An adaptive sampling method for global sensitivity analysis based on least-squares support vector regression

  • In the field of engineering, surrogate models are commonly used for approximating the behavior of a physical phenomenon in order to reduce the computational costs. Generally, a surrogate model is created based on a set of training data, where a typical method for the statistical design is the Latin hypercube sampling (LHS). Even though a space filling distribution of the training data is reached,In the field of engineering, surrogate models are commonly used for approximating the behavior of a physical phenomenon in order to reduce the computational costs. Generally, a surrogate model is created based on a set of training data, where a typical method for the statistical design is the Latin hypercube sampling (LHS). Even though a space filling distribution of the training data is reached, the sampling process takes no information on the underlying behavior of the physical phenomenon into account and new data cannot be sampled in the same distribution if the approximation quality is not sufficient. Therefore, in this study we present a novel adaptive sampling method based on a specific surrogate model, the least-squares support vector regresson. The adaptive sampling method generates training data based on the uncertainty in local prognosis capabilities of the surrogate model - areas of higher uncertainty require more sample data. The approach offers a cost efficient calculation due to the properties of the least-squares support vector regression. The opportunities of the adaptive sampling method are proven in comparison with the LHS on different analytical examples. Furthermore, the adaptive sampling method is applied to the calculation of global sensitivity values according to Sobol, where it shows faster convergence than the LHS method. With the applications in this paper it is shown that the presented adaptive sampling method improves the estimation of global sensitivity values, hence reducing the overall computational costs visibly.show moreshow less

Download full text files

Export metadata

Metadaten
Document Type:Preprint
Author:M.Sc. Maria Steiner, Jean-Marc BourinetORCiD, Prof. Tom LahmerORCiDGND
DOI (Cite-Link):https://doi.org/10.25643/BAUHAUS-UNIVERSITAET.3832Cite-Link
URN (Cite-Link):https://nbn-resolving.org/urn:nbn:de:gbv:wim2-20181218-38320Cite-Link
Language:English
Date of Publication (online):2018/12/17
Date of first Publication:2018/12/17
Release Date:2018/12/18
Publishing Institution:Bauhaus-Universität Weimar
Institutes and partner institutions:Fakultät Bauingenieurwesen / Professur Informatik im Bauwesen
Pagenumber:33
First Page:1
Last Page:33
Tag:Adaptive sampling method; Global sensitivity analysis; Least-squares support vector regression; Sampling; Surrogate models
GND Keyword:Approximation; Sensitivitätsanalyse; Abtastung
Dewey Decimal Classification:000 Informatik, Informationswissenschaft, allgemeine Werke
600 Technik, Medizin, angewandte Wissenschaften
BKL-Classification:31 Mathematik
54 Informatik
Licence (German):License Logo Creative Commons 4.0 - Namensnennung (CC BY 4.0)
Note:
This is the pre-peer reviewed version of the following article: https://www.sciencedirect.com/science/article/pii/S0951832017311808, which has been published in final form at https://doi.org/10.1016/j.ress.2018.11.015.