This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Adjoint-Based Model Tuning and Machine Learning Strategy for Turbulence Model Improvement
Technical Paper
2022-01-0899
ISSN: 0148-7191, e-ISSN: 2688-3627
Annotation ability available
Sector:
Language:
English
Abstract
As turbulence modeling has become an indispensable approach to perform flow simulation in a wide range of industrial applications, how to enhance the prediction accuracy has gained increasing attention during the past years. Of all the turbulence models, RANS is the most common choice for many OEMs due to its short turn-around time and strong robustness. However, the default setting of RANS is usually benchmarked through classical and well-studied engineering examples, not always suitable for resolving complex flows in specific circumstances.
Many previous researches have suggested a small tuning in turbulence model coefficients could achieve higher accuracy on a variety of flow scenarios. Instead of adjusting parameters by trial and error from experience, this paper introduced a new data-driven method of turbulence model recalibration using adjoint solver, based on Generalized k-ω (GEKO) model, one variant of RANS. In this algorithm, the model coefficients are tuned locally to match the measured drag force and surface pressure of DrivAer generic cars as closely as possible. After well-tuned process, neural network strategy was utilized with a training dataset to learn the correlation between calibrated coefficients and certain flow features. Once trained, the improved model can be employed in similar cases, like geometrical modifications and varying working conditions. Results show that through tuning and machine learning, prediction of drag and surface pressure presented better agreement with test data, while with only CD as the target, the trained model application scope is quite limited, thus, a more comprehensive input of accessible data is needed to obtain a more generalized model.
Adjoint-based GEKO tuning, validated by the cases of this paper and correlated with wind tunnel test, proved to be very helpful in turbulence model optimization. Hence, this study indicated the promising potential of turbulence model tuning and machine learning in vehicle aerodynamic application, and at the same time, discussed the related issues encountered during this work.
Recommended Content
Authors
Citation
Wu, H., Zhou, H., Xu, S., Ren, C. et al., "Adjoint-Based Model Tuning and Machine Learning Strategy for Turbulence Model Improvement," SAE Technical Paper 2022-01-0899, 2022, https://doi.org/10.4271/2022-01-0899.Also In
References
- Makowski , F. and Kim , S. Advances in External-Aero Simulation of Ground Vehicles Using the Steady RANS Equations SAE Technical Paper 2000-01-0484 2000 https://doi.org/10.4271/2000-01-0484
- Kok , C. Resolving the Dependence on Freestream Values for the k-Turbulence Model AIAA Journal 38 7 2000 1292 1295
- Catalano , P. and Amato , M. An Evaluation of RANS Turbulence Modelling for Aerodynamic Applications Aerospace science and Technology 7 7 2003 493 509
- Corson , D. , Jaiman , R. , and Shakib , F. Industrial Application of RANS Modelling: Capabilities and Needs International journal of Computational Fluid dynamics 23 4 2009 337 347
- Rodi , W. Comparison of LES and RANS Calculations of the Flow around Bluff Bodies Journal of Wind Engineering and Industrial Aerodynamics 69 1997 55 75
- Strelets , M. Detached Eddy Simulation of Massively Separated Flows 39th Aerospace Sciences Meeting and Exhibit 2001
