Assessment of supervised machine learning methods for fluid flows
Abstract
We apply supervised machine learning techniques to a number of regression problems in fluid dynamics. Four machine learning architectures are examined in terms of their characteristics, accuracy, computational cost, and robustness for canonical flow problems. We consider the estimation of force coefficients and wakes from a limited number of sensors on the surface for flows over a cylinder and NACA0012 airfoil with a Gurney flap. The influence of the temporal density of the training data is also examined. Furthermore, we consider the use of convolutional neural network in the context of superresolution analysis of twodimensional cylinder wake, twodimensional decaying isotropic turbulence, and threedimensional turbulent channel flow. In the concluding remarks, we summarize on findings from a range of regressiontype problems considered herein.
 Publication:

Theoretical and Computational Fluid Dynamics
 Pub Date:
 February 2020
 DOI:
 10.1007/s0016202000518y
 arXiv:
 arXiv:2001.09618
 Bibcode:
 2020ThCFD.tmp...13F
 Keywords:

 Supervised machine learning;
 Wake dynamics;
 Turbulence;
 Physics  Fluid Dynamics;
 Physics  Computational Physics
 EPrint:
 doi:10.1007/s0016202000518y