Journal ArticleDOI
Uncertainty Analysis of a Dynamic Model of a Novel Remotely Piloted Airship
Reads0
Chats0
TLDR
In this paper, a probabilistic Bayesian uncertainty analysis was performed, whereby the output of a model is sampled over a set of input points and the response of the model is emulated by a Gaussian process, allowing a reliable uncertainty analysis to be performed for considerably less computational expense than conventional Monte Carlo methods.Abstract:
As the sophistication of finite element models increases, the need to investigate the effects of uncertainties in model inputs becomes increasingly important in the interests of developing a robust model. A novel design of an unmanned airship is under development at the Politecnico di Torino, Italy. Structural and fluid. structure interaction models were developed to ascertain the response of the structure under operational and aerodynamic loading; however, uncertainties in material properties (due to variation with temperature), pressure, and aerodynamic loading necessitated an uncertainty analysis. Since the computational expense of a large finite element model makes Monte Carlo methods impractical, a probabilistic Bayesian uncertainty analysis was performed, whereby the output of the model is sampled over a set of input points and the response of the model is emulated by a Gaussian process. Quantities pertaining to uncertainty analysis were then inferred, allowing a reliable uncertainty analysis to be performed for considerably less computational expense than conventional Monte Carlo methods. The results showed that uncertainty in model stress and deflection is indeed significant, although the highly nonlinear response of the fluid. structure interaction model required a high sampling density in order to accurately emulate its response. Copyright © 2011 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.read more
Citations
More filters
Journal ArticleDOI
Bayesian sensitivity analysis of a nonlinear finite element model
TL;DR: The work here investigates the use of an emulator known as a Gaussian process (GP), which is an advanced probabilistic form of regression, which is particularly suited to uncertainty analysis since it is able to emulate a wide class of models, and accounts for its own emulation uncertainty.
Journal ArticleDOI
Empirical Assessment of Deep Gaussian Process Surrogate Models for Engineering Problems
Dushhyanth Rajaram,Tejas G. Puranik,S. Ashwin Renganathan,WoongJe Sung,Olivia Pinon Fischer,Dimitri N. Mavris,Arun Ramamurthy +6 more
TL;DR: In recent years, multilayered hierarchical compositions of the well-known and widely used Gaussian process models called deep Gaussian processes are finding use in the approximation of black-box fu...
Journal ArticleDOI
An Unmanned Lighter-Than-Air Platform for Large Scale Land Monitoring
TL;DR: In this article, the authors presented the concept and preliminary design of an unmanned lighter-than-air (LTA) platform instrumented with different remote sensing technologies, which includes equipment for high-definition visual, thermal, and hyperspectral imaging as well as LiDAR scanning.
References
More filters
Book
Gaussian Processes for Machine Learning
TL;DR: The treatment is comprehensive and self-contained, targeted at researchers and students in machine learning and applied statistics, and deals with the supervised learning problem for both regression and classification.
Journal ArticleDOI
The design and analysis of computer experiments
TL;DR: This paper presents a meta-modelling framework for estimating Output from Computer Experiments-Predicting Output from Training Data and Criteria Based Designs for computer Experiments.
Journal ArticleDOI
Probabilistic sensitivity analysis of complex models: a Bayesian approach
Jeremy E. Oakley,Anthony O'Hagan +1 more
TL;DR: In this article, the authors present a Bayesian framework which unifies the various tools of probabilistic sensitivity analysis, which allows effective sensitivity analysis to be achieved by using far smaller numbers of model runs than standard Monte Carlo methods.
Reference EntryDOI
Arbitrary Lagrangian–Eulerian Methods
TL;DR: In this paper, the authors provide an in-depth survey of arbitrary Lagrangian-Eulerian (ALE) methods, including both conceptual aspects of the mixed kinematical description and numerical implementation details.
Related Papers (5)
Characterization of Dynamic Response of Structures With Uncertainty by Using Gaussian Processes
Z. Xia,Jiong Tang +1 more