- Ekman , P. , Wieser , D. , Virdung , T. , and Karlsson , M. Assessment of Hybrid RANS-LES Methods for Accurate Automotive Aerodynamic Simulations Journal of Wind Engineering and Industrial Aerodynamics 206 2020 104301
- Oliver , A. and Moser , D. Bayesian Uncertainty Quantification Applied to RANS Turbulence Models Journal of Physics: Conference Series 318 4 2011 042032
- Parente , A. , Gorlé , C. , Van , J. et al. Improved k-ε Model and Wall Function Formulation for the RANS Simulation of ABL Flows Journal of Wind Engineering and Industrial Aerodynamics 99 4 2011 267 278
- Fu , C. , Uddin , M. , Robinson , C. et al. Turbulence Models and Model Closure Coefficients Sensitivity of NASCAR Racecar RANS CFD Aerodynamic Predictions SAE International Journal of Passenger Cars-Mechanical Systems 10 1 2017 330 345
- Zhao , Y. , Akolekar , D. , Weatheritt , J. et al. RANS Turbulence Model Development Using CFD-Driven Machine Learning Journal of Computational Physics 411 2020 109413
- Margheri , L. , Meldi , M. , Salvetti , V. et al. Epistemic Uncertainties in RANS Model Free Coefficients Computers & Fluids 102 2014 315 335
- Jung , K. , Chang , K. , and Bae , H. Uncertainty Quantification of GEKO Model Coefficients on Compressible Flows International Journal of Aerospace Engineering 2021
- Singh , P. , Medida , S. , and Duraisamy , K. Machine-Learning-Augmented Predictive Modeling of Turbulent Separated Flows over Airfoils AIAA Journal 55 7 2017 2215 2227
- Tracey , B. , Duraisamy , K. , and Alonso , J. A Machine Learning Strategy to Assist Turbulence Model Development AIAA, SciTech Forum 2015 Florida
- Kochkov , D. , Smith , A. , Alieva , A. et al. Machine Learning-Accelerated Computational Fluid Dynamics Proceedings of the National Academy of Sciences 118 21 2021
- Singh , P. and Duraisamy , K. Using Field Inversion to Quantify Functional Errors in Turbulence Closures Physics of Fluids 28 4 2016 045110
- Menter , R. , Matyushenko , A. and Lechner , R. Development of a Generalized k-ω Two-Equation Turbulence Model Numerical Fluid Mechanics and Multidisciplinary Design 142 2018
- Brezgin , V. and Aronson , E. Nozzle Design Influence on the Steam-Driven Ejector Journal of Physics Conference Series 1683 2020 042013
- Zhang , J. and Duraisamy , K. Machine Learning Methods for Data-Driven Turbulence Modeling 22nd AIAA Computational Fluid Dynamics Conference 2015 2460
- Duraisamy , K. , Zhang , J. , and Singh , P. New Approaches in Turbulence and Transition Modeling Using Data-Driven Techniques 53rd AIAA Aerospace Sciences Meeting 2015 1284
- Parish , J. , and Duraisamy , K. A Paradigm for Data-Driven Predictive Modeling Using Field Inversion and Machine Learning Journal of Computational Physics 305 2016 758 774
- Brunton , L. , Noack , R. , and Koumoutsakos , P. Machine Learning for Fluid Mechanics Annual Review of Fluid Mechanics 2020
- Meijering , E. A Chronology of Interpolation: from Ancient Astronomy to Modern Signal and Image Processing Proceedings of the IEEE 90 3 2002 319 342
- Ling , J. , Kurzawski , A. , and Templeton , J. Reynolds Averaged Turbulence Modeling Using Deep Neural Networks with Embedded Invariance Journal of Fluid Mechanics 807 2016 155 166
- Holland , R. , Baeder , D. and Duraisamy , K. Field Inversion and Machine Learning with Embedded Neural Networks: Physics-Consistent Neural Network Training AIAA Aviation 2019 Forum 2019 3200
- Sekar , V. , Zhang , M. , Shu , C. et al. Inverse Design of Airfoil Using a Deep Convolutional Neural Network AIAA Journal 57 3 2019 993 1003
