Recommendations
- A neural network method for the unsteady flow past a circular cylinder
- Deep learning observables in computational fluid dynamics
- Prediction of aerodynamic flow fields using convolutional neural networks
- Deep learning in fluid dynamics
- Data-driven modelling of nonlinear spatio-temporal fluid flows using a deep convolutional generative adversarial network
- Using machine learning to detect the turbulent region in flow past a circular cylinder
- Stationary flow predictions using convolutional neural networks
- Assessment of end-to-end and sequential data-driven learning for non-intrusive modeling of fluid flows
- Data-driven, physics-based feature extraction from fluid flow fields using convolutional neural networks
- Multi-Fidelity Machine Learning Applied to Steady Fluid Flows
Cites work
- scientific article; zbMATH DE number 3448257 (Why is no real title available?)
- Analysis of Fluid Flows via Spectral Properties of the Koopman Operator
- DNS of a plane mixing layer for the investigation of sound generation mechanisms
- Direct numerical simulation of turbulence in a nominally zero-pressure-gradient flat-plate boundary layer
- Discrete conservation principles in large-eddy simulation with application to separation control over an airfoil
- Dynamic mode decomposition of numerical and experimental data
- Fish swimming and bird/insect flight
- Koopman-mode decomposition of the cylinder wake
- Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
- Turbulence and the dynamics of coherent structures. I. Coherent structures
Cited in
(38)- On the locality of local neural operator in learning fluid dynamics
- Airfoil-based convolutional autoencoder and long short-term memory neural network for predicting coherent structures evolution around an airfoil
- Numerical solutions of boundary problems in partial differential equations: a deep learning framework with Green's function
- Four-dimensional variational data assimilation of a turbulent jet for super-temporal-resolution reconstruction
- WCNS3-MR-NN: a machine learning-based shock-capturing scheme with accuracy-preserving and high-resolution properties
- Deep learning-based reduced order model for three-dimensional unsteady flow using mesh transformation and stitching
- Spatial-temporal prediction model for unsteady near-wall flow around cylinder based on hybrid neural network
- Numerical study and parameter prediction of a full-free flexible flament in wakes of flapping wing
- A deep learning approach for the transonic flow field predictions around airfoils
- Neural-network-based fluid-structure interaction applied to vortex-induced vibration
- Physics-inspired architecture for neural network modeling of forces and torques in particle-laden flows
- Mesh-Conv: convolution operator with mesh resolution independence for flow field modeling
- Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows
- Physics-data combined machine learning for parametric reduced-order modelling of nonlinear dynamical systems in small-data regimes
- Deep learning of the spanwise-averaged Navier-Stokes equations
- A transformer-based synthetic-inflow generator for spatially developing turbulent boundary layers
- Reconstruction of three-dimensional turbulent flow structures using surface measurements for free-surface flows based on a convolutional neural network
- Data-driven modelling of nonlinear spatio-temporal fluid flows using a deep convolutional generative adversarial network
- Neural network training using \(\ell_1\)-regularization and bi-fidelity data
- Unsteady flow prediction from sparse measurements by compressed sensing reduced order modeling
- A robust super-resolution reconstruction model of turbulent flow data based on deep learning
- A reduced order modeling method with variable-separation-based domain decomposition for parametric dynamical systems
- Predicting drag on rough surfaces by transfer learning of empirical correlations
- Physics-guided deep learning for generating turbulent inflow conditions
- Using machine learning to detect the turbulent region in flow past a circular cylinder
- Super-resolution reconstruction of turbulent flows from a single Lagrangian trajectory
- ReF-nets: physics-informed neural network for Reynolds equation of gas bearing
- Hybrid deep learning and iterative methods for accelerated solutions of viscous incompressible flow
- Unsupervised deep learning for super-resolution reconstruction of turbulence
- Modular operator superposition (MOS): a physics-guided machine learning framework for addressing the curse of dimensionality and multiscale challenges in computational fluid dynamics
- A deep-neural-network accelerated precursor-based method for atmospheric boundary layers
- Accurate prediction of the particle image velocimetry flow field and rotor thrust using deep learning
- DUE: a deep learning framework and library for modeling unknown equations
- Deep learning of vortex-induced vibrations
- Differentiability in unrolled training of neural physics simulators on transient dynamics
- Bridging the gap: machine learning to resolve improperly modeled dynamics
- Prediction of turbulent heat transfer using convolutional neural networks
- A highly accurate strategy for data-driven turbulence modeling
This page was built for publication: Data-driven prediction of unsteady flow over a circular cylinder using deep learning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5235730)
