Efficient shape optimization for fluid–structure interaction problems

Aghajari, Nima; Schäfer, Michael

doi:10.1016/j.jfluidstructs.2015.06.011

Cited by 14 publications

(10 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…f (1) is the inviscid MF-surrogate with predictions the closest to the viscous ones, albeit with the least dependences on the trimming parameters compared to the other inviscid surrogates b f (2,3) , which exhibit more noticeable dependences on the optimization variables. In contrast, the viscous MF-surrogates b f (4) and b f (5) have predictions in close agreement to each other over the whole domain ⌦. This proximity is not surprising since, compared to b f (4) , the construction of b f (5) incorporates just a few evaluations of the objective function f (5) , based on a more refined mesh and a different boundary flow conditions, but the same physical models otherwise.…”

Section: Optimization Resultsmentioning

confidence: 72%

“…Efficient Global Optimization (EGO) strategies [1] have been used in various application domains, and many works demonstrated their applicability for the optimization of complex systems with costly objective function evaluation [2]. Examples of applications are the aerodynamic drag reduction [3], the vibration reduction for rotating aircraft [4], the optimization of Fluid-Structure Interactions (FSI) problems [5] and the sail trimming optimization [6]. In a nutshell, the EGO method builds a statistical surrogate model of the costly objective function f to be optimized, usually a Gaussian Processes [7] (GP), and a statistical criterion accounting for the Expected Improvement (EI), often referred to as merit function, is maximized to select a new design point.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Non-Nested Infilling Strategy for Multifidelity Based Efficient Global Optimization

Sacher

Maître

Duvigneau

et al. 2021

Int. J. UncertaintyQuantification

View full text Add to dashboard Cite

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. Multi-fidelity 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 multi-fidelity approaches in the context of EGO is still a research topic. In this work, we propose a new effective infilling strategy for multi-fidelity 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 multi-fidelity 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 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.

show abstract

Section: Optimization Resultsmentioning

confidence: 72%

Section: Introductionmentioning

confidence: 99%

A Non-Nested Infilling Strategy for Multifidelity Based Efficient Global Optimization

Sacher

Maître

Duvigneau

et al. 2021

Int. J. UncertaintyQuantification

View full text Add to dashboard Cite

show abstract

“…GP models have also been found especially appealing for optimization, in the framework of the Efficient Global Optimization (EGO) [10] method, because their statistical nature allows to provide both a prediction of the objective function, in terms of model mean, and an error estimate, in terms of model variance. EGO strategies have been used in several engineering applications, such as aerodynamic drag reduction of transonic wings [11], vibration reduction for rotating aircrafts [12,13], optimization of FSI problems [14] and sail trimming optimization [15]. The EGO efficiency has been demonstrated for the optimization of complex systems with costly objective function evaluations [1].…”

Section: Introductionmentioning

confidence: 99%

A classification approach to efficient global optimization in presence of non-computable domains

Sacher

Duvigneau

Maître

et al. 2018

Struct Multidisc Optim

View full text Add to dashboard Cite

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 statistical surrogate model, using a training set of computed objective function values, which is refined according to a prescribed infilling strategy. However, this sequential optimization procedure can stop prematurely if the objective function cannot be computed at a proposed point. Such a situation can occur when the search space encompasses design points corresponding to an unphysical configuration, an ill-posed problem, or a non-computable problem due to the limitation of numerical solvers. To avoid such a premature stop in the optimization procedure, we propose to use a classification model to learn non-computable areas and to adapt the infilling strategy accordingly. Specifically, the proposed method splits the training set into two subsets composed of computable and non-computable points. A surrogate model for the objective function is built using the training set of computable points, only, whereas a probabilistic classification model is built using the union of the computable and non-computable training sets. The classifier is then incorporated in the surrogate-based optimization procedure to avoid proposing new points in the noncomputable domain while improving the classification uncertainty if needed. The method has the advantage to automatically adapt both the surrogate of the objective function and the classifier during the iterative optimization process. Therefore, non-computable areas do not need to be a priori known. The proposed method is applied to several analytical problems presenting different types of difficulty, and to the optimization of a fully nonlinear fluid-structure interaction system. The latter problem concerns the drag minimization of a flexible hydrofoil with cavitation constraints. The efficiency of the proposed method compared favorably to a reference evolutionary algorithm, except for situations where the feasible domain is a small portion of the design space.

show abstract

“…In [17] a partitioned approach for a NURBS-parametrized shape optimization problem is studied. In the recent work [18] the authors propose a reduced order model, based on a sequential quadratic programming approach, to accelerate the shape optimization process.…”

Section: Introductionmentioning

confidence: 99%

Adjoint Optimal Control Problems for Fluid-Structure Interaction Systems

Cerroni¹,

Davià²,

Manservisi³

et al. 2016

Proceedings of the VII European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS Congress 2016)

View full text Add to dashboard Cite

Abstract. In the last years adjoint optimal control has been increasingly used for design and simulations in several research fields. Applications to Computational Fluid Dynamics problems dedicated to the study of transient-diffusion equations, shape optimization problems, fluid-solid conjugate heat transfer and turbulent flows can be found in literature. The study of FluidStructure Interaction problems gained popularity recently because of many interesting applications in engineering and biomedical fields. In this paper we study adjoint optimal control problems for Fluid-Structure Interaction systems in order to improve the advantages of using FSI simulations when designing engineering devices where fluid-dynamical interactions between a fluid and a solid play a significant role. We assess distributed optimal control problems with the purpose to control the fluid behavior by moving the solid region to obtain a desired fluid velocity in specific parts of the domain. The adjoint equations of the FSI monolithic system are derived and the optimality system solved for some simple cases with an in-house finite element code with mesh-moving capabilities for the study of large displacements in the solid. The approach presented in this work is general and can be used to assess different objectives and types of control in future works.

show abstract

Efficient shape optimization for fluid–structure interaction problems

Cited by 14 publications

References 26 publications

A Non-Nested Infilling Strategy for Multifidelity Based Efficient Global Optimization

A Non-Nested Infilling Strategy for Multifidelity Based Efficient Global Optimization

A classification approach to efficient global optimization in presence of non-computable domains

Adjoint Optimal Control Problems for Fluid-Structure Interaction Systems

Contact Info

Product

Resources

About