Skip to main content
main-content

Tipp

Weitere Kapitel dieses Buchs durch Wischen aufrufen

2021 | OriginalPaper | Buchkapitel

Application of Multitask Learning for Enhancement of Spalart-Allmaras Turbulence Model

verfasst von: Rohit Pochampalli, Emre Özkaya, Beckett Y. Zhou, Guillermo Suarez, Nicolas R. Gauger

Erschienen in: New Results in Numerical and Experimental Fluid Mechanics XIII

Verlag: Springer International Publishing

share
TEILEN

Abstract

Correcting functional terms using data driven techniques is one recent approach to improve the predictive accuracy of turbulence models. We present a multitask learning framework that can be used to generalize functional corrections of turbulence models to different flow conditions and test cases. The approach is developed with a view to generalize the correction of turbulence models using data from different flow conditions. The machine learning model first learns a compressed latent representation of flow variables that are most correlated with the correction term and has enhanced predictive accuracy for new test cases. The approach is shown to improve the performance of Spalart-Allmaras turbulence model in massively separated flow regimes.
Literatur
1.
Zurück zum Zitat Spalart, P.R.: Comments on the feasibility of LES for wings, and on a hybrid RANS/LES approach. In: Proceedings of first AFOSR international conference on DNS/LES. Greyden Press (1997) Spalart, P.R.: Comments on the feasibility of LES for wings, and on a hybrid RANS/LES approach. In: Proceedings of first AFOSR international conference on DNS/LES. Greyden Press (1997)
2.
Zurück zum Zitat Spalart, P.R., Deck, S., Shur, M.L., Squires, K.D., Strelets, M.K., Travin, A.: A new version of detached-eddy simulation, resistant to ambiguous grid densities. Theoret. Comput. Fluid Dyn. 20(3), 181 (2006) Spalart, P.R., Deck, S., Shur, M.L., Squires, K.D., Strelets, M.K., Travin, A.: A new version of detached-eddy simulation, resistant to ambiguous grid densities. Theoret. Comput. Fluid Dyn. 20(3), 181 (2006)
3.
Zurück zum Zitat Shur, M., Spalart, P.R., Strelets, M., Travin, A.: Detached-eddy simulation of an airfoil at high angle of attack. In: Engineering Turbulence Modelling and Experiments, vol. 4, pp. 669–678. Elsevier (1999) Shur, M., Spalart, P.R., Strelets, M., Travin, A.: Detached-eddy simulation of an airfoil at high angle of attack. In: Engineering Turbulence Modelling and Experiments, vol. 4, pp. 669–678. Elsevier (1999)
4.
Zurück zum Zitat Spalart, P., Allmaras, S.: A one-equation turbulence model for aerodynamic flows. In: 30th Aerospace Sciences Meeting and Exhibit, p. 439 (1992) Spalart, P., Allmaras, S.: A one-equation turbulence model for aerodynamic flows. In: 30th Aerospace Sciences Meeting and Exhibit, p. 439 (1992)
5.
Zurück zum Zitat Tracey, B.D., Duraisamy, K., Alonso, J.J.: A machine learning strategy to assist turbulence model development. In: 53rd AIAA Aerospace Sciences Meeting, p. 1287 (2015) Tracey, B.D., Duraisamy, K., Alonso, J.J.: A machine learning strategy to assist turbulence model development. In: 53rd AIAA Aerospace Sciences Meeting, p. 1287 (2015)
6.
Zurück zum Zitat Duraisamy, K., Zhang, Z.J., Singh, A.P.: New approaches in turbulence and transition modeling using data-driven techniques. In: 53rd AIAA Aerospace Sciences Meeting, p. 1284 (2015) Duraisamy, K., Zhang, Z.J., Singh, A.P.: New approaches in turbulence and transition modeling using data-driven techniques. In: 53rd AIAA Aerospace Sciences Meeting, p. 1284 (2015)
7.
