Published October 22, 2012 | Version 11030
Journal article Open

Optimization of Parametric Studies Using Strategies of Sampling Techniques

Description

To improve the efficiency of parametric studies or tests planning the method is proposed, that takes into account all input parameters, but only a few simulation runs are performed to assess the relative importance of each input parameter. For K input parameters with N input values the total number of possible combinations of input values equals NK. To limit the number of runs, only some (totally N) of possible combinations are taken into account. The sampling procedure Updated Latin Hypercube Sampling is used to choose the optimal combinations. To measure the relative importance of each input parameter, the Spearman rank correlation coefficient is proposed. The sensitivity and the influence of all parameters are analyzed within one procedure and the key parameters with the largest influence are immediately identified.

Files

11030.pdf

Files (190.9 kB)

Name Size Download all
md5:849cf4f62cdfda281409408e859566c9
190.9 kB Preview Download

Additional details

References

  • A. Florian, "An Efficient Sampling Scheme: Updated Latin Hypercube Sampling," J. Probabilistic Engineering Mechanics, vol. 7, no. 2, 1992, pp. 123-130.
  • M. McKay, R. J. Beckman, and W. J. Conover, "A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code," Technometrics, no. 2, 1979, pp. 239-245.
  • R. L. Iman, and W. J. Conover, "A Distribution-Free Approach to Inducing Rank Correlation Among Input Variables," Commun. Statist., no. B11, 1982, pp. 311-334.
  • ANSYS Academic Research, Release 13, Help System, Theory Reference Guide for Mechanical APDL, ANSYS, Inc., 2011.