Search:
Computing and Library Services - delivering an inspiring information environment

Bayesian cross-entropy methodology for optimal design of validation experiments

Jiang, Xiang and Mahadevan, S. (2006) Bayesian cross-entropy methodology for optimal design of validation experiments. Measurement Science and Technology, 17. pp. 1895-1908. ISSN 0957-0233

Metadata only available from this repository.

Abstract

An important concern in the design of validation experiments is how to incorporate the mathematical model in the design in order to allow conclusive comparisons of model prediction with experimental output in model assessment. The classical experimental design methods are more suitable for phenomena discovery and may result in a subjective, expensive, time-consuming and ineffective design that may adversely impact these comparisons. In this paper, an integrated Bayesian cross-entropy methodology is proposed to perform the optimal design of validation experiments incorporating the computational model. The expected cross entropy, an information-theoretic distance between the distributions of model prediction and experimental observation, is defined as a utility function to measure the similarity of two distributions. A simulated annealing algorithm is used to find optimal values of input variables through minimizing or maximizing the expected cross entropy. The measured data after testing with the optimum input values are used to update the distribution of the experimental output using Bayes theorem. The procedure is repeated to adaptively design the required number of experiments for model assessment, each time ensuring that the experiment provides effective comparison for validation. The methodology is illustrated for the optimal design of validation experiments for a three-leg bolted joint structure and a composite helicopter rotor hub component.

Item Type: Article
Subjects: T Technology > T Technology (General)
T Technology > TJ Mechanical engineering and machinery
T Technology > TA Engineering (General). Civil engineering (General)
Q Science > QC Physics
Schools: School of Computing and Engineering
School of Computing and Engineering > Centre for Precision Technologies
School of Computing and Engineering > Centre for Precision Technologies > Surface Metrology Group
Related URLs:
Depositing User: Briony Heyhoe
Date Deposited: 26 Mar 2008 16:19
Last Modified: 02 Dec 2010 12:30
URI: http://eprints.hud.ac.uk/id/eprint/596

Item control for Repository Staff only:

View Item

University of Huddersfield, Queensgate, Huddersfield, HD1 3DH Copyright and Disclaimer All rights reserved ©