Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

Bouhlel, Mohamed Amine; Bartoli, Nathalie; Regis, Rommel G.; Otsmane, Abdelkader; Morlier, Joseph

doi:10.1080/0305215x.2017.1419344

Cited by 89 publications

(55 citation statements)

References 40 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bouhlel et al [42] subsequently improved KPLS by adding a new step in the construction of the surrogate model, which improves the accuracy for high-dimensional problems (KPLS+K). These approaches were used within the SEGO algorithm, and they were demonstrated to be efficient for problems with up to 50 design variables [43].…”

Section: Handling a Large Number Of Design Variablesmentioning

confidence: 99%

“…We recently proposed to use EGO with a new type of kriging model adapted to high-dimensional problems (namely, KPLS and KPLS+K) [41,42]. The resulting optimization approach, which we call SEGOKPLS(+K), was demonstrated in problems with up to 50 design variables that were solved using approximately a hundred function evaluations [43]. To handle nonlinear functions that vary significantly within a wide domain, researchers have proposed to cluster the data and construct an assembly of local surrogates, known as mixture of experts (MOE), which facilitate global optimization [44][45][46][47][48].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design

Bartoli

Lefèbvre

Dubreuil

et al. 2019

Aerospace Science and Technology

Self Cite

View full text Add to dashboard Cite

Surrogate models are often used to reduce the cost of design optimization problems that involve computationally costly models, such as computational fluid dynamics simulations. However, the number of evaluations required by surrogate models usually scales poorly with the number of design variables, and there is a need for both better constraint formulations and multimodal function handling. To address this issue, we developed a surrogate-based gradient-free optimization algorithm that can handle cases where the function evaluations are expensive, the computational budget is limited, the functions are multimodal, and the optimization problem includes nonlinear equality or inequality constraints. The proposed algorithm-super efficient global optimization coupled with mixture of experts (SEGOMOE)can tackle complex constrained design optimization problems through the use of an enrichment strategy based on a mixture of experts coupled with adaptive surrogate models. The performance of this approach was evaluated for analytic constrained and unconstrained problems, as well as for a multimodal aerodynamic shape optimization problem with 17 design variables and an equality constraint. Our results showed that the method is efficient and that the optimum is much less dependent on the starting point than the conventional gradient-based optimization.

show abstract

Section: Handling a Large Number Of Design Variablesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design

Bartoli

Lefèbvre

Dubreuil

et al. 2019

Aerospace Science and Technology

Self Cite

View full text Add to dashboard Cite

show abstract

“…With high damping coefficients and multiple tuning resonance frequencies, this system can obtain wide regions of attenuation frequency. The global optimization techniques (Bouhlel et al 2018;Chen et al 2019;Kadlec and Šeděnka 2018) are also very effective to improve the performance of LMs. The propagation of elastic wave in a beam lattice material characterized by the anti-chiral cell was investigated by (Bacigalupo et al 2016).…”

Section: Introductionmentioning

confidence: 99%

Improved impact responses of a honeycomb sandwich panel structure with internal resonators

et al. 2019

Engineering Optimization

View full text Add to dashboard Cite

In this paper, a decahedral honeycomb sandwich panel structure with periodic internal resonators is developed in order to effectively mitigate the impact force and dissipate the impact energy. Five types of local resonance microstructures (LMs) with the property of negative effective mass are applied on the sandwich structure to study the effects of different resonators on load attenuation, and the theoretical and numerical results show that the multi-resonator model (MM) exhibits the best attenuation effect. In addition, the damping characteristic is added to the MM to further improve the structure responses. The analysis results have clearly demonstrated that the combination of damping characteristic and local resonance of multi-resonators can significantly dissipate the impact energy and mitigate the dynamic load. In addition, the optimization analyses are also carried out to optimize the spring stiffness, lumped mass and damping coefficient. The performance of the sandwich panel structure is significantly improved with a relatively small structure mass.

show abstract

“…Besides, the machine learning community has generalized Kriging's theory into a versatile computational framework called GPML [19]. This methodology is well known in machine learning/engineering optimization for reaching a global optimum at a fixed budget even at in a high dimensional constrained optimization problem [20]. It also has been validated in aerodynamic shape optimization [21,22] using a modified EGO algorithm based on Mixture of Experts (MOE).…”

Section: Introductionmentioning

confidence: 99%

An EGO-like optimization framework for sensor placement optimization in modal analysis

Morlier¹,

Basile²,

Chiplunkar³

et al. 2018

Smart Mater. Struct.

Self Cite

View full text Add to dashboard Cite

In aircraft design, Ground/Flight Vibration Tests (GVT/FVT) are conducted to extract aircraft's modal parameters (natural frequencies, damping ratios and mode shapes) also known as the modal basis. The main problem in aircraft modal identification is the large number of sensors needed, which increases operational time and costs. The goal of this paper is to minimize the number of sensors by optimizing their locations in order to reconstruct a truncated modal basis of N mode shapes with a high level of accuracy in the reconstruction. There are several methods to solve Sensors Placement Optimization (SPO) problems, but for this case an original approach has been established based on an iterative process for mode shapes reconstruction through an adaptive Kriging Metamodeling approach so-called EGO-SPO. The main idea in this publication is to solve an optimization problem where the sensors locations are variables and the objective function is defined by maximizing the trace of criteria so-called AutoMAC. The results on a 2D wing demonstrate a reduction of sensors by 30% using our EGO-SPO strategy.

show abstract

Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

Cited by 89 publications

References 40 publications

Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design

Adaptive modeling strategy for constrained global optimization with application to aerodynamic wing design

Improved impact responses of a honeycomb sandwich panel structure with internal resonators

An EGO-like optimization framework for sensor placement optimization in modal analysis

Contact Info

Product

Resources

About