Abstract
This paper deals with the advantages of using variable-fidelity metamodeling strategies in order to develop a valid metamodel more rapidly than by using traditional methods. In our mechanical assembly design, we use the term “variable-fidelity” in reference to the convergence (or accuracy) level of the iterative solver being used. Variable-fidelity metamodeling is a way to improve the prediction of the output of a complex system by incorporating rapidly available auxiliary lower-fidelity data. This work uses two fidelity levels, but more levels can be added. The LATIN iterative algorithm is used along with a “multiparametric” strategy to calculate the various data and their different fidelity levels by means of an error indicator. Three main categories of variable-fidelity strategies are currently available. We tested at least one method from each of these categories, which comes to a total of five methods for calculating a valid metamodel using low- and high-fidelity data. Here, our objective is to compare the performances of these five methods in solving three mechanical examples.
Similar content being viewed by others
Abbreviations
- \({{\mathcal {D}}}\) :
-
The design space
- n :
-
The dimension of \({{\mathcal {D}}}\)
- \(\mathbf{x } ^{i}\) :
-
A point \(\in {{\mathcal {D}}}\)
- \(y(\mathbf{x } )\) :
-
The objective function at point \(\mathbf{x } \)
- \(\hat{y}(\mathbf{x } )\) :
-
An approximation of the objective function at point \(\mathbf{x } \) obtained through a metamodel
- \(\mathbf X \) :
-
The matrix of all the calculated points \(\mathbf X =\left[ \begin{array}{ccc} {\mathbf{x } }^{1}&\cdots&{\mathbf{x } }^{p} \end{array} \right] ^{T} \)
- \(y^{i}\) :
-
The objective function at point \(\mathbf{x } ^{i}\)
- \(\mathbf Y \) :
-
The vector of all the objective function \(\mathbf Y =\left[ \begin{array}{ccc} y^{1}&\cdots&y^{p} \end{array} \right] ^{T} \)
- \(\bullet _{fcv}\) :
-
Any of the above in the case of fully converged data
- \(\bullet _{pcv}\) :
-
Any of the above in the case of partially converged data
- \(\mathbf{f } (\mathbf{x } )\) :
-
The regression function. \(\Big (\)If one chooses a mean regression, \(\mathbf{f } (\mathbf{x } )=1\); for a linear regression, \(\mathbf{f } (\mathbf{x } )=\left[ \begin{array}{cccc} 1&x_{1}&\cdots&x_{n} \end{array} \right] ^{T}\Big )\)
- \(\varvec{\beta }\) :
-
The parameters of the regression function. \(\Big (\)For a linear regression, \(\varvec{\beta }=\left[ \begin{array}{ccc} \beta _{1}&\cdots&\beta _{n+1} \end{array} \right] ^{T} \Big )\)
- \(\mathbf F = \left[ \begin{array}{ccc}{\mathbf{f } }({\mathbf{x } }^{1})&\cdots&{\mathbf{f } }({\mathbf{x } }^{p})\end{array}\right] ^{T}\) :
-
The matrix of the regression functions at the calculation points (for universal kriging case)
- \(z(\mathbf{x } )\) :
-
A stochastic function
- \(\mathbf Z \) :
-
The stochastic vector of the calculated approximations \(\mathbf Z =\left[ \begin{array}{ccc} z(\mathbf{x } ^{1})&\cdots&z(\mathbf{x } ^{p}) \end{array} \right] ^{T} \)
- \(\mathbf{C } \) :
-
The covariance matrix
- \(\mathbf R \) :
-
The correlation matrix
References
Jones, D.R.: A taxonomy of global optimization methods based on response surface. J. Glob. Optim. 21, 345–383 (2001)
Robinson, T.D., Eldred, M.R., Willcox, K.E., Haimes, R.: Surrogate-based optimization using multifidelity models with variable parametrization and corrected space mapping. AIAA J. 46, 2814–2822 (2008)
