Efficient uncertainty quantification using a two-step approach with chaos collocation

Conference paper (2006)

Authors

A. Loeven

J.A.S. Witteveen

H. Bijl

Faculty

Aerospace Engineering (Aerospace Engineering) (TU Delft)

Non-intrusive Computational Fluid Dynamics Uncertainty quantification Polynomial chaos Fluid-structure interaction Stochastic collocation

To reference this document use:

http://resolver.tudelft.nl/uuid:b52a1655-621f-4590-9ad5-0206c090d251

More Info

expand_more

Published Date

06-09-2006

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Source:

ECCOMAS CFD 2006: Proceedings of the European Conference on Computational Fluid Dynamics, Egmond aan Zee, The Netherlands, September 5-8, 2006

Faculty

Aerospace Engineering

Abstract

In this paper a Two Step approach with Chaos Collocation for efficient uncertainty quantification in computational fluid-structure interactions is followed. In Step I, a Sensitivity Analysis is used to efficiently narrow the problem down from multiple uncertain parameters to one parameter which has the largest influence on the solution. In Step II, for this most important parameter the Chaos Collocation method is employed to obtain the stochastic response of the solution. The Chaos Collocation method is presented in this paper, since a previous study showed that no efficient method was available for arbitrary probability distributions. The Chaos Collocation method is compared on efficiency with Monte Carlo simulation, the Polynomial Chaos method, and the Stochastic Collocation method. The Chaos Collocation method is non-intrusive and shows exponential convergence with respect to the polynomial order for arbitrary parameter distributions. Finally, the efficiency of the Two Step approach with Chaos Collocation is demonstrated for the linear piston problem with an unsteady boundary condition. A speed-up of a factor of 100 is obtained compared to a full uncertainty analysis for all parameters.

Files

Loeven.pdf

(pdf | 0.264 Mb)