Adaptive infill sampling criterion for multi-fidelity gradient-enhanced kriging model

Hao, Peng; Feng, Shaojun; Li, Yuwei; Wang, Bo; Chen, Huihan

doi:10.1007/s00158-020-02493-8

Cited by 49 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this issue, the multi-fidelity (MF) metamodel may be a good alternative to make a trade-off between the high computational cost and accuracy by combining LF and HF samples. [25][26][27][28] Co-GPM (also known as multi-fidelity GPM) is a popular MF metamodel, which can be considered as an extension of a multivariable Kriging model with the assistance of secondary information. Kennedy and O'Hagan (KOH) 29 extended the co-Kriging model from mining engineering to computer experiments considering both LF and HF data.…”

Section: Introductionmentioning

confidence: 99%

“…On the other hand, the low‐fidelity (LF) numerical models cannot provide samples that are reliable enough to train an accurate GPM. To address this issue, the multi‐fidelity (MF) metamodel may be a good alternative to make a trade‐off between the high computational cost and accuracy by combining LF and HF samples 25–28 . Co‐GPM (also known as multi‐fidelity GPM) is a popular MF metamodel, which can be considered as an extension of a multivariable Kriging model with the assistance of secondary information.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Analytical uncertainty quantification approach based on adaptive generalized co‐Gaussian process model

Wan

Zhang

Luo

et al. 2022

Numerical Meth Engineering

View full text Add to dashboard Cite

Gaussian process modeling (GPM) is widely used to replace physical models for uncertainty quantification (UQ). The fidelity of training samples has a major impact on the prediction performance of the GPM, and the computational time of obtaining high‐fidelity (HF) samples from simulations may be very prohibitively expensive. Therefore, this study proposes the use of the multi‐fidelity GPM for UQ to make a trade‐off between the computational cost and accuracy by combining a few HF samples with many computationally‐cheaper low‐fidelity (LF) ones. Currently, the co‐Gaussian process model (co‐GPM, also known as multi‐fidelity GPM) requires that the LF and HF samples be nested. A generalized co‐Gaussian process model (GC‐GPM) is developed that can mix LF and HF samples under both nested and un‐nested conditions. In addition, to further alleviate the computational cost, an adaptive sampling strategy is introduced into GC‐GPM to reduce the number of training samples using a new infill criterion. Within the adaptive GC‐GPM framework, the high dimensional integrals associated with the statistical moments (mean and variance) of structural response can be decomposed to single dimensional ones that are analytically tractable, which finally allows for analytical calculation of the mean and variance. Three full‐scale structures are adopted to verify the reliability and efficiency of the proposed adaptive GC‐GPM method, and traditional GC‐GPM and Monte Carlo simulation (MCS) are used for comparison purpose. The comparison results show that the UQ results of the adaptive GC‐GPM match well with those of the MCS and compared to the traditional GC‐GPM, the adaptive GC‐GPM owns higher computational accuracy and efficiency.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Analytical uncertainty quantification approach based on adaptive generalized co‐Gaussian process model

Wan

Zhang

Luo

et al. 2022

Numerical Meth Engineering

View full text Add to dashboard Cite

show abstract

“…Indeed, surrogate models can reduce significantly the computational burden needed to solve advanced physical problems, concerned with design of earthquake-resistant systems [25], nonlinear analysis of carbon nanotubes [26], reliability of large-scale structures [27,28], fluid mechanics in turbulent flows [29,30]. Despite being seldom and only recently employed in the specific field of metamaterial design [31][32][33], surrogate optimization techniques have been successfully applied in a variety of advanced applications, especially related to optimal material design problems involving -for instance -matrix and precipitate crystal structures [34], stiffened shells [35], hierarchical stiffened plates and shells [36][37][38], carbon fiber reinforced polymer laminates [39], hysteretic multilayer nanocomposites [40], variable-stiffness composite panels [41].…”

Section: Introductionmentioning

confidence: 99%

Computational design of innovative mechanical metafilters via adaptive surrogate-based optimization

Bacigalupo

Gnecco

Lepidi

et al. 2021

Computer Methods in Applied Mechanics and Engineering

View full text Add to dashboard Cite

“…An extensive review of surrogate modelling techniques in support of engineering design optimization can be found in Queipo et al (2005). The latest and significant studies on surrogate based optimization include Hao et al (2018Hao et al ( , 2020 and Fern andez-Godino et al (2019).…”

Section: Introductionmentioning

confidence: 99%

Structural optimization of a composite launch tube of man portable air defense system

Yar

Acar

2021

AEAT

View full text Add to dashboard Cite

Purpose The purpose of this paper is to find the optimum configuration of the composite launch tube currently being developed in Roketsan. The winding thicknesses and winding angles of the launch tube are selected as design variables, and three different composite material alternatives are evaluated: glass/epoxy, carbon/epoxy and aramid/epoxy. Design/methodology/approach In this study, structural optimization of a composite launch tube of man portable air defense system is conducted. To achieve a cost-effective design, a cost scoring table that includes structural weight, material cost, availability and manufacturability is first introduced. Then, optimization for minimum weight is conducted, where the winding thicknesses and winding angle are taken as design variables, and the safety factor value obtained by using the Tsai–Wu damage criterion is used as constraint. A surrogate-based optimization approach is used where various options for surrogate models are evaluated. Glass/epoxy, carbon/epoxy and aramid/epoxy are considered as alternative materials for the launch tube. Finally, the selection of the most cost-effective design is performed to achieve optimum cost. Findings Carbon fiber-reinforced epoxy matrix material provides the optimum cost-effective design for the launch tube. Practical implications The findings of the paper can be used to design more cost-efficient composite launch tube currently being developed in Roketsan. Originality/value The existing studies are based on a design approach to achieve minimum weight of the launch tubes, whereas this study introduces a design approach to achieve optimum cost.

show abstract

Adaptive infill sampling criterion for multi-fidelity gradient-enhanced kriging model

Cited by 49 publications

References 47 publications

Analytical uncertainty quantification approach based on adaptive generalized co‐Gaussian process model

Analytical uncertainty quantification approach based on adaptive generalized co‐Gaussian process model

Computational design of innovative mechanical metafilters via adaptive surrogate-based optimization

Structural optimization of a composite launch tube of man portable air defense system

Contact Info

Product

Resources

About