<  Back to the Polytechnique Montréal portal

Probabilistic modeling of heteroscedastic laboratory experiments using Gaussian process regression

Lucie Tabor, James-A. Goulet, Jean-Philippe Charron and Clelia Desmettre

Article (2018)

[img]
Preview
Accepted Version
Terms of Use: All rights reserved.
Download (1MB)
Cite this document: Tabor, L., Goulet, J.-A., Charron, J.-P. & Desmettre, C. (2018). Probabilistic modeling of heteroscedastic laboratory experiments using Gaussian process regression. Journal of Engineering Mechanics, 144(6), p. 1-10. doi:10.1061/(asce)em.1943-7889.0001466
Show abstract Hide abstract

Abstract

This paper proposes an extension to Gaussian process regression (GPR) for data sets composed of only a few replicated specimens and displaying a heteroscedastic behavior. Because there are several factors that are out of the control of experimenters, it is often impossible to reproduce identical specimens for a same experiment. Moreover, observations from laboratory experiments typically display a heteroscedastic interspecimen variability. Because experiments and specimen manufacturing are expensive, it is uncommon to have more than three specimens to build a model for the observed responses. The method proposed in this paper uses GPR to predict each tested specimen using a shared prior structure and models the global heteroscedastic behavior by combining observations using conjugate prior distributions. An application of the method to high-performance fiber-reinforced concrete experiments highlights fiber addition benefits for reducing water permeability caused by macrocracks.

Open Access document in PolyPublie
Subjects: 1000 Génie civil > 1000 Génie civil
Department: Département des génies civil, géologique et des mines
Research Center: Non applicable
Date Deposited: 19 Apr 2018 12:44
Last Modified: 24 Oct 2018 16:12
PolyPublie URL: https://publications.polymtl.ca/3026/
Document issued by the official publisher
Journal Title: Journal of Engineering Mechanics (vol. 144, no. 6)
Publisher: ASCE
Official URL: https://doi.org/10.1061/(asce)em.1943-7889.0001466

Statistics

Total downloads

Downloads per month in the last year

Origin of downloads

Dimensions

Repository Staff Only