Validation of Sequentially Linear Analysis by simulating structural behaviour of masonry components
C.S. Meiring (TU Delft - Civil Engineering & Geosciences)
Max Hendriks – Mentor
Manimaran Pari – Graduation committee member
Geert J.P. Ravenshorst – Graduation committee member
J.G. Rots – Graduation committee member
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
The unreinforced masonry construction of Dutch terraced houses in Groningen is prone to lateral earthquake loading. To analyse the capacity of these houses, the faculty of Civil Engineering at the TU Delft started a test campaign. In this thesis, the experimental results served as benchmarks to validate a relatively new numerical method: Sequentially Linear Analysis (SLA). This method is developed to overcome numerical instability, which is a problem for regular non-linear finite element analyses (NLFEA). The objective is to investigate to what extent SLA is able to predict the behaviour of the two main components of a Dutch house, namely the shear wall and the out-of-plane loaded transversal wall, during a monotonic pushover test. Both pre-stressed components are modelled, using the (new) non-proportional loading algorithm and two modelling approaches were applied: the smeared crack approach and the discrete crack approach. Besides, two finite element types were used: the new implementation of the SLA code for shell elements is validated by modelling the shear wall with plane stress elements as well.
Results showed that SLA is able to predict the behaviour of pre-stressed masonry components using the non-proportional loading algorithm, except for the post-peak behaviour of the out-of-plane loaded transversal wall. As SLA is under development, the method still has difficulties to overcome. The defined stop criteria were helpful to temporarily deal with the identified problems, but further research needs to be done to actually solve the errors.