22–24 Mar 2021
University of Zagreb Faculty of Civil Engineering, Zagreb, Croatia
Europe/Zagreb timezone

EXPERIMENTAL VERSUS NUMERICAL RESPONSE OF R.C. WALLS SUBJECTED TO EARTHQUAKE LOADING

Not scheduled
20m
VP (University of Zagreb Faculty of Civil Engineering, Zagreb, Croatia)

VP

University of Zagreb Faculty of Civil Engineering, Zagreb, Croatia

Kačićeva 26 10 000 Zagreb
Seismic Performance of Structures

Speaker

Mr Emir Hodžić (University of Sarajevo)

Description

Reinforced-concrete walls have been frequently used as a bearing system for seismic loads in the last half of the century. They partially replaced the traditional European way of building with brick walls, and they are most often used for the construction of residential and business buildings of medium height. The aim of this study is to numerically examine the influence of different reinforcement layouts on the load bearing capacity and ductility of slender shear walls. Three walls experimentally tested at ETH-Zürich under in-plane static cyclic action were used for comparison with finite element models. The walls were designed for different ductility classes according to Eurocode 8. Two types of modelling techniques were made employing SAP2000nl. A frame model with nonlinearity localized in a plastic hinge was made using Takeda and Pivot hysteretic rules. A nonlinear layered shell model was applied using Mander’s definition of confined and unconfined concrete while reinforcement was assumed as hardening plasticity. Advanced numerical analysis on 2D models was performed using DIANAFEA software package assuming smeared cracking approach. Behaviour of concrete was described with total strain rotating crack constitutive law while von Mises yield criterion defines the stress-strain relationship of embedded reinforcement. Plastic hinge models are simple, require short calculation time and predict load-bearing capacity quite well. Shell layered model was mesh dependent and extremely time-consuming resulting in premature failure or mismatch in energy dissipation with respect to experimental findings. Refined numerical modelling using powerful solvers in DIANA proved to be the most successful regarding computational time and feasible mesh densities. Good agreement of dissipated energy, load bearing capacity and crack layout with test results was obtained.

Keywords reinforced-concrete wall, earthquake, capacity design method, nonlinear analysis
DOI https://doi.org/10.5592/CO/1CroCEE.2021.106

Primary authors

Mr Emir Hodžić (University of Sarajevo) Dr Senad Medic (Faculty of Civil Engineering University of Sarajevo) Prof. Mustafa Hrasnica ( Faculty of Civil Engineering at University of Sarajevo)

Presentation materials