A platform for research: civil engineering, architecture and urbanism
FEA Model Errors for the Load Capacity Prediction of Cyclically Loaded Unreinforced Masonry Shear Walls
A finite element analysis (FEA) model was developed for predicting the strength of unreinforced masonry walls subjected to a cyclic, in-plane shear loading. A stochastic FEA was then developed that included the random and spatial variability of tensile bond strength, shear bond strength, masonry unit tensile strength, and bed and perpend joint thickness. This probabilistic assessment allowed the mean and standard deviation of wall capacity to be estimated from Monte-Carlo simulations. Seven distinct stochastic finite element analyses were applied to capture the distinct material properties and structural configurations of sixteen laboratory wall specimens. The subsequent comparison of these numerical predictions with the experimental results allows for the estimation of model error (or model uncertainty), defined as the actual (experimental) capacity divided by the model predicted (FEA) capacity. Consideration of the distinct failure mechanism of each experimental wall specimen and numerical model has been made in order to probabilistically characterise the model error specific to the failure modes of unreinforced masonry shear walls; namely, flexural tension, flexural compression, shear sliding, and diagonal tension failures. Determining the model error associated with a modelling technique is essential to ensuring both that the adopted modelling technique is suitable, and that predictions made from the application of a model are accurate.
FEA Model Errors for the Load Capacity Prediction of Cyclically Loaded Unreinforced Masonry Shear Walls
A finite element analysis (FEA) model was developed for predicting the strength of unreinforced masonry walls subjected to a cyclic, in-plane shear loading. A stochastic FEA was then developed that included the random and spatial variability of tensile bond strength, shear bond strength, masonry unit tensile strength, and bed and perpend joint thickness. This probabilistic assessment allowed the mean and standard deviation of wall capacity to be estimated from Monte-Carlo simulations. Seven distinct stochastic finite element analyses were applied to capture the distinct material properties and structural configurations of sixteen laboratory wall specimens. The subsequent comparison of these numerical predictions with the experimental results allows for the estimation of model error (or model uncertainty), defined as the actual (experimental) capacity divided by the model predicted (FEA) capacity. Consideration of the distinct failure mechanism of each experimental wall specimen and numerical model has been made in order to probabilistically characterise the model error specific to the failure modes of unreinforced masonry shear walls; namely, flexural tension, flexural compression, shear sliding, and diagonal tension failures. Determining the model error associated with a modelling technique is essential to ensuring both that the adopted modelling technique is suitable, and that predictions made from the application of a model are accurate.
FEA Model Errors for the Load Capacity Prediction of Cyclically Loaded Unreinforced Masonry Shear Walls
Lecture Notes in Civil Engineering
Milani, Gabriele (editor) / Ghiassi, Bahman (editor) / Gooch, Lewis J. (author) / Stewart, Mark G. (author) / Masia, Mark J. (author)
International Brick and Block Masonry Conference ; 2024 ; Birmingham, United Kingdom
2024-12-13
12 pages
Article/Chapter (Book)
Electronic Resource
English
Seismic Rehabilitation of Unreinforced Masonry Shear Walls
British Library Conference Proceedings | 2002
|SEISMIC REHABILITATION OF UNREINFORCED MASONRY SHEAR WALLS
British Library Conference Proceedings | 2001
|Capacity of vertically sheared unreinforced masonry walls
British Library Conference Proceedings | 1998
|Cyclic Load Testing of Unreinforced Masonry Walls
NTIS | 1992
|Seismic Fragility Assessment of Unreinforced Masonry Shear Walls
Springer Verlag | 2020
|