|
iSYS R&D(Intelligent SYStem Reliability & Design Laboratory) |
||
|
Home | Professor
|
People
| Research | Publications | Resources |
||
|
|
|
|
|
Research Prognosis & Health Management(PHM) |
Motivation: As the
role of predictive models has increased, the fidelity of computational
results has been of great concern to engineering decision makers in both
government and industry. Model V&V is the primary process for quantifying
and building the fidelity and predictive accuracy in predictive models.
Successful V&V activity can provides decision makers with the information
necessary for making high-consequence decision. To
address a growing concern about the fidelity of the predictive models, this
research aims at a development of guideline and procedures
for conducting a model V&V in a statistical manner. Project Highlight: • Previous research focused on a
development general validation framework in a statistical manner and
integration of this framework to engineering system design and analysis. • In a validation process, our limited
understanding of complex systems leads to building inappropriate predictive
computer models. A hierarchical
model validation procedure was suggested as the validation framework. Figure
below briefly shows a hierarchical model validation framework of cellular
phone model with top-down planning and bottom-up execution activity. • Uncertainty quantification,
model calibration with likelihood function, statistical data processing and
validity check activity are appropriate used in the validation
framework. • To facilitate the computer model
validation process for several different engineering applications, validation
software on Microsoft Excel platform is developed. • Suggested V&V
activity/software are demonstrated with cellular phone problem, thermal
challenge problem and tread block problem. Relevant Publications: 1. Youn B. D.,
Jung B. C., Kim S. B., Seung Y.H., and Kim T.Y. “A Hierarchical Validation of
Predictive Models for Engineered System Design,” 5th China-Japan-Korea Joint
Symposium on Optimization of Structural and Mechanical Systems, Jeju Island,
Korea, 2008 2. Youn B. D.,
Jung B. C., Xi Z., and Kim S. B., “A Hierarchical Model Validation of
Predictive Models for Engineered Systems Design,” IEEE Transactions on
Reliability, Submitted, 2008 |
|
|
|
|
|
