Efficient Modelling of a Nonlinear Gust Loads Process for Uncertainty Quantification of Highly Flexible Aircraft
Year: 2018
Type: conference-paper
Abstract: An efficient gust loads process is presented which can predict worst case loads on a highly flexible aircraft, for use in uncertainty quantification. This process generates a neural network surrogate model of an aeroelastic system using linearised system equations to rapidly determine worst gust cases, including the capability to model atypical gust excitations. The surrogate model is used to produce a reduced set of identified gust cases which cause the largest loads; these cases are then run in the nonlinear code. Uncertainty quantification of this gust process is carried out using polynomial chaos expansion (PCE) techniques, considering uncertain structural properties. Convergence studies of the gust loads using PCE indicate significantly fewer samples are required than would be for a Monte Carlo simulation. It is demonstrated how oblique gusts exceed the loads envelope from a traditional gust process, justifying the need to consider alternative gust excitations, but interestingly for the particular test case in this work, uncertain structural properties can be seen to have little effect on the uncertainty of the static g and gust loads.
DOI: 10.2514/6.2018-1681
Authors: Robert G. Cook, Christopher Wales, Ann Gaitonde, Dorian Jones, Jonathan E. Cooper
Institution: University of Bristol
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2781501512