Homogeneous Orthotropic Masonry Material Model

Homogeneous Orthotropic Masonry Material Model: Research, development and implementation for explicit analysis

Aukselis, A.

Rots, J.G. (mentor)
Hendriks, M.A.N. (mentor)
Yang, Y. (mentor)
Hoorn, H. (mentor)
Koot, A. (mentor)

Civil Engineering and Geosciences

Structural Engineering

2017-01-18

The Groningen region, in the Netherlands, experiences earthquakes since 1986. These earthquakes are new to the region and the structures built there are not designed for it. Therefore, they pose a great risk for the historical structures and people living in the area. Most of the buildings there are built using masonry and to analyze such structures on how they will be affected by earthquakes using currently available tools and methods is rather complicated. In this report, the author investigates the use of available continuum damage mechanics models for the purpose of simulating orthotropic masonry behavior during an explicit integration analysis. The material behavior is described in Fortran programming language and is used as a custom user subroutine (VUMAT) in Simulia Abaqus finite element analysis software. The developed material model exhibits 3D elasticity and 2D plane plasticity. Furthermore, it is assumed that two general failure mechanisms are present. One associated with tensile and shear brittle fracture represented by Rankine type yield surface and other with distributed crushing of a material represented by Hill type yield surface. The model exhibits uncoupled damage evolution in the tension regime and coupled in compression. Additionally, the model supports tensile crack closure, while in compression it accumulates the plastic deformations and if an element is crushed it can be flagged for deletion from the mesh. The model is formulated in such a way that most of the properties in material directions are independent of one another. The developed model was tested by examining its behavior in analyses where numerical models were composed out of one or few elements. Additionally, for experimental comparison, four shear walls were modelled, three subjected to monotonic loading and one to cyclic. The analyses closely agree to experimental results even when using raw test data. The material model is stable due to the explicit approach and provides qualitative results as it is flexible enough to be used for various types of analyses, either static or cyclic.In the final part of the report, further developments are considered, including improvements to the code base, additional testing, and development of a custom element.

material model
Masonry
FEM
damage mechanics
earthquakes
Groningen
Fortran
Python

http://resolver.tudelft.nl/uuid:b0ca8953-de6f-49b1-9692-3e7bb1b1f687

Student theses

master thesis

(c) 2017 A. Aukselis