Random and gradient based fields in discrete particle models of heterogeneous materials

conference paper

English

The proper characterization of the microstructure and macroscopic properties of random heterogeneous materials may help to interpret the observed scatter in many engineering problems. In this contribution, a lattice discrete particle model is presented in a civil engineering framework with a special emphasis on concrete structures and infrastructure. The implications of spatial variability are investigated with regard to the observed scatter in standard concrete tests. In particular, classical experiments, such as three-point bending tests, may exhibit various levels of scatter which have to be accounted for if material characteristics are to be derived from such experiments and consequently used for the analysis and design of structural systems. Therefore, random and gradient based fields are combined in this paper in an attempt to realistically capture the material properties and associated variability stemming from the microstructure and macroscopic features such as the placement of coarse aggregates. Various mechanical and statistical aspects of simulated test series are investigated, such as macroscopic fracture energy or distribution of load bearing capacity. The presented application example of three-point bend specimens incorporates not only the inherent spatial variability owing to the placement of aggregates, modelled by random fields, but also various production artefacts, such as concrete compacting characterized here by gradient based fields.

LDPM, random fields, gradient-based fields, spatial variability, coarse aggregates.

PODROUŽEK, J.; VOREL, J.; WENDNER, R.

1. 8. 2017

Eccomas Proceedia

Rhodes Island

978-618-82844-4-9

Uncertainty Quantification in Computational Sciences and Engineering

605

615

11