You are here
Uncertainty Quantification (UQ) in Computational Science Workshop
I would like to inform you that the Computational Science and Engineering at the University of Illinois at Urbana-Champaign is organizing a workshop on Uncertainty Quantification (UQ) in Computational Science on June 11-12, 2007. There is no registration fee for this workshop !!!
Please visit: http://www.cse.uiuc.edu/uq/
| Attachment | Size |
|---|---|
 UQ_workshop_poster.pdf | 1.62 MB |
»
- karelmatous's blog
- Log in or register to post comments
- 6531 reads
Comments
sensitivity analysis and optimization
Uncertainty quantification (UQ) has received a great deal of attention within the ASCI community (e.g., C-SAFE http://imechanica.org/node/786#comment-2760, CSAR http://imechanica.org/node/786#comment-2740) due to its potential for use as a validation tool for the large multi-scale, multi-physics application codes being developed.
http://csmr.ca.sandia.gov/projects/sequoia/
For the simulations of accidental fires and explosions, the equations modeling ignition, fire spread, and other phenomena of interest depend on parameters. These may include reaction rate constants, activation energies, thermodynamic constants, and transport coefficients as well as initial, boundary, or other auxiliary conditions. They are seldom known to high precision, leading to uncertainty in outcomes predicted by the model.
Sensitivity analysis is aimed at quantifying the influence of parameters on model behavior. Goals of sensitivity analysis that are important here include assessing the effects of parameter uncertainty on predicted outcomes and determining the accuracy with which parameters must be evaluated for desired model reliability.
This information will be helpful in deciding which detailed components should be included in the integrated models.