Stochastic modelling of uncertainty in timber engineering. (15th September 2015)

• Record Type: Journal Article
• Authors: Jenkel, Christian; Leichsenring, Ferenc; Graf, Wolfgang; Kaliske, Michael
• Is Part Of: Engineering structures. Volume 99(2015:Sep. 15)
• Page Start: 296
• Page End: 310
• Publication Date: 2015-09-15
• Subjects: Wood mechanics -- Experiments -- Variance of material parameter -- Stochastic modelling -- Fuzzy-randomness; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2015.04.049
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: The variation of material parameters of wood is considered in structural analyses. A data basis for Norway spruce obtained by experiments is utilised. The data is statistically analysed and modelled stochastically. Point estimators and the choice of probability distribution types are discussed. Interval estimation and further data models to describe uncertainty are presented. Abstract: The material properties and parameters of wood are strongly varying due to natural growth and environmental conditions. To include this variation in structural analyses, available data need to be modelled appropriately, e.g. by means of probability. In this publication, a limited empirical data basis for Norway spruce obtained by experiments according to DIN EN 408 and further building codes is utilised. The data is statistically analysed including correlation analyses and statistical tests. On this basis, the material parameters are modelled stochastically. Thereby, several admissible probability distribution types are applied and their influence on the structural results is investigated. Therefore, two exemplary timber structures are analysed, one by means of the FEM and one according to the design rules of DIN EN 1995. The results and the limitations of the data model probability are discussed critically. Finally, means to improve the accuracy of the stochastic description of the data at hand are examined, e.g. by applying interval estimation instead of point estimation methods to determine the distribution parameters. Further data models to describe uncertainty are presented additionally.