, Le groupe avec une portée moyenne est composée de x 4 , x 5 , x 6 , x 12 , x 13 et x 14 . Le dernier groupe avec une valeur de portée élevée contient les variables x 2 , x 3 , x 10 et x 11, L'algorithme a détecté quatre groupes, les variables avec des portées faibles sont réparties en deux groupes x 1 puis x 7, vol.15

, Les conclusions et perspectives des trois chapitres pré-cédents qui correspondent aux propositions originales de ce manuscrit sont données ici. formation d'un front des solutions robustes pour le rendement. Seuls les trois calages intermédiaires subissent des perturbations. Les méthodes ont donné des fronts de Pareto proches. De plus, la méthode MyqEI ressort en donnant le

, Une autre solution est d'utiliser un modèle d'agrégation de code pour conduire l'optimisation robuste en utilisant les informations données par les deux simulateurs. En effet, les dé-rivées du code 1D sont observables et sont de bonnes approximations des dérivées du code 3D. Il faudrait donc créer un modèle capable de prédire le code 3D en se servant de l, l'optimisation robuste a uniquement été conduite sur le code 1D

, De plus, l'optimisation robuste à uniquement été menée sur le rendement, l'idée serait de procéder à une optimisation multi-objectif robuste sur la pression et le couple. Dans ce cas, il faudrait utiliser l'algorithme NSGA III (cf [Mkaouer et al., 2014]) qui permet de traiter plus d'objectifs. D'autre part, le noyau isotrope par groupe présenté dans la section "groupe de portée" pourrait être utilisé pour conduire une optimisation robuste avec beaucoup de variables d'entrées

. Dans-le, Il faudrait introduire un effet de pépite dans le modèle de co-krigeage avec dérivées pour prendre en compte cette erreur. Aussi, les codes industriels prennent en entrée des variables géométriques ainsi que le débit. Cette dernière variable est différente des autres car les sorties s, il arrive que les simulations fournies n'aient pas convergées

, L'idée serait de construire un krigeage fonctionnel pour prédire directement le rendement en tant que sortie fonctionnelle (fonction du débit) et non plus en tant que sortie scalaire

[. Bibliographie and . Apley, Understanding the effects of model uncertainty in robust design with computer experiments, Journal of Mechanical Design, vol.128, issue.4, pp.945-958, 2006.

[. Salem and M. , Model selection and adaptive sampling in surrogate modeling : Kriging and beyong, 2018.

S. Beyer, H. Beyer, and B. Sendhoff, Robust optimization -a comprehensive survey, Computer Methods in Applied Mechanics and Engineering, vol.196, issue.33, pp.3190-3218, 2007.

M. Binois, Uncertainty quantification on pareto fronts and high-dimensional strategies in bayesian optimization, with applications in multi-objective automotive design, 2015.
URL : https://hal.archives-ouvertes.fr/tel-01310521

. Binois, Quantifying uncertainty on Pareto fronts with Gaussian process conditional simulations, European J. Oper. Res, vol.243, issue.2, pp.386-394, 2015.
URL : https://hal.archives-ouvertes.fr/hal-00904811

P. Binois, M. Binois, and V. Picheny, Gpareto : An r package for gaussian-process based multi-objective optimization and analysis

[. Booker, Optimization using surrogate objectives on a helicopter test example, Computational methods for optimal design and control, vol.24, pp.49-58, 1997.

[. Byrd, A limited memory algorithm for bound constrained optimization, SIAM Journal on Scientific Computing, vol.16, issue.5, pp.1190-1208, 1995.

[. Chambers, Statistical models in S, vol.251, 1992.

. Ginsbourger, C. Chevalier, and D. Ginsbourger, Fast computation of the multi-points expected improvement with applications in batch selection, Learning and Intelligent Optimization, pp.59-69, 2013.
URL : https://hal.archives-ouvertes.fr/hal-00732512

[. Coco, Robust optimization criteria : state-of-the-art and new issues, pp.2266-5064, 2014.

