TU Darmstadt / ULB / TUprints

A blackbox yield estimation workflow with Gaussian process regression applied to the design of electromagnetic devices

Fuhrländer, Mona ; Schöps, Sebastian (2022):
A blackbox yield estimation workflow with Gaussian process regression applied to the design of electromagnetic devices. (Publisher's Version)
In: Journal of Mathematics in Industry, 10, Springer Nature, e-ISSN 2190-5983,
DOI: 10.26083/tuprints-00021111,
[Article]

[img] Text
s13362-020-00093-1(1).pdf
Available under: CC BY 4.0 International - Creative Commons, Attribution.

Download (1MB)
Item Type: Article
Origin: Secondary publication via sponsored Golden Open Access
Status: Publisher's Version
Title: A blackbox yield estimation workflow with Gaussian process regression applied to the design of electromagnetic devices
Language: English
Abstract:

In this paper an efficient and reliable method for stochastic yield estimation is presented. Since one main challenge of uncertainty quantification is the computational feasibility, we propose a hybrid approach where most of the Monte Carlo sample points are evaluated with a surrogate model, and only a few sample points are reevaluated with the original high fidelity model. Gaussian process regression is a non-intrusive method which is used to build the surrogate model. Without many prerequisites, this gives us not only an approximation of the function value, but also an error indicator that we can use to decide whether a sample point should be reevaluated or not. For two benchmark problems, a dielectrical waveguide and a lowpass filter, the proposed methods outperform classic approaches.

Journal or Publication Title: Journal of Mathematics in Industry
Volume of the journal: 10
Place of Publication: Darmstadt
Publisher: Springer Nature
Collation: 17 Seiten
Classification DDC: 500 Naturwissenschaften und Mathematik > 510 Mathematik
600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften
Divisions: 18 Department of Electrical Engineering and Information Technology > Institute for Accelerator Science and Electromagnetic Fields > Computational Electromagnetics
18 Department of Electrical Engineering and Information Technology > Institute for Accelerator Science and Electromagnetic Fields
Date Deposited: 08 Apr 2022 11:56
Last Modified: 22 Aug 2022 13:18
DOI: 10.26083/tuprints-00021111
Corresponding Links:
URN: urn:nbn:de:tuda-tuprints-211115
Additional Information:

Keywords: Yield analysis; Failure probability; Uncertainty quantification; Monte Carlo; Gaussian process regression; Surrogate model; Blackbox

URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/21111
PPN: 493433422
Export:
Actions (login required)
View Item View Item