Non-linear simulation of shaking-table tests on 3- and 7-storey X-Lam timber buildings. (15th April 2016)

• Record Type: Journal Article
• Authors: Rinaldin, G.; Fragiacomo, M.
• Is Part Of: Engineering structures. Volume 113(2016:Apr. 15)
• Page Start: 133
• Page End: 148
• Publication Date: 2016-04-15
• Subjects: X-Lam timber structures -- FE modelling -- Seismic analysis -- Cyclic behaviour -- Connections