TY - JOUR
T1 - Crack growth in masonry
T2 - Numerical analysis and sensitivity study for discrete and smeared crack modelling
AU - Bejarano-Urrego, Leidy
AU - Verstrynge, Els
AU - Giardina, Giorgia
AU - Van Balen, Koen
PY - 2018
Y1 - 2018
N2 - One of the most common obstacles faced by engineers when making numerical models to assess damage in historical masonry lies in defining the most suitable constitutive models when there is shortage of either material characterization or experimental data. This paper presents the implementation of a 2D finite element model (FEM) of a masonry wall by means of two strategies: a discrete cracking meso-model and a continuum smeared cracking macro-model. A sensitivity study is performed to investigate the effect of material properties variation on both modelling strategies, each of which considers the highly non-linear behaviour as well as the brittle cracking of the masonry. The numerical models are validated through the results obtained from an experimental testing campaign which considered a brick masonry wall subjected to cyclic three-point bending. The results of both modelling strategies compared with experimental results are presented, as well as the criteria considered for material characterization and the sensitivity analysis. Results indicate the suitability of both models to reproduce experimentally observed load capacity, failure mechanism and horizontal deformations. However, the meso-model showed higher accuracy in terms of failure mechanism and plastic deformations. The sensitivity analysis indicated that some material parameters, such as fracture energy, cohesion and tensile strength, significantly govern the final cracking. This is an important criterion for adequately choosing the parameters for further models in which crack width is considered, e.g. for settlement-induced cracking analysis.
AB - One of the most common obstacles faced by engineers when making numerical models to assess damage in historical masonry lies in defining the most suitable constitutive models when there is shortage of either material characterization or experimental data. This paper presents the implementation of a 2D finite element model (FEM) of a masonry wall by means of two strategies: a discrete cracking meso-model and a continuum smeared cracking macro-model. A sensitivity study is performed to investigate the effect of material properties variation on both modelling strategies, each of which considers the highly non-linear behaviour as well as the brittle cracking of the masonry. The numerical models are validated through the results obtained from an experimental testing campaign which considered a brick masonry wall subjected to cyclic three-point bending. The results of both modelling strategies compared with experimental results are presented, as well as the criteria considered for material characterization and the sensitivity analysis. Results indicate the suitability of both models to reproduce experimentally observed load capacity, failure mechanism and horizontal deformations. However, the meso-model showed higher accuracy in terms of failure mechanism and plastic deformations. The sensitivity analysis indicated that some material parameters, such as fracture energy, cohesion and tensile strength, significantly govern the final cracking. This is an important criterion for adequately choosing the parameters for further models in which crack width is considered, e.g. for settlement-induced cracking analysis.
KW - Discrete crack modelling
KW - Masonry
KW - Model calibration
KW - Numerical analysis
KW - Sensitivity study
KW - Smeared crack modelling
UR - http://www.scopus.com/inward/record.url?scp=85044785247&partnerID=8YFLogxK
U2 - 10.1016/j.engstruct.2018.03.030
DO - 10.1016/j.engstruct.2018.03.030
M3 - Article
AN - SCOPUS:85044785247
SN - 0141-0296
VL - 165
SP - 471
EP - 485
JO - Engineering Structures
JF - Engineering Structures
ER -