Research Papers

Autonomous Uncertainty Quantification for Discontinuous Models Using Multivariate Padé Approximations

[+] Author and Article Information
Richard Ahlfeld

Uncertainty Quantification Lab,
Department of Aeronautics Imperial
College London,
London SW7 2AZ, UK
e-mail: r.ahlfeld14@imperial.ac.uk

Francesco Montomoli

Uncertainty Quantification Lab,
Department of Aeronautics Imperial
College London,
London SW7 2AZ, UK

Mauro Carnevale

Osney Thermo-Fluids Laboratory,
Department of Engineering Science,
University of Oxford,
Oxford OX2 0ES, UK

Simone Salvadori

Department of Industrial Engineering,
University of Florence,
Florence 50121, Italy

1Corresponding author.

Contributed by the International Gas Turbine Institute (IGTI) of ASME for publication in the JOURNAL OF TURBOMACHINERY. Manuscript received May 12, 2017; final manuscript received October 14, 2017; published online January 23, 2018. Editor: Kenneth Hall.

J. Turbomach 140(4), 041004 (Jan 23, 2018) (10 pages) Paper No: TURBO-17-1062; doi: 10.1115/1.4038826 History: Received May 12, 2017; Revised October 14, 2017

Problems in turbomachinery computational fluid dynamics (CFD) are often characterized by nonlinear and discontinuous responses. Ensuring the reliability of uncertainty quantification (UQ) codes in such conditions, in an autonomous way, is challenging. In this work, we suggest a new approach that combines three state-of-the-art methods: multivariate Padé approximations, optimal quadrature subsampling (OQS), and statistical learning. Its main component is the generalized least-squares multivariate Padé–Legendre (PL) approximation. PL approximations are globally fitted rational functions that can accurately describe discontinuous nonlinear behavior. They need fewer model evaluations than local or adaptive methods and do not cause the Gibbs phenomenon like continuous polynomial chaos methods. A series of modifications of the Padé algorithm allows us to apply it to arbitrary input points instead of optimal quadrature locations. This property is particularly useful for industrial applications, where a database of CFD runs is already available, but not in optimal parameter locations. One drawback of the PL approximation is that it is nontrivial to ensure reliability. To improve stability, we suggest to couple it with OQS. Our reasoning is that least-squares errors, caused by an ill-conditioned design matrix, are the main source of error. Finally, we use statistical learning methods to check smoothness and convergence. The resulting method is shown to efficiently and correctly fit thousands of partly discontinuous response surfaces for an industrial film cooling and shock interaction problem using only nine CFD simulations.

Copyright © 2018 by ASME
Your Session has timed out. Please sign back in to continue.