Courrier, N., Boucard, P.-A., Soulier, B.: The use of partially converged simulations in building surrogate models. Adv. Eng. Softw. 67, 186–197 (2014)
Chinesta, F., Ammar, A., Cueto, E.: Recent advances and new challenges in the use of the proper generalized decomposition for solving multidimensional models. Arch. Comput. Methods Eng. 17, 327–350 (2010)
Chinesta, F., Ladeveze, P., Cueto, E.: A short review on model order reduction based on proper generalized decomposition. Arch. Comput. Methods Eng. 18(4), 395–404 (2011)
Lamari, H., Chinesta, F., Ammar, A., Cuesto, E.: Nonconventional numerical strategies in the advanced simulation of materials and processes. Int. J. Mod. Manuf. Technol. 1, 49–56 (2009)
Pruliere, E., Chinesta, F., Ammar, A.: On the deterministic solution of multidimensional parametric models using the proper generalized decomposition. Math. Comput. Simul. 81, 791–810 (2010)
Heyberger, C., Boucard, P.-A., Neron, D.: Multiparametric analysis within the proper generalized decomposition framework. Comput. Mech. 49, 277–289 (2012)
Gano, S.E., Renaud, J.E., Martin, J.D., Simpson, T.W.: Update strategies for kriging models used in variable fidelity optimization. Struct. Multidiscipl. Optim. 32(4), 287–298 (2006)
Vitali, R., Haftka, R.T., Sankar, B.V.: Multi-fidelity design of stiffened composite panel with a crack. Struct. Multidiscipl. Optim. 23(5), 347–356 (2002)
Knill, D.L., Giunta, A.A., Baker, C.A., Grossman, B., Mason, W.H., Haftka, R.T., Watson, L.T.: Response surface model combining linear and euler aerodynamics for super sonic transport design. J. Aircr. 36(1), 75–86 (1999)
Forrester, A.I.J., Bressloff, N.W., Keane, A.J.: Optimization using surrogate models and partially converged computational fluid dynamics simulations. Proc. R. Soc. A 462, 2177–2204 (2006)
Dadone, A., Grossman, B.: Progressive optimization of inverse fluid dynamic design problems. Comput. Fluids 29, 1–32 (2000)
Jones, D.R., Schonlau, M., Welch, W.J.: Efficient global optimization of expensive black-box functions. J. Glob. Optim. 13, 455–492 (1998)
Krige, D.G.: A statistical approach to some mine valuation and allied problems on the witwatersrand. Master’s thesis, University of the Witwatersrand (1951)
Matheron, G.: Principles of geostatics. Econ. Geol. 58(8), 1246–1266 (1963)
Laurent, L., Boucard, P.-A., Soulier, B.: A dedicated multiparametric strategy for the fast construction of a cokriging metamodel. Comput. Struct. 124, 61–73 (2013)
Morris, M.D., Mitchell, T.J., Ylvisaker, D.: Bayesian desing and analysis of computer experiments: use of derivatives in surface prediction. Technometrics 35(3), 243–255 (1993)
Koehler, J., Owen, A.: Computer experiments. Handb. Stat. 13, 261–308 (1996)
Praveen, C., Duvigneau, R.: Radial basis function and kriging metamodels for aerodynamic optimization. Technical report 6151 INRIA (2007)
Buhmann, M.D.: Radial basis functions. Acta Numer. 9, 1–38 (2000)
Zimmermann, R., Han, Z.-H.: Simplified cross-correlation estimation for multi-fidelity surrogate cokriging models. Adv. Appl. Math. Sci. 7(2), 181–201 (2010)
Han, Z.-H., Zimmermann, R., Görtz, S.: Alternative cokriging model for variable-fidelity surrogate modeling. AIAA J. 50(5), 1205–1210 (2012)
Forrester, A.I.J., Keane, A.J.: Recent advances in surrogate-based optimization. Prog. Aerosp. Sci. 45, 50–79 (2009)
Rosenbaum, B., Schulz, V.: Efficient response surface method based on generic surrogate models. SIAM J. Sci. Comput. 35(2), B529–B550 (2013)
