The Influence of Correlation Length on the Strength of Stochastic Discrete Models

The Influence of Correlation Length on the Strength of Stochastic Discrete Models

Paper in proceedings (conference paper)

This paper presents a meso-scale discrete model suitable for simulation of concrete fracturing. The model is enhanced by spatial fluctuation of the model parameters related to tensile behavior. The spatial variability of the parameters are represented using random fields. The main objective of the paper is to investigate the effect of correlation length of the random fields on the peak loads simulated by the discrete model. Three point bending of concrete beams is selected for the study presented. In specimens failing at crack initiation from a smooth surface, an interesting dependency is reported. For a correlation length close to a size of the fracture process zone (a zone with a concentrated damage), a significant drop in the peak load occurs. The paper also reports clear dependency of the peak load variability on the correlation length. In specimens containing a deep notch, the average peak load is almost insensitive to the correlation length but the variability seems to be influenced.

This paper presents a meso-scale discrete model suitable for simulation of concrete fracturing. The model is enhanced by spatial fluctuation of the model parameters related to tensile behavior. The spatial variability of the parameters are represented using random fields. The main objective of the paper is to investigate the effect of correlation length of the random fields on the peak loads simulated by the discrete model. Three point bending of concrete beams is selected for the study presented. In specimens failing at crack initiation from a smooth surface, an interesting dependency is reported. For a correlation length close to a size of the fracture process zone (a zone with a concentrated damage), a significant drop in the peak load occurs. The paper also reports clear dependency of the peak load variability on the correlation length. In specimens containing a deep notch, the average peak load is almost insensitive to the correlation length but the variability seems to be influenced.

concrete, lattice-particle model, crack, random field, variability, autocorrelation length, notch, beam, bending, size effect

concrete, lattice-particle model, crack, random field, variability, autocorrelation length, notch, beam, bending, size effect

KADĚROVÁ, J.; ELIÁŠ, J.; VOŘECHOVSKÝ, M.

2016

01.09.2015

Civil-Comp Press

Stirlingshire, United Kingdom

978-1-905088-63-8

Proceedings of the Fifteenth International Conference on Civil, Structural and Environmental Engineering Computing

Civil Comp Proceedings

1759-3433

Civil-Comp Proceedings

108

United Kingdom of Great Britain and Northern Ireland

1

13

13

http://dx.doi.org/10.4203/ccp.108.25