Machine learning methods for turbulence modeling in subsonic flows around airfoils

Reynolds-Averaged Navier-Stokes(RANS) method will still play a vital role in the following several decade in aerospace engineering. Although RANS models are widely used, empiricism and large discrepancies between models reduce the reliability of simulating complex flows. Therefore, in recent years, data-driven turbulence model has aroused widespread concern in fluid mechanics. Based on the experimental/numerical simulation results, this approach aims to modify or construct the turbulence model for specific purposes by machine learning technologies. In this paper, we take the results calculated by SA model as training data. Different from low Reynolds number turbulent flows, the data from high Reynolds number flows shows an apparent scaling effect, thus leading to difficulties in the data-driven modeling. In order to improve the fitting accuracy, we divided the flow field into near-wall region, wake region, and far-field region, and built individual model for every region. In this paper, we adopted the radial basis function neural network (RBFNN) and some auxiliary optimization algorithms to reconstruct a mapping function between mean variables and the eddy viscosity. Since this model reflects the relationship between local flow characteristics and turbulent eddy viscosity, it is independent on the airfoil shape and flow condition. The training data in this paper is generated from only three subsonic flow calculations of NACA0012 airfoil. By coupling the proposed approach with N-S equations, we calculated various flow cases as well as two different airfoils and showed the eddy viscosity contours, velocity profiles along the normal direction of wall and skin friction coefficient distributions, etc. Compared with the SA model, the results show a reasonable accuracy and better efficiency, which indicates the positive prospect of data-driven methods in turbulence modeling.

[1]  H. Lomax,et al.  Thin-layer approximation and algebraic model for separated turbulent flows , 1978 .

[2]  James A. Stover,et al.  A Neural Network For L1 Norm Linear Regression , 1990, Other Conferences.

[3]  James D. Keeler,et al.  Layered Neural Networks with Gaussian Hidden Units as Universal Approximations , 1990, Neural Computation.

[4]  Shang-Liang Chen,et al.  Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[5]  Jerry S. Brock.  A modified Baldwin-Lomax turbulence model for turbomachinery wakes , 1991 .

[6]  P. Spalart.  A One-Equation Turbulence Model for Aerodynamic Flows , 1992 .

[7]  T. Coakley,et al.  Turbulence Modeling Validation, Testing, and Development , 1997 .

[8]  T. Coakley,et al.  TURBULENCE MODELING VALIDATION , 1997 .

[9]  S. Pope.  Turbulent Flows: FUNDAMENTALS , 2000 .

[10]  Dingli Yu,et al.  Selecting radial basis function network centers with recursive orthogonal least squares training , 2000, IEEE Trans. Neural Networks Learn. Syst..

[11]  Friedhelm Schwenker,et al.  Three learning phases for radial-basis-function networks , 2001, Neural Networks.

[12]  Michele Milano,et al.  Neural network modeling for near wall turbulent flow , 2002 .

[13]  C. Shu,et al.  Local radial basis function-based differential quadrature method and its application to solve two-dimensional incompressible Navier–Stokes equations , 2003 .

[14]  Nikos A. Vlassis,et al.  The global k-means clustering algorithm , 2003, Pattern Recognit..

[15]  B. Eisfeld.  Implementation of Reynolds stress models into the DLR-FLOWer code , 2004 .

[16]  M.K. Muezzinoglu,et al.  Projection-based gradient descent training of radial basis function networks , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[17]  A Turbulent-Wake Estimation Using Radial Basis Function Neural Networks , 2005 .

[18]  Sundaram Suresh,et al.  A Fully Complex-Valued Radial Basis Function Network and its Learning Algorithm , 2009, Int. J. Neural Syst..

[19]  M. P. Sebastian,et al.  Improving the Accuracy and Efficiency of the k-means Clustering Algorithm , 2009 .

[20]  Anil K. Jain.  Data clustering: 50 years beyond K-means , 2010, Pattern Recognit. Lett..

[21]  Mostafa Y. El-Bakry.  Radial basis function neural network model for mean velocity and vorticity of capillary flow , 2011 .

[22]  Weiwei Zhang,et al.  Efficient Method for Limit Cycle Flutter Analysis Based on Nonlinear Aerodynamic Reduced-Order Models , 2012 .

[23]  P. A. Durbin,et al.  Transition modeling using data driven approaches , 2014 .

[24]  Haiyan Hu,et al.  Nonlinear Reduced-Order Modeling for Multiple-Input/Multiple-Output Aerodynamic Systems , 2014 .

[25]  William Gropp,et al.  CFD Vision 2030 Study: A Path to Revolutionary Computational Aerosciences , 2014 .

[26]  Julia Ling,et al.  Using Machine Learning to Understand and Mitigate Model Form Uncertainty in Turbulence Models , 2015, 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA).

