Application of Monte Carlo method for the reliability treatment of wind load variables using Bayesian hierarchical models
Abstract eng: This paper discusses the probabilistic approaches and methods used in the reliability modeling of wind load components, specifically when considering variables subject to multiple sources of uncertainty. The use of the Bayesian approach to model such variables in hierarchical models is discussed. A numerical Monte Carlo method is presented as an alternative to analytical solutions such as conjugate functions when determining posterior predictive distribution parameters for hierarchical models. The application of the method to the three primary wind load components, namely free-field wind, pressure coefficients and terrain roughness factors, is presented.
Contributors:
Publisher:
Taylor and Francis Group, London, UK
Conference Title:
Conference Title:
Sixth International Conference on Structural Engineering, Mechanics and Computation
Conference Venue:
Cape Town, South Africa
Conference Dates:
2016-09-05 / 2016-09-07
Rights:
Text je chráněný podle autorského zákona č. 121/2000 Sb.
Record appears in:
Record created 2016-09-20, last modified 2016-09-20
guest ::