[. Coello, . Sierra, C. A. Coello, and M. R. Sierra, A study of the parallelization of a coevolutionary multi-objective evolutionary algorithm, Mexican International Conference on Artificial Intelligence, pp.688-697, 2004.

[. Cornford, Modelling frontal discontinuities in wind fields, Statistical models and methods for discontinuous phenomena, vol.14, pp.43-58, 1998.

N. A. Cressie, Statistics for spatial data, Wiley Series in Probability and Mathematical Statistics : Applied Probability and Statistics, 1993.

[. Darlington, An algorithm for constrained nonlinear optimization under uncertainty, Automatica, vol.35, issue.2, pp.217-228, 1999.

[. Deb, A fast and elitist multiobjective genetic algorithm : Nsga-ii, IEEE transactions on evolutionary computation, vol.6, issue.2, pp.182-197, 2002.

[. Dupuy, Dicedesign and diceeval : Two r packages for design and analysis of computer experiments, Journal of Statistical Software, vol.65, issue.11, pp.1-38, 2015.
URL : https://hal.archives-ouvertes.fr/hal-02065877

N. Durrande, Étude de classes de noyaux adaptées à la simplification et à l'interprétation des modèles d'approximation. Une approche fonctionnelle et probabiliste, 2001.

[. Durrande, Additive covariance kernels for high-dimensional Gaussian process modeling, Ann. Fac. Sci. Toulouse Math, vol.21, issue.6, pp.481-499, 2012.
URL : https://hal.archives-ouvertes.fr/hal-00644934

[. Emmerich, , 2011.

, Hypervolume-based expected improvement : Monotonicity properties and exact computation, Evolutionary Computation (CEC), pp.2147-2154, 2011.

[. Everitt, Cluster analysis. Wiley Series in Probability and Statistics, 2011.

[. Fonseca, An improved dimension-sweep algorithm for the hypervolume indicator, Proceedings of the 2006 Congress on Evolutionary Computation (CEC 2006), pp.1157-1163, 2006.

, Multivariate Gaussian process emulators with nonseparable covariance structures, Technometrics, vol.55, issue.1, pp.47-56, 2013.

[. Friedman, The elements of statistical learning, Springer series in statistics, vol.1, 2001.

[. Gabrel, Recent advances in robust optimization : An overview, European journal of operational research, vol.235, issue.3, pp.471-483, 2014.
URL : https://hal.archives-ouvertes.fr/hal-01495346

[. Gao, Multiples métamodèles pour l'approximation et l'optimisation de fonctions numériques multivariables, AI 2002 : Advances in artificial intelligence, vol.2557, pp.395-406, 2002.

[. Ginsbourger, Kriging is wellsuited to parallelize optimization, Computational Intelligence in Expensive Optimization Problems, pp.131-162, 2010.
URL : https://hal.archives-ouvertes.fr/emse-00436126

[. Ginsbourger, On ANOVA decompositions of kernels and Gaussian random field paths. In Monte Carlo and quasi-Monte Carlo methods, Springer Proc. Math. Stat, vol.163, pp.315-330, 2016.
URL : https://hal.archives-ouvertes.fr/emse-01339368

. Springer,

[. Göhler, Robustness metrics : Consolidating the multiple approaches to quantify robustness, Journal of the Royal Statistical Society. Series C (Applied Statistics), vol.138, issue.11, pp.100-108, 1979.

. Henkenjohann, N. Kunert-;-henkenjohann, and J. Kunert, An efficient sequential optimization approach based on the multivariate expected improvement criterion, Quality Engineering, vol.19, issue.4, pp.267-280, 2007.

B. Iooss and P. Lemaître, A Review on Global Sensitivity Analysis Methods, pp.101-122, 2015.

R. Darlington, ]. J. Darlington, C. Pantelides, B. T. Rustem, B. Janusevskis et al., An Efficient Approximate Algorithm for Robust Optimal Decisions under Uncertainty, Journal of Global Optimization, vol.55, issue.2, pp.313-336, 2013.