[27]  Brendan D. Tracey,et al.  A Machine Learning Strategy to Assist Turbulence Model Development , 2015 .

[28]  J. Templeton.  Evaluation of machine learning algorithms for prediction of regions of high Reynolds averaged Navier Stokes uncertainty , 2015 .

[29]  Anand Pratap Singh,et al.  New Approaches in Turbulence and Transition Modeling Using Data-driven Techniques , 2015 .

[30]  Haiyan Hu,et al.  Open/Closed-Loop Aeroservoelastic Predictions via Nonlinear, Reduced-Order Aerodynamic Models , 2015 .

[31]  Heng Xiao,et al.  Physics-Informed Machine Learning for Predictive Turbulence Modeling: Towards a Complete Framework. , 2016 .

[32]  J. Templeton,et al.  Reynolds averaged turbulence modelling using deep neural networks with embedded invariance , 2016, Journal of Fluid Mechanics.

[33]  Heng Xiao,et al.  A Priori Assessment of Prediction Confidence for Data-Driven Turbulence Modeling , 2017, Flow, Turbulence and Combustion.

[34]  K. Duraisamy,et al.  Using field inversion to quantify functional errors in turbulence closures , 2016 .

[35]  Weiwei Zhang,et al.  A high-order finite volume method on unstructured grids using RBF reconstruction , 2016, Comput. Math. Appl..

[36]  Karthik Duraisamy,et al.  A paradigm for data-driven predictive modeling using field inversion and machine learning , 2016, J. Comput. Phys..

[37]  Julia Ling,et al.  Machine learning strategies for systems with invariance properties , 2016, J. Comput. Phys..

[38]  Yuji Hattori,et al.  Searching for turbulence models by artificial neural network , 2016, 1607.01042.

[39]  Weiwei Zhang,et al.  Nonlinear Aerodynamic Reduced-Order Model for Limit-Cycle Oscillation and Flutter , 2016 .

[40]  Karthik Duraisamy,et al.  Machine Learning-augmented Predictive Modeling of Turbulent Separated Flows over Airfoils , 2016, ArXiv.

[41]  Weiwei Zhang,et al.  Layered reduced-order models for nonlinear aerodynamics and aeroelasticity , 2017 .

[42]  Z. She,et al.  N ov 2 01 2 Lie-group derivation of a multi-layer mixing length formula for turbulent channel and pipe flows , 2012 .

[43]  J. Nathan Kutz,et al.  Deep learning in fluid dynamics , 2017, Journal of Fluid Mechanics.

[44]  P. Gualtieri,et al.  Turbulent separation in a channel with a curved wall , 2015, 1512.05527.

[45]  Jinlong Wu,et al.  Physics-informed machine learning approach for reconstructing Reynolds stress modeling discrepancies based on DNS data , 2016, 1606.07987.

[46]  Karthik Duraisamy,et al.  Status, Emerging Ideas and Future Directions of Turbulence Modeling Research in Aeronautics , 2017 .

[47]  Physics-Informed, Data-Driven Framework for Model-Form Uncertainty Estimation and Reduction in RANS Simulations , 2017 .

[48]  K. Duraisamy,et al.  Augmentation of turbulence models using field inversion and machine learning , 2017 .

[49]  Z. She,et al.  Quantifying wall turbulence via a symmetry approach. Part 2. Reynolds stresses , 2018, Journal of Fluid Mechanics.

[50]  Weiwei Zhang,et al.  Reduced-Order Modeling for Nonlinear Aeroelasticity with Varying Mach Numbers , 2018, Journal of Aerospace Engineering.

[51]  Di Peng,et al.  Data mining of a clean signal from highly noisy data based on compressed data fusion: A fast-responding pressure-sensitive paint application , 2018, Physics of Fluids.

[52]  A data assimilation model for turbulent flows using continuous adjoint formulation , 2018, Physics of Fluids.

[53]  P. Durbin.  Some Recent Developments in Turbulence Closure Modeling , 2018 .

[54]  Hui Li,et al.  Prediction model of velocity field around circular cylinder over various Reynolds numbers by fusion convolutional neural networks based on pressure on the cylinder , 2018 .

[55]  P. Gualtieri,et al.  Turbulence dynamics in separated flows: the generalised Kolmogorov equation for inhomogeneous anisotropic conditions , 2018, Journal of Fluid Mechanics.

[56]  Subhransu Maji,et al.  Bilinear Convolutional Neural Networks for Fine-Grained Visual Recognition , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[57]  Weiwei Zhang,et al.  A hybrid reduced-order framework for complex aeroelastic simulations , 2019, Aerospace Science and Technology.

[58]  Karthik Duraisamy,et al.  Turbulence Modeling in the Age of Data , 2018, Annual Review of Fluid Mechanics.