From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information

Ander  Gray; Alexander Wimbush; Marco de Angelis; Peter O. Hristov; Dominic Calleja; Enrique Miralles-Dolz; Roberto Rocchetta

doi:10.1016/j.ymssp.2021.108210

From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information

Ander Gray, Alexander Wimbush, Marco de Angelis (Corresponding author), Peter O. Hristov (Corresponding author), Dominic Calleja, Enrique Miralles-Dolz, Roberto Rocchetta (Corresponding author)

Onderzoeksoutput: Bijdrage aan tijdschrift › Tijdschriftartikel › Academic › peer review

1 Downloads (Pure)

Samenvatting

In this paper we present a framework for addressing a variety of engineering design challenges with limited empirical data and partial information. This framework includes guidance on the characterisation of a mixture of uncertainties, efficient methodologies to integrate data into design decisions, and to conduct reliability analysis, and risk/reliability based design optimisation. To demonstrate its efficacy, the framework has been applied to the NASA 2020 uncertainty quantification challenge. The results and discussion in the paper are with respect to this application.

Originele taal-2	Engels
Artikelnummer	108210
Aantal pagina's	39
Tijdschrift	Mechanical Systems and Signal Processing
Volume	165
DOI's	https://doi.org/10.1016/j.ymssp.2021.108210
Status	Gepubliceerd - 15 feb 2022

Toegang tot document

10.1016/j.ymssp.2021.108210Licentie: CC BY

published versionDefinitieve gepubliceerde versie, 13,6 MBLicentie: CC BY

Citeer dit

@article{12c1fd1e30a94f9b86838ec9018b7e74,

title = "From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information",

abstract = "In this paper we present a framework for addressing a variety of engineering design challenges with limited empirical data and partial information. This framework includes guidance on the characterisation of a mixture of uncertainties, efficient methodologies to integrate data into design decisions, and to conduct reliability analysis, and risk/reliability based design optimisation. To demonstrate its efficacy, the framework has been applied to the NASA 2020 uncertainty quantification challenge. The results and discussion in the paper are with respect to this application.",

keywords = "Bayesian calibration, Probability bounds analysis, Epistemic uncertainty, Uncertainty propagation, Uncertainty reduction, Optimisation under uncertainty",

author = "Ander Gray and Alexander Wimbush and {de Angelis}, Marco and Hristov, {Peter O.} and Dominic Calleja and Enrique Miralles-Dolz and Roberto Rocchetta",

year = "2022",

month = feb,

day = "15",

doi = "10.1016/j.ymssp.2021.108210",

language = "English",

volume = "165",

journal = "Mechanical Systems and Signal Processing",

issn = "0888-3270",

publisher = "Academic Press Inc.",

}

From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information. / Gray, Ander ; Wimbush, Alexander; de Angelis, Marco (Corresponding author); Hristov, Peter O. (Corresponding author); Calleja, Dominic; Miralles-Dolz, Enrique; Rocchetta, Roberto (Corresponding author).

In: Mechanical Systems and Signal Processing, Vol. 165, 108210, 15.02.2022.

Onderzoeksoutput: Bijdrage aan tijdschrift › Tijdschriftartikel › Academic › peer review

TY - JOUR

T1 - From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information

AU - Gray, Ander

AU - Wimbush, Alexander

AU - de Angelis, Marco

AU - Hristov, Peter O.

AU - Calleja, Dominic

AU - Miralles-Dolz, Enrique

AU - Rocchetta, Roberto

PY - 2022/2/15

Y1 - 2022/2/15

N2 - In this paper we present a framework for addressing a variety of engineering design challenges with limited empirical data and partial information. This framework includes guidance on the characterisation of a mixture of uncertainties, efficient methodologies to integrate data into design decisions, and to conduct reliability analysis, and risk/reliability based design optimisation. To demonstrate its efficacy, the framework has been applied to the NASA 2020 uncertainty quantification challenge. The results and discussion in the paper are with respect to this application.

AB - In this paper we present a framework for addressing a variety of engineering design challenges with limited empirical data and partial information. This framework includes guidance on the characterisation of a mixture of uncertainties, efficient methodologies to integrate data into design decisions, and to conduct reliability analysis, and risk/reliability based design optimisation. To demonstrate its efficacy, the framework has been applied to the NASA 2020 uncertainty quantification challenge. The results and discussion in the paper are with respect to this application.

KW - Bayesian calibration

KW - Probability bounds analysis

KW - Epistemic uncertainty

KW - Uncertainty propagation

KW - Uncertainty reduction

KW - Optimisation under uncertainty

U2 - 10.1016/j.ymssp.2021.108210

DO - 10.1016/j.ymssp.2021.108210

M3 - Article

VL - 165

JO - Mechanical Systems and Signal Processing

JF - Mechanical Systems and Signal Processing

SN - 0888-3270

M1 - 108210

ER -

From inference to design: A comprehensive framework for uncertainty quantification in engineering with limited information

Samenvatting

Toegang tot document

Vingerafdruk

Citeer dit