Accession Number

ADA564357

Title

Experimental Resource Allocation for Statistical Simulation of Fretting Fatigue Problem (Preprint).

Publication Date

Aug 2012

Media Count

17p

Personal Author

C. Dubinsky G. Singh H. R. Millwater P. Golden

Abstract

Estimation of statistical moments from simulation, i.e., mean and standard deviation of an output, may involve large uncertainty caused by the variability in the input random variables. The allocation of resources to obtain more experimental data can reduce the variance of the output moments (mean and standard deviation). The methodology proposed and executed used an optimization method to determine the optimal number of additional experiments required to minimize the variance of the output moments given a constraint. A method to generate the output moments based on the moments of the input variables was implemented. The method used the multivariate tdistribution and the Wishart distribution to generate realizations of the population mean and population covariance of the input variables, respectively. This method was sufficient to handled independent and correlated variables. A fretting fatigue problem was explored to minimize the variance of cyclesto failure mean and standard deviation.

Keywords

Allocations Covariance Estimates Experimental data Failure Fatigue(Mechanics) Fretting Fretting fatigues Multivariate analysis Population(Mathematics) Random variables Resource management Standard deviation Statistics Uncertainty Wishart matrices


Source Agency

Non Paid ADAS

NTIS Subject Category

71N  Nonferrous Metals & Alloys 46E  Structural Mechanics

Corporate Author

Air Force Research Lab., WrightPatterson AFB., OH. Metals Branch.

Document Type

Technical report

Title Note

Interim rept.

NTIS Issue Number

1303

Contract Number

N/A
