A Non-Nested Infilling Strategy for Multi-Fidelity based Efficient Global Optimization
Article dans une revue avec comité de lecture
JournalInternational Journal for Uncertainty Quantification
Efficient global optimization (EGO) has become a standard approach for the global optimization of complex systems with high computational costs. EGO uses a training set of objective function values computed at selected input points to construct a statistical surrogate model, with low evaluation cost, on which the optimization procedure is applied. The training set is sequentially enriched, selecting new points, according to a prescribed infilling strategy, in order to converge to the optimum of the original costly model. Multifidelity approaches combining evaluations of the quantity of interest at different fidelity levels have been recently introduced to reduce the computational cost of building a global surrogate model. However, the use of multifidelity approaches in the context of EGO is still a research topic. In this work, we propose a new effective infilling strategy for multifidelity EGO. Our infilling strategy has the particularity of relying on non-nested training sets, a characteristic that comes with several computational benefits. For the enrichment of the multifidelity training set, the strategy selects the next input point together with the fidelity level of the objective function evaluation. This characteristic is in contrast with previous nested approaches, which require estimation of all lower fidelity levels and are more demanding to update the surrogate. The resulting EGO procedure achieves a significantly reduced computational cost, avoiding computations at useless fidelity levels whenever possible, but it is also more robust to low correlations between levels and noisy estimations. Analytical problems are used to test and illustrate the efficiency of the method. It is finally applied to the optimization of a fully nonlinear fluid-structure interaction system to demonstrate its feasibility on real large-scale problems, with fidelity levels mixing physical approximations in the constitutive models and discretization refinements.
Files in this item
Showing items related by title, author, creator and subject.
Communication avec acteSACHER, Matthieu; DURAND, Mathieu; BERRINI, Elisa; HAUVILLE, Frédéric; DUVIGNEAU, Régis; LE MAITRE, Olivier; ASTOLFI, Jacques-André (2017)This paper investigates the use of constrained surrogate models to solve the multi-design optimization problem of a flexible hydrofoil. The surrogate-based optimization (EGO) substitutes the complex objective function of ...
Efficient optimization procedure in non-linear fluid-structure interaction problem: Application to mainsail trimming in upwind conditions Article dans une revue avec comité de lectureSACHER, Matthieu; HAUVILLE, Frédéric; DUVIGNEAU, Régis; LE MAÎTRE, Olivier; AUBIN, Nicolas; DURAND, Mathieu (Elsevier, 2017)This paper investigates the use of Gaussian processes to solve sail trimming optimization problems. The Gaussian process, used to model the dependence of the performance with the trimming parameters, is constructed from a ...
Communication avec acteSACHER, Matthieu; HAUVILLE, Frédéric; DUVIGNEAU, Régis; LE MAITRE, Olivier; AUBIN, Nicolas; DURAND, Mathieu (2016)This paper investigates the use of meta-models for optimizing sails trimming. A Gaussian process is used to robustly approximate the dependence of the performance with the trimming parameters to be optimized. The Gaussian ...
Surrogates and Classification approaches for Efficient Global Optimization (EGO) with Inequality Constraints Communication avec acteSACHER, Matthieu; DUVIGNEAU, Régis; LE MAITRE, Olivier; DURAND, Mathieu; BERRINI, Elisa; HAUVILLE, Frédéric; ASTOLFI, Jacques-André (2017-05)In this work, we compare the use of Gaussian Process (GP) models for the constraints [Schonlau 1997] with a classification approach relying on a Least-Squares Support Vector Machine (LS-SVM) [Suykens and Vandewalle 1999]. ...
Article dans une revue avec comité de lectureSACHER, Matthieu; DUVIGNEAU, Régis; LE MAÎTRE, Olivier; DURAND, Mathieu; BERRINI, Elisa; HAUVILLE, Frédéric; ASTOLFI, Jacques-André (Springer Verlag (Germany), 2018)Gaussian-Process based optimization methods have become very popular in recent years for the global optimization of complex systems with high computational costs. These methods rely on the sequential construction of a ...