COMPUTATIONAL MODEL VALIDATION UNDER UNCERTAINTY
Abstract eng: This paper proposes a new method to assess the validity of large-scale computational models by combining system reliability concepts with a Bayesian model validation approach. The concept of Bayesian hypothesis testing is extended to system-level problems where full-scale testing is impossible. This paper shows how component-level validation results may be used to derive a system-level validation measure. This derivation depends on the knowledge of inter-relationships between component modules. The proposed methods are demonstrated for series and parallel systems.
Contributors:
Publisher:
Columbia University in the City of New York
Conference Title:
Conference Title:
15th ASCE Engineering Mechanics Division Conference
Conference Venue:
New York (US)
Conference Dates:
2002-06-02 / 2002-06-05
Rights:
Text je chráněný podle autorského zákona č. 121/2000 Sb.
Record appears in:
Record created 2014-11-19, last modified 2014-11-19