Abstract
In this paper, a novel kriging-based multi-fidelity method is proposed. Firstly, the model uncertainty of low-fidelity and high-fidelity models is quantified. On the other hand, the prediction uncertainty of kriging-based surrogate models(SM) is confirmed by its mean square error. After that, the integral uncertainty is acquired by math modeling. Meanwhile, the SMs are constructed through data from low-fidelity and high-fidelity models. Eventually, the low-fidelity (LF) and high-fidelity (HF) SMs with integral uncertainty are obtained and a proposed fusion algorithm is implemented. The fusion algorithm refers to the Kalman filter’s idea of optimal estimation to utilize the independent information from different models synthetically. Through several mathematical examples implemented, the fused SM is certified that its variance is decreased and the fused results tend to the true value. In addition, an engineering example about autonomous underwater vehicles’ hull design is provided to prove the feasibility of this proposed multi-fidelity method in practice. In the future, it will be a helpful tool to deal with reliability optimization of black-box problems and potentially applied in multidisciplinary design optimization.
Similar content being viewed by others
References
Allaire DL, Willcox KE, Toupet O (2010) A bayesian-based approach to multifidelity multidisciplinary design optimization. AIAA 2010–9183
Balabanov VO, Venter G (2004) Multi-fidelity optimization with high-fidelity analysis and low-fidelity gradients. AIAA 2004–4459
Box GE, Draper NR (1987a) Empirical model building and response surfaces. Wiley, New York
Box EP, Draper NR (1987b) Empirical model-building and response surfaces. Wiley, New York
Broomhead D, Loewe D (1988) Multivariate functional interpolation and adaptive networks. Complex Syst 2:321–355
Burnham K, Anderson D (2002) Model selection and multi-model inference: a practical guide information-theoretic approach. Springer, New York
Cadini F, Santos F, Zio E (2014) An improved adaptive kriging-based importance technique for sampling multiple failure regions of low probability. Reliab Eng Syst Saf 131:109–117
Chang KJ, Haftka RT, Giles GL et al (1993) Sensitivity-based scaling for approximating structural response. J Aircraft 30(2):283–288
Degroote J, Couckuyt I, Vierendeels J et al (2012) Inverse modelling of an aneurysm’s stiffness using surrogate-based optimization and fluid-structure interaction simulations. Struct Multidiscip Optim 46:457–469
Eves J, Toropov VV, Thompson HM et al (2012) Design optimization of supersonic jet pumps using high fidelity flow analysis. Struct Multidiscip Optim 45:739–745
Forrester AIJ, Keane AJ (2009) Recent advances in surrogate-based optimization. Prog Aerosp Sci 45:50–79
Forrester AIJ, Sóbester A, Keane AJ (2007) Multi-fidelity optimization via surrogate modeling. Proc Roy Soc A 463(2088):3251–3269
Forrester AIJ, Sóbester A, Keane AJ (2008) Engineering design via surrogate modeling — a practical guide. Wiley, New York
Foytik J, Sankaran P, Asari V (2011) Tracking and recognizing multiple faces using Kalman filtering and ModularPCA. Procedia Comput Sci 6:256–261
Han ZH, Zimmermann R, Görtz S (2010) A new cokriging method for variable-fidelity surrogate modeling of aerodynamic data. AIAA 2010–1225
Huang LK, Gao Z, Zhang D (2013) Research on multi-fidelity aerodynamic optimization methods. Chin J Aeronaut 26:279–286
Jones DR, Schonlau M, Welch WJ (1998) Efficient global optimization of expensive black-box functions. J Global Optim 13:455–492
Kalman RE (1960) A new approach to linear filtering and prediction problems. Trans ASME J Basic Eng 82:35–45
Kennedy M, O’Hagan A (2000) Predicting the output from a complex computer code when fast approximations are available. Biometrika 87(1):1–13
Kennedy M, O’Hagan A (2001) Bayesian calibration of computer models. J R Stat Soc B 63(3):425–464
Koziel S, Leifsson L (2013) Surrogate-based aerodynamic shape optimization by variable-resolution models. AIAA 51(1):94–105
Koziel S, Ogurtsov S (2013) Multi-objective design of antennas using variable-fidelity simulations and surrogate models. IEEE Trans Antennas Propag 61(12):5931–5939
Koziel S, Bandler JW, Madsen K (2006) A space-mapping framework for engineering optimization-theory and implementation. IEEE Trans Microw Theory 54(10):3721–3730
Li M, Li G, Azarm S (2008) A kriging metamodel assisted multi-objective genetic algorithm for design optimization. J Mech Design 130(3): 031 401-1–10
Link W, Barker R (2006) Model weights and the foundations of multimodel inference. Ecology 87(10):2626–2635
Morris MD, Mitchell TJ (1995) Exploratory designs for computational experiments. J Stat Plan Infer 43:381–402
Oberkampf WL, Roy CJ (2010) Verification and validation in scientific computing. Cambridge, UK