Carnevale, M. , Montomoli, F. , D'Ammaro, A. , Salvadori, S. , and Martelli, F. , 2013, “ Uncertainty Quantification: A Stochastic Method for Heat Transfer Prediction Using LES,” ASME J. Turbomach., 135(5), p. 051021. [CrossRef]
Wan, X. , and Karniadakis, G. E. , 2005, “ An Adaptive Multi-Element Generalized Polynomial Chaos Method for Stochastic Differential Equations,” J. Comput. Phys., 209(2), pp. 617–642. [CrossRef]
Lin, G. , and Tartakovsky, A. M. , 2007, “ An Efficient, High-Order Multi-Element Probabilistic Collocation Method on Sparse Grids for Three-Dimensional Flow in Random Porous Media,” American Geophysical Union, Washington, DC, AGU Fall Meeting Abstracts, Vol. 1, p. 1318. http://abstractsearch.agu.org/meetings/2007/FM/H23B-1318.html
Sargsyan, K. , Safta, C. , Debusschere, B. , and Najm, H. , 2012, “ Uncertainty Quantification Given Discontinuous Model Response and a Limited Number of Model Response Evaluations,” SIAM J. Sci. Comput., 34(1), pp. B44–B64. [CrossRef]
Lucor, D. , Witteveen, J. , Constantine, P. , Schiavazzi, D. , and Iaccarino, G. , 2012, “ Comparison of Adaptive Uncertainty Quantification Approaches for Shock Wave-Dominated Flows,” Summer Program Center for Turbulence Research, Stanford, CA, June 24–July 20, pp. 219–228. https://web.stanford.edu/group/ctr/Summer/SP12/03.04_lucor.pdf
Lucor, D. , Su, C.-H. , and Karniadakis, G. E. , 2004, “ Generalized Polynomial Chaos and Random Oscillators,” Int. J. Numer. Methods Eng., 60(3), pp. 571–596. [CrossRef]
Chantrasmi, T. , Doostan, A. , and Iaccarino, G. , 2009, “ Padé-Legendre Approximants for Uncertainty Analysis With Discontinuous Response Surfaces,” J. Comput. Phys., 228(19), pp. 7159–7180. [CrossRef]
Le Maitre, O. P. , Knio, O. M. , Najm, H. N. , and Ghanem, R. G. , 2004, “ Uncertainty Propagation Using Wiener-Haar Expansions,” J. Comput. Phys., 197(1), pp. 28–57. [CrossRef]
Witteveen, J. A. S. , and Iaccarino, G. , 2012, “ Simplex Stochastic Collocation With Random Sampling and Extrapolation for Nonhypercube Probability Spaces,” SIAM J. Sci. Comput., 34(2), pp. A814–A838. [CrossRef]
Edeling, W. N. , Dwight, R. P. , and Cinnella, P. , 2016, “ Simplex-Stochastic Collocation Method With Improved Scalability,” J. Comput. Phys., 310, pp. 301–328. [CrossRef]
Gorodetsky, A. , and Marzouk, Y. , 2014, “ Efficient Localization of Discontinuities in Complex Computational Simulations,” SIAM J. Sci. Comput., 36(6), pp. A2584–A2610. [CrossRef]
Sudret, B. , Marelli, S. , and Lataniotis, C. , 2015, “ Sparse Polynomial Chaos Expansions as a Machine Learning Regression Technique,” International Symposium on Big Data and Predictive Computational Modeling, Munich, Germany, May 18–21, pp. 1–37. https://www.research-collection.ethz.ch/bitstream/handle/20.500.11850/111718/eth-48586-01.pdf
Seshadri, P. , Narayan, A. , and Mahadevan, S. , 2016, “Optimal Quadrature Subsampling for Least Squares Polynomial Approximations,” e-print arXiv:1601.05470.
Montomoli, F. , Carnevale, M. , D'Ammaro, A. , Massini, M. , and Salvadori, S. , 2015, Uncertainty Quantification in Computational Fluid Dynamics and Aircraft Engines, Springer, Berlin. [CrossRef]
Carnevale, M. , D'Ammaro, A. , Montomoli, F. , and Salvadori, S. , 2014, “Film Cooling and Shock Interaction: An Uncertainty Quantification Analysis With Transonic Flows,” ASME Paper No. GT2014-25024.
Bungartz, H.-J. , and Griebel, M. , 2004, “ Sparse Grids,” Acta Numer., 13, pp. 147–269. [CrossRef]
Blatman, G. , and Sudret, B. , 2011, “ Adaptive Sparse Polynomial Chaos Expansion Based on Least Angle Regression,” J. Comput. Phys., 230(6), pp. 2345–2367. [CrossRef]
Donoho, D. L. , 2006, “ For Most Large Underdetermined Systems of Linear Equations the Minimal /1-Norm Solution is Also the Sparsest Solution,” Commun. Pure Appl. Math., 59(6), pp. 797–829. [CrossRef]
Chantrasmi, T. , Iaccarino, G. , Najm, H. N. , and Papanicolaou, G. , 2011, Padé-Legendre Method for Uncertainty Quantification With Fluid Dynamics Applications, Stanford University, Stanford, CA.
Palar, P. S. , Tsuchiya, T. , and Parks, G. T. , 2016, “ Multi-Fidelity Non-Intrusive Polynomial Chaos Based on Regression,” Comput. Methods Appl. Mech. Eng., 305, pp. 579–606. [CrossRef]
Hosder, S. , Walters, R. W. , and Balch, M. , 2007, “ Efficient Sampling for Non-Intrusive Polynomial Chaos Applications With Multiple Uncertain Input Variables,” AIAA Paper No. 2007-1939.
Loeven, A. , and Bijl, H. , 2009, “ An Efficient Framework for Uncertainty Quantification in CFD Using Probabilistic Collocation,” AIAA Paper No. 2009-2275.
Guillaume, P. , Huard, A. , and Robin, V. , 1998, “ Generalized Multivariate Padé Approximants,” J. Approximation Theory, 95(2), pp. 203–214. [CrossRef]
Salvadori, S. , Montomoli, F. , and Martelli, F. , 2013, “ Film-Cooling Performance in Supersonic Flows: Effect of Shock Impingement,” Proc. Inst. Mech. Eng., Part A, 227(3), pp. 295–305. [CrossRef]
Ochs, M. , Schulz, A. , and Bauer, H.-J. , 2007, “Investigation of the Influence of Trailing Edge Shock Waves on Film Cooling Performance of Gas Turbine Airfoils,” ASME Paper No. GT2007-27482.
Garg, V. K. , and Rigby, D. L. , 1999, “ Heat Transfer on a Film-Cooled Blade–Effect of Hole Physics,” Int. J. Heat Fluid Flow, 20(1), pp. 10–25. [CrossRef]
Acharya, S. , and Leedom, D. H. , 2013, “ Large Eddy Simulations of Discrete Hole Film Cooling With Plenum Inflow Orientation Effects,” ASME J. Heat Transfer, 135(1), p. 011010. [CrossRef]
Menter, F. R. , Langtry, R. , and Völker, S. , 2006, “ Transition Modelling for General Purpose CFD Codes,” Flow, Turbul. Combust., 77(1–4), pp. 277–303. [CrossRef]
Hagen, J. P. , and Kurosaka, M. , 1993, “ Corewise Cross-Flow Transport in Hairpin Vortices—The Tornado Effect,” Phys. Fluids A, 5(12), pp. 3167–3174. [CrossRef]
Zhou, C. , 2015, “ Thermal Performance of Transonic Cooled Tips in a Turbine Cascade,” J. Propul. Power, 31(5), pp. 1268–1280. [CrossRef]


