Investigating Bayesian robust experimental design with principles of global sensitivity analysis

Fei He, Hong Yue, Martin Brown

Research output: Contribution to conferencePaper

Abstract

The purpose of model-based experimental design is to maximise the information gathered for quantitative model identification. Instead of the commonly used optimal experimental design, robust experimental design aims to address parametric uncertainties in the design process. In this paper, the Bayesian robust experimental design is investigated, where both a Monte Carlo sampling strategy and local sensitivity evaluation at each sampling point are employed to achieve the robust solution. The link between global sensitivity analysis (GSA) and the Bayesian robust experimental design is established. It is revealed that a lattice sampling based GSA strategy, the Morris method, can be explicitly interpreted as the Bayesian A-optimal design for the uniform hypercube type uncertainties.

Conference

Conference9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010)
CountryBelgium
CityLueven
Period5/07/107/07/10

Fingerprint

Design of experiments
Sensitivity analysis
Sampling
Identification (control systems)
Uncertainty

Keywords

  • bayesian analysis
  • global sensitivity analysis
  • modeling

Cite this

He, F., Yue, H., & Brown, M. (2010). Investigating Bayesian robust experimental design with principles of global sensitivity analysis. 563-568. Paper presented at 9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010), Lueven, Belgium. https://doi.org/10.3182/20100705-3-BE-2011.00096
He, Fei ; Yue, Hong ; Brown, Martin. / Investigating Bayesian robust experimental design with principles of global sensitivity analysis. Paper presented at 9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010), Lueven, Belgium.6 p.
@conference{4c80592579594759a3f75d2cc09ac49f,
title = "Investigating Bayesian robust experimental design with principles of global sensitivity analysis",
abstract = "The purpose of model-based experimental design is to maximise the information gathered for quantitative model identification. Instead of the commonly used optimal experimental design, robust experimental design aims to address parametric uncertainties in the design process. In this paper, the Bayesian robust experimental design is investigated, where both a Monte Carlo sampling strategy and local sensitivity evaluation at each sampling point are employed to achieve the robust solution. The link between global sensitivity analysis (GSA) and the Bayesian robust experimental design is established. It is revealed that a lattice sampling based GSA strategy, the Morris method, can be explicitly interpreted as the Bayesian A-optimal design for the uniform hypercube type uncertainties.",
keywords = "bayesian analysis, global sensitivity analysis , modeling",
author = "Fei He and Hong Yue and Martin Brown",
year = "2010",
doi = "10.3182/20100705-3-BE-2011.00096",
language = "English",
pages = "563--568",
note = "9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010) ; Conference date: 05-07-2010 Through 07-07-2010",

}

He, F, Yue, H & Brown, M 2010, 'Investigating Bayesian robust experimental design with principles of global sensitivity analysis' Paper presented at 9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010), Lueven, Belgium, 5/07/10 - 7/07/10, pp. 563-568. https://doi.org/10.3182/20100705-3-BE-2011.00096

Investigating Bayesian robust experimental design with principles of global sensitivity analysis. / He, Fei; Yue, Hong; Brown, Martin.

2010. 563-568 Paper presented at 9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010), Lueven, Belgium.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Investigating Bayesian robust experimental design with principles of global sensitivity analysis

AU - He, Fei

AU - Yue, Hong

AU - Brown, Martin

PY - 2010

Y1 - 2010

N2 - The purpose of model-based experimental design is to maximise the information gathered for quantitative model identification. Instead of the commonly used optimal experimental design, robust experimental design aims to address parametric uncertainties in the design process. In this paper, the Bayesian robust experimental design is investigated, where both a Monte Carlo sampling strategy and local sensitivity evaluation at each sampling point are employed to achieve the robust solution. The link between global sensitivity analysis (GSA) and the Bayesian robust experimental design is established. It is revealed that a lattice sampling based GSA strategy, the Morris method, can be explicitly interpreted as the Bayesian A-optimal design for the uniform hypercube type uncertainties.

AB - The purpose of model-based experimental design is to maximise the information gathered for quantitative model identification. Instead of the commonly used optimal experimental design, robust experimental design aims to address parametric uncertainties in the design process. In this paper, the Bayesian robust experimental design is investigated, where both a Monte Carlo sampling strategy and local sensitivity evaluation at each sampling point are employed to achieve the robust solution. The link between global sensitivity analysis (GSA) and the Bayesian robust experimental design is established. It is revealed that a lattice sampling based GSA strategy, the Morris method, can be explicitly interpreted as the Bayesian A-optimal design for the uniform hypercube type uncertainties.

KW - bayesian analysis

KW - global sensitivity analysis

KW - modeling

U2 - 10.3182/20100705-3-BE-2011.00096

DO - 10.3182/20100705-3-BE-2011.00096

M3 - Paper

SP - 563

EP - 568

ER -

He F, Yue H, Brown M. Investigating Bayesian robust experimental design with principles of global sensitivity analysis. 2010. Paper presented at 9th International Symposium on Dynamics and Control of Process Systems (DYCOPS 2010), Lueven, Belgium. https://doi.org/10.3182/20100705-3-BE-2011.00096