Paz J, Diaz J, Romera L et al (2014) Crushing analysis and multi-objective crashworthiness optimization of GFRP honeycomb-filled energy absorption devices. Finite Elem Anal Des 91:30–39
Queipo NV, Haftka RT, Shyy W et al (2005) Surrogate-based analysis and optimization. Prog Aerosp Sci 41:1–28
Reinert J, Apostolakis G (2006) Including model uncertainty in risk-informed decision making. Ann Nucl Energy 33(4):354–369
Robinson TD, Eldred MS, Willcox KE et al (2008) Surrogate-based optimization using multifidelity models with variable parameterization and corrected space mapping. AIAA 46(11):2814–2821
Simpson TW, Mauery TM, Korte JJ et al (2001) Kriging metamodels for global approximation in simulation-based multidisciplinary design optimization. AIAA 39(12):2233–2241
Sun G, Li G, Zhou S et al (2011) Multi-fidelity optimization for sheet forming process. Struct Multidiscip Optim 44:111–124
Tenne Y, Armfield SW (2009) A framework for memetic optimization using variable global and local surrogate models. Soft Comput 13(8–9):781–793
Wang GG, Shan S (2007) Review of metamodeling techniques in support of engineering design optimization. J Mech Des 129(4):370–380
Wankhede MJ, Bressloff NW, Keane AJ (2011) Combustor design optimization using co-kriging of steady and unsteady turbulent combustion. J Eng Gas Turbines Power 133:121504–121511
Xiong Y, Chen W, Tsui KL (2008) A new variable-fidelity optimization framework based on model fusion and objective-oriented sequential sampling. J Mech Des 130:111401–111409
Yao W, Chen X, Ouyang Q (2011) A surrogate based multistage-multilevel optimization procedure for multidisciplinary design optimization. Struct Multidiscip Optim 45:559–574
Yelten MB, Zhu T, Koziel S et al (2012) Demystifying surrogate modeling for circuits and systems. IEEE Circ Syst Mag 12(1):45–63
Yu K, Yang X, Yue Z (2011) Aerodynamic and heat transfer design optimization of internally cooling turbine blade based different surrogate models. Struct Multidiscip Optim 44:75–83
Zadeh PM, Toropov VV, Wood AS (2009) Metamodel-based collaborative optimization framework. Struct Multidiscip Optim 38:103–115
Zheng J, Shao X, Gao L et al (2013) A hybrid variable-fidelity global approximation modeling method combing tuned radial basis function base and kriging correction. J Eng Des 24(8):604–622
Zio E, Apostolakis G (1996) Two methods for the structured assessment of model uncertainty by experts in performance assessments of radioactive waste repositories. Reliab Eng Syst Saf 54(2–3):225–241
Acknowledgments
The author is grateful to the editor and the anonymous referees for their insightful and constructive comments and suggestions, which have been very helpful for improving this paper. This research was supported by the National Natural Science Foundation of China (Grant No. 51375389) and the National High Technology Research.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A
Consider the approximation of a function f(x). The design variable x is a vector with n dimensions. A stochastic process F(x) is defined to realize the deterministic response of f(x) as follows:
μ is defined as constant and Z(x) is defined as a stochastic process. Z(x) has the following stochastic behaviors:
σ 2 is the process variance for the response and R(θ, x, w) is the correlation model between any two points x and x '. Θ = {θ 1, θ 2, ⋯ θ n } is a set of parameters which determines the gradient of R(Θ, x, x '). This paper uses the Gaussian correlation function, which is defined as
Next, assume that there are N sample points x (1), x (2), ⋯, x (N) given by true function f(x). The Kriging model realizes all the given points as follows:
In Kriging method, these parameters μ, σ 2, Θ are obtained by maximum likelihood estimation (MLE) (Forrester and Keane 2009). Here the estimated values are given:
Where f = [f(x (1)), f(x (2)), ⋯, f(x (N)) ]T, R is a N × N correlations matrix whose element in the the j-th column of the i-th line is defined as R(Θ, x (i), x (j)).
Finally, minimize the mean squared error (MSE)
Meanwhile meet the following unbiasedness constraint:
The best linear unbiased predictor (BLUP) \( \widehat{f}\left(\boldsymbol{x}\right) \) results in the form as
Where r(x) is a N-dimensional vector. The i-th element of r(x) is R(Θ, x, x (i)) where x is the any location to be estimated. The final form of MSE is:
Appendix B
Six-hump camel-back function (SC)
Branin function (BR)
Booth function (BT)
Bohachevsky function (BC)
Himmelblau function (HM)
Rights and permissions
About this article
Cite this article
Dong, H., Song, B., Wang, P. et al. Multi-fidelity information fusion based on prediction of kriging. Struct Multidisc Optim 51, 1267–1280 (2015). https://doi.org/10.1007/s00158-014-1213-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00158-014-1213-9