Grahic Jump Location
Fig. 1

Response surfaces of hyperbolic tangent using a 16 points full tensor grid (a) and an 18 points optimally subsampled grid (b). The analytical function is shown in (c): (a) full tensor grid, (b) optimal quadrature subsampled, and (c) true function

Grahic Jump Location
Fig. 2

Demonstration of automatic parameter selection

Grahic Jump Location
Fig. 3

(a) Control volume and (b) flow features

Grahic Jump Location
Fig. 4

Grid details in (a) shock interaction region and (b) coolant region

Grahic Jump Location
Fig. 5

(a) Comparison between CFD (upper) and experimental data (lower) for the deterministic reference configuration case. (b) ηaw distribution over the centerline

Grahic Jump Location
Fig. 6

(a) Flow streamlines and interaction with the shock and (b) details of the tornado vortex after the shock

Grahic Jump Location
Fig. 7

Temperature maps and streamlines between hole and shock

Grahic Jump Location
Fig. 8

The graphs illustrates how the shock moves through the stochastic response surface of the thermal efficiency

Grahic Jump Location
Fig. 9

Example of choice of the automatic response surface error and smoothness control

Grahic Jump Location
Fig. 10

Three examples of automatically fitted Padé response surfaces before (1), at (2) and after (3) the shock location. The bottom row shows three PDFs to the x-locations before, at and after the shock.

Grahic Jump Location
Fig. 11

(a) Mean and standard deviation of the film cooling effectiveness and (b) sensitivity analysis of input parameters using sobol indices



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In