- Song , X. , Zhang , Z. , Wang , Y. , Ye , S. , et al. Reconstruction of RANS Model and Cross-Validation of Flow Field based on Tensor Basis Neural Network Fluids Engineering Division Summer Meeting, American Society of Mechanical Engineers 2019 59032: V002T02A081
- Wu , J. , Li , J. , Qiu , X. et al. A Comparative Analysis of Multi-Machine Learning Algorithms for Data-Driven RANS Turbulence Modelling Journal of Physics: Conference Series. IOP Publishing 1684 1 2020 012043
- Hornik , K. , Stinchcombe , M. , and White , H. Multilayer Feedforward Networks are Universal Approximators Neural Networks 2 5 1989 359 366
- Bolcskei , H. , Grohs , P. , Kutyniok , G. et al. Optimal Approximation with Sparsely Connected Deep Neural Networks SIAM Journal on Mathematics of Data Science 1 1 2019 8 45
- Lawrence , N. and Jordan , M. Semi-Supervised Learning via Gaussian Processes Advances in Neural Information Processing Proceedings 2004
- de Sousa , C.A.R. An Overview on the Gaussian Fields and Harmonic Functions Method for Semi-Supervised Learning 2015 International Joint Conference on Neural Networks (IJCNN) 2015
- Izmailov , P. , Kirichenko , P. , and Finzi , M. Semi-Supervised Learning with Normalizing Flows Andrew Gordon Wilson Proceedings of the 37th International Conference on Machine Learning, PMLR 119 4615 4630 2020
- Pise , N. , and Kulkarni , P. A Survey of Semi-Supervised Learning Methods 2008 International Conference on Computational Intelligence and Security 2008
- Heft , I. , Indinger , T. , and Adams , A. Introduction of A New Realistic Generic Car Model for Aerodynamic Investigations SAE Technical Paper 2012-01-0168 2012 https://doi.org/10.4271/2012-01-0168
- Ahmed , S. , Ramm , G. , and Faltin , G. Some Salient Features of the Time-Averaged Ground Vehicle Wake SAE Technical Paper 840300 1984 https://doi.org/10.4271/840300
- Cooper , R. Closed-Test-Section Wind Tunnel Blockage Corrections for Road Vehicles SAE SP-1176 Warrendale 1996
- Windsor , S. IMechE Autotech UK 1991
- Hoffman , J. , Martindale , B. , Arnette , S. , Williams , J. et al. Effect of Test Section Configuration on Aerodynamic Drag Measurements SAE Technical Paper 2001-01-0631 2001 https://doi.org/10.4271/2001-01-0631
- Wieser , D. , Schmidt , H. , Müller , S. , Strangfeld , C. et al. Experimental Comparison of the Aerodynamic Behavior of Fastback and Notchback DrivAer Models SAE Int. J. Passeng. Cars - Mech. Syst. 7 2 2014 682 691 https://doi.org/10.4271/2014-01-0613
- Ashton , N. and Revell , A. Comparison of RANS and DES Methods for the DrivAer Automotive Body SAE Technical Paper 2015-01-1538 2015 https://doi.org/10.4271/2015-01-1538
- Strangfeld , C. , Wieser , D. , Schmidt , J. , Woszidlo , R. et al. Experimental Study of Baseline Flow Characteristics for the Realistic Car Model DrivAer SAE Technical Paper 2013-01-1251 2013 https://doi.org/10.4271/2013-01-1251
- Collin , C. , Mack , S. , Indinger , T. , and Mueller , J. A Numerical and Experimental Evaluation of Open Jet Wind Tunnel Interferences using the DrivAer Reference Model SAE Int. J. Passeng. Cars - Mech. Syst. 9 2 2016 657 679 https://doi.org/10.4271/2016-01-1597
- John , M. , Buga , D. , and Monti , I. Experimental and Numerical Study of the DrivAer Model Aerodynamics SAE Technical Paper 2018-01-0741 2018 https://doi.org/10.4271/2018-01-0741
- Zhou , H. , Ren , C. , Wu , H. , Chen , Q. et al. Multi-Objective Aerodynamic Optimization of Vehicle Shape Using Adjoint Approach Based on Steady-State and Transient Flow Solutions SAE Technical Paper 2021-01-0945 2021 https://doi.org/10.4271/2021-01-095