O. Jeong, S. Obayashi, and S. , Efficient global optimization (ego) for multi-objective problem and data mining, Evolutionary Computation, 2005. The 2005 IEEE Congress on, vol.3, pp.2138-2145, 2005.

[. Jin, An efficient algorithm for constructing optimal design of computer experiments, Journal of Statistical Planning and Inference, vol.134, issue.1, pp.268-287, 2005.

D. R. Jones, A taxonomy of global optimization methods based on response surfaces, Journal of global optimization, vol.21, issue.4, pp.345-383, 2001.

[. Jones, Efficient global optimization of expensive black-box functions, Journal of Global optimization, vol.13, issue.4, pp.455-492, 1998.

N. Jozefowiez, Optimisation combinatoire multi-objectif : des méthodes aux problèmes, de la Terre à (presque) la Lune, 2013.

J. Knowles-;-knowles, Parego : A hybrid algorithm with on-line landscape approximation for expensive multiobjective optimization problems, IEEE Transactions on Evolutionary Computation, vol.10, issue.1, pp.50-66, 2006.

[. Gratiet and L. , Multi-fidelity Gaussian process regression for computer experiments, 2013.
URL : https://hal.archives-ouvertes.fr/tel-00866770

[. Lelièvre, On the consideration of uncertainty in design : optimization -reliability -robustness. Structural and Multidisciplinary Optimization, vol.54, pp.1423-1437, 2016.

[. Liu, On the performance of metamodel assisted moea/d, International Symposium on Intelligence Computation and Applications, pp.547-557, 2007.

H. Maatouk-;-maatouk, Correspondance entre régression par processus Gaussien et splines d'interpolation sous contraintes linéaires de type inégalité, 2015.

[. Marmin, Differentiating the multipoint expected improvement for optimal batch design, Machine Learning, Optimization, and Big Data, pp.37-48, 2015.
URL : https://hal.archives-ouvertes.fr/hal-01133220

. Marrel, An efficient methodology for modeling complex computer codes with Gaussian processes, Comput. Statist. Data Anal, vol.52, issue.10, pp.4731-4744, 2008.
URL : https://hal.archives-ouvertes.fr/hal-00239492

[. Marzat, Worst-case global optimization of black-box functions through kriging and relaxation, Journal of Global Optimization, vol.55, issue.4, pp.707-727, 2013.
URL : https://hal.archives-ouvertes.fr/hal-00682475

. [mebane, Genetic optimization using derivatives : the rgenoud package for r, Journal of Statistical Software, vol.42, issue.11, pp.1-26, 2011.

[. Mkaouer, High dimensional search-based software engineering : finding tradeoffs among 15 objectives for automating software refactoring using nsga-iii, Proceedings of the 2014 Annual Conference on Genetic and Evolutionary Computation, vol.4, pp.55-100, 2006.

M. Morris, M. D. Morris, and T. J. Mitchell, Exploratory designs for computational experiments, Journal of statistical planning and inference, vol.43, issue.3, pp.381-402, 1995.

[. Muehlenstaedt, Datadriven Kriging models based on FANOVA-decomposition, Stat. Comput, vol.22, issue.3, pp.723-738, 2012.
URL : https://hal.archives-ouvertes.fr/hal-00537781

C. J. Paciorek and M. J. Schervish, Spatial modelling using a new class of nonstationary covariance functions, Environmetrics, vol.17, issue.5, pp.483-506, 2006.

E. Padonou and O. Roustant, Polar Gaussian processes and experimental designs in circular domains, SIAM/ASA J. Uncertain. Quantif, vol.4, issue.1, pp.1014-1033, 2016.
URL : https://hal.archives-ouvertes.fr/emse-01412189

J. Parr, Improvement criteria for constraint handling and multiobjective optimization, 2013.

V. Picheny-;-picheny, Multiobjective optimization using gaussian process emulators via stepwise uncertainty reduction, Statistics and Computing, vol.25, issue.6, pp.1265-1280, 2015.
URL : https://hal.archives-ouvertes.fr/hal-00868472

[. Ponweiser, Dicekriging, diceoptim : Two r packages for the analysis of computer experiments by kriging-based metamodeling and optimization, International Conference on Parallel Problem Solving from Nature, vol.36, pp.1-55, 2003.