Zurück zum Zitat Ling, J., Kurzawski, A., Templeton, J.: Reynolds averaged turbulence modelling using deep neural networks with embedded invariance. J. Fluid Mech. 807, 155–166 (2016) MathSciNetCrossRef Ling, J., Kurzawski, A., Templeton, J.: Reynolds averaged turbulence modelling using deep neural networks with embedded invariance. J. Fluid Mech. 807, 155–166 (2016) MathSciNetCrossRef
8.
Zurück zum Zitat Sotgiu, C., Weigand, B., Semmler, K., Wellinger, P.: Towards a general data-driven explicit algebraic reynolds stress prediction framework. Int. J. Heat Fluid Flow 79 (2019) Sotgiu, C., Weigand, B., Semmler, K., Wellinger, P.: Towards a general data-driven explicit algebraic reynolds stress prediction framework. Int. J. Heat Fluid Flow 79 (2019)
10.
Zurück zum Zitat Nocedal, J., Wright, S.: Numerical Optimization. Springer (2006) Nocedal, J., Wright, S.: Numerical Optimization. Springer (2006)
11.
Zurück zum Zitat Albring, T., Sagebaum, M., Gauger, N.R.: A consistent and robust discrete adjoint solver for the SU2 framework - validation and application. Notes on Numerical Fluid Mechanics and Multidisciplinary Design, vol. 132, pp. 77–86 (2016) Albring, T., Sagebaum, M., Gauger, N.R.: A consistent and robust discrete adjoint solver for the SU2 framework - validation and application. Notes on Numerical Fluid Mechanics and Multidisciplinary Design, vol. 132, pp. 77–86 (2016)
12.
Zurück zum Zitat Ladson, C.L.: Effects of independent variation of Mach and Reynolds numbers on the low-speed aerodynamic characteristics of the NACA 0012 airfoil section, volume 4074. National Aeronautics and Space Administration, Scientific and Technical \(\ldots \) (1988) Ladson, C.L.: Effects of independent variation of Mach and Reynolds numbers on the low-speed aerodynamic characteristics of the NACA 0012 airfoil section, volume 4074. National Aeronautics and Space Administration, Scientific and Technical \(\ldots \) (1988)
13.
Zurück zum Zitat Sheldahl, R.E., Klimas, P.C.: Aerodynamic characteristics of seven symmetrical airfoil sections through 180-degree angle of attack for use in aerodynamic analysis of vertical axis wind turbines. Technical report, Sandia National Labs., Albuquerque, NM (USA) (1981) Sheldahl, R.E., Klimas, P.C.: Aerodynamic characteristics of seven symmetrical airfoil sections through 180-degree angle of attack for use in aerodynamic analysis of vertical axis wind turbines. Technical report, Sandia National Labs., Albuquerque, NM (USA) (1981)
14.
Zurück zum Zitat White, H., Domowitz, I.: Nonlinear regression with dependent observations. Econometrica J. Econ. Soc. 143–161 (1984) White, H., Domowitz, I.: Nonlinear regression with dependent observations. Econometrica J. Econ. Soc. 143–161 (1984)
15.
Zurück zum Zitat Kramer, M.A.: Nonlinear principal component analysis using autoassociative neural networks. AIChE J. 37(2), 233–243 (1991) Kramer, M.A.: Nonlinear principal component analysis using autoassociative neural networks. AIChE J. 37(2), 233–243 (1991)
17.
Zurück zum Zitat Paszke, A., et al.: Pytorch: an imperative style, high-performance deep learning library. In: Advances in Neural Information Processing Systems, pp. 8024–8035 (2019) Paszke, A., et al.: Pytorch: an imperative style, high-performance deep learning library. In: Advances in Neural Information Processing Systems, pp. 8024–8035 (2019)
Metadaten
Titel
Application of Multitask Learning for Enhancement of Spalart-Allmaras Turbulence Model
verfasst von
Rohit Pochampalli
Emre Özkaya
Beckett Y. Zhou
Guillermo Suarez
Nicolas R. Gauger
Copyright-Jahr
2021
DOI
https://doi.org/10.1007/978-3-030-79561-0_8

Premium Partner

    Marktübersichten

    Die im Laufe eines Jahres in der „adhäsion“ veröffentlichten Marktübersichten helfen Anwendern verschiedenster Branchen, sich einen gezielten Überblick über Lieferantenangebote zu verschaffen.