Mechanisms of a Convolutional Neural Network for Learning Three-dimensional Unsteady Wake Flow

Abstract

Convolutional neural networks (CNNs) have recently been applied to predict or model fluid dynamics. However, mechanisms of CNNs for learning fluid dynamics are still not well understood, while such understanding is highly necessary to optimize the network or to reduce trial-and-errors during the network optmization. In the present study, a CNN to predict future three-dimensional unsteady wake flow using flow fields in the past occasions is developed. Mechanisms of the developed CNN for prediction of wake flow behind a circular cylinder are investigated in two flow regimes: the three-dimensional wake transition regime and the shear-layer transition regime. Feature maps in the CNN are visualized to compare flow structures which are extracted by the CNN from flow at the two flow regimes. In both flow regimes, feature maps are found to extract similar sets of flow structures such as braid shear-layers and shedding vortices. A Fourier analysis is conducted to investigate mechanisms of the CNN for predicting wake flow in flow regimes with different wave number characteristics. It is found that a convolution layer in the CNN integrates and transports wave number information from flow to predict the dynamics. Characteristics of the CNN for transporting input information including time histories of flow variables is analyzed by assessing contributions of each flow variable and time history to feature maps in the CNN. Structural similarities between feature maps in the CNN are calculated to reveal the number of feature maps that contain similar flow structures. By reducing the number of feature maps that contain similar flow structures, it is also able to successfully reduce the number of parameters to learn in the CNN by 85\% without affecting prediction performances.

References in corpus (21)

• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Understanding Neural Networks Through Deep Visualization
• Super-resolution reconstruction of turbulent flows with machine learning
• Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
• Travelling waves in pipe flow
• Deep multi-scale video prediction beyond mean square error
• Data-driven prediction of unsteady flow fields over a circular cylinder using deep learning
• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• Convolutional neural network based hierarchical autoencoder for nonlinear mode decomposition of fluid field data
• Machine-learning-based reduced order modeling for unsteady flows around bluff bodies of various shapes
• A new prediction method of unsteady wake flow by the hybrid deep neural network
• Probabilistic neural networks for fluid flow surrogate modeling and data recovery
• Coherent structures and extreme events in rotating multiphase turbulent flows
• Deep unsupervised learning of turbulence for inflow generation at various Reynolds numbers
• Experimental velocity data estimation for imperfect particle images using machine learning
• An Efficient Deep Learning Technique for the Navier-Stokes Equations: Application to Unsteady Wake Flow Dynamics
• Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
• Three-dimensional fluid motion in Faraday waves: creation of vorticity and generation of two-dimensional turbulence
• Networked oscillator based modeling and control of unsteady wakes
• Prediction of laminar vortex shedding over a cylinder using deep learning
• Predicting Hurricane Trajectories using a Recurrent Neural Network

Cited by in corpus (4)

• Experimental velocity data estimation for imperfect particle images using machine learning
• Data-driven nonlinear turbulent flow scaling with Buckingham Pi variables
• Detection of magnetohydrodynamic waves by using machine learning
• Machine Learning for Naval Architecture, Ocean and Marine Engineering