[. Saltelli, Sensitivity analysis, vol.1, 2000.
URL : https://hal.archives-ouvertes.fr/inria-00386559

[. Santner, The design and analysis of computer experiments, Springer Series in Statistics, 2003.

J. R. Schott-;-schott, ;. Schwarz, and G. , Fault tolerant design using single and multicriteria genetic algorithm optimization, Ann. Statist, vol.6, issue.2, pp.461-464, 1978.

E. L. Snelson, Flexible and efficient Gaussian process models for machine learning, ProQuest LLC, 2008.

I. M. Sobol-;-sobol, Sensitivity estimates for nonlinear mathematical models, Mathematical modelling and computational experiments, vol.1, issue.4, pp.407-414, 1993.

M. L. Stein, Interpolation of spatial data, Springer Series in Statistics, 1999.

. Stitson, Support vector regression with anova decomposition kernels, Advances in kernel methods-Support vector learning, pp.285-292, 1997.

B. Sudret and J. Svenson, Computer experiments : Multiobjective optimization and sensitivity analysis, 2011.

J. Svenson and T. Santner, Multiobjective optimization of expensive-to-evaluate deterministic computer simulator models, Computational Statistics & Data Analysis, vol.94, pp.250-264, 2016.
DOI : 10.1016/j.csda.2015.08.011
URL : https://manuscript.elsevier.com/S0167947315001991/pdf/S0167947315001991.pdf

[. Tarantola, Estimating the approximation error when fixing unessential factors in global sensitivity analysis, Reliability Engineering & System Safety, vol.92, issue.7, pp.957-960, 2007.

[. Troian, Methodology for the design of the geometry of a cavity and its absorption coefficients as random design variables under vibroacoustic criteria, Journal of Computational Acoustics, vol.24, issue.02, p.1650006, 2016.

S. Ur-rehman and M. Langelaar, Efficient global robust optimization of unconstrained problems affected by parametric uncertainties. Structural and Multidisciplinary Optimization, vol.52, pp.319-336, 2015.

. [ur-rehman, Efficient kriging-based robust optimization of unconstrained problems, Journal of Computational Science, vol.5, issue.6, pp.872-881, 2014.

[. Van-loan-;-van-loan and C. F. , Matrix computations (johns hopkins studies in mathematical sciences), 1996.

[. Van-veldhuizen-;-van-veldhuizen, D. A. Vazquez, E. Bect, and J. , Convergence properties of the expected improvement algorithm with fixed mean and covariance functions, Journal of Statistical Planning and Inference, vol.140, issue.11, pp.3088-3095, 1999.

. Villa-vialaneix, A comparison of eight metamodeling techniques for the simulation of n 2 o fluxes and n leaching from corn crops, Environmental Modelling & Software, vol.34, pp.51-66, 2012.
URL : https://hal.archives-ouvertes.fr/hal-00654753

[. Wagner, On expectedimprovement criteria for model-based multi-objective optimization, International Conference on Parallel Problem Solving from Nature, pp.718-727, 2010.

[. Welch, Screening, predicting, and computer experiments, Technometrics, vol.34, issue.1, pp.15-25, 1992.

G. Yi, Variable selection with penalized Gaussian process regression models, 2009.

[. Zhang, Expensive multiobjective optimization by moea/d with gaussian process model, IEEE Transactions on Evolutionary Computation, vol.14, issue.3, pp.456-474, 2010.

T. Zitzler, E. Zitzler, and L. Thiele, Multiobjective evolutionary algorithms : a comparative case study and the strength pareto approach, IEEE Transactions on Evolutionary Computation, vol.3, issue.4, pp.257-271, 1999.