Han, Z.-H., Görtz, S., Zimmermann, R.: Improving variable-fidelity surrogate modeling via gradient-enhanced kriging and a generalized hybrid bridge function. Aerosp. Sci. Technol. 25(1), 177–189 (2013)
Forrester, A.I.J., Sobester, A., Keane, A.J.: Multi-fidelity optimization via surrogate modelling. Proc. R. Soc. A 463, 3251–3269 (2007)
Yamazaki, W., Mavriplis, D.J.: Derivative-enhanced variable fidelity surrogate modeling for aerodynamic functions. AIAA J. 51(1), 126–137 (2013)
Rosenbaum, B.: Efficient global surrogate models for responses of expensive simulations. PhD thesis, Universität Trier (2013)
Han, Z.-H., Görtz, S.: A hierarchical kriging model for variable-fidelity surrogate modeling of aerodynamic functions. AIAA J. 50(9), 1885–1896 (2012)
Ladevèze, Pierre: Nonlinear computational structural mechanics: new approaches and non-incremental methods of calculation. Mechanical Engineering Series. Springer, Berlin (1999)
Wachernagel, P.: Multivariate geostatics: an introduction with applications. Springer, Berlin (2003)
Soulier, B., Boucard, P.-A.: A multiparametric strategy for the two step optimization of structural assemblies. Struct. Multidiscipl. Optim. 47(4), 539–553 (2013)
Boucard, P.-A., Ladevèze, P.: A multiple solution method for non-linear structural mechanics. Mech. Eng. 50(5), 317–328 (1999)
Soulier, B., Boucard, P.-A.: A multiparametric strategy for the large-scale multilevel optimization of structural assemblies. In: 8th World Congress on Structural and Multidisciplinary Optimization, Lisbon, Portugal (2009)
Sacks, J., Welch, W.J., Mitchell, T.J., Wynn, H.P.: Design and analysis of computer experiments. Stat. Sci. 4, 409–423 (1989)
Han, Z.-H., Zimmermann, R., Görtz, S.: A new cokriging method for variable-fidelity surrogate modeling of aerodynamic data. In: 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, Orlando, FL (2010). AIAA 2010-1225
Kennedy, M.C., O’Hagan, A.: Predicting the output from a complex computer code when fast approximations are available. Biometrika 87(1), 1–13 (2000)
Han, Z.-H., Görtz, S., Hain, R.: A variable-fidelity modeling method for aero-loads prediction. New Results Numer. Exp. Fluid Mech. VII 112, 17–25 (2010)
McKay, M.D., Beckman, R.J., Conover, W.J.: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code. Technometrics 21(2), 239–245 (1979)
Morris, M.D., Mitchell, T.J.: Exploratory designs for computational experiments. J. Stat. Plan. Interf. 43(3), 381–402 (1995)
Lawrence, I., Lin, K.: A concordance correlation coefficient to evaluate reproducibility. Biometrics 45, 255–268 (1989)
Edwards, A.L.: An introduction to linear regression and correlation. W.H.Freeman, New York (1976)
Forrester, A.I.J.: Efficient global aerodynamic optimisation using expensive computational fluid dynamics simulations. PhD thesis, University of Southampton, School of Engineering Sciences, Southampton, England (2004)
Huang, D., Allen, T.T., Notz, W.I., Miller, R.A.: Sequential kriging optimization using multiple-fidelity evaluations. Struct. Multidiscipl. Optim. 32(5), 369–382 (2006)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Courrier, N., Boucard, PA. & Soulier, B. Variable-fidelity modeling of structural analysis of assemblies. J Glob Optim 64, 577–613 (2016). https://doi.org/10.1007/s10898-015-0345-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10898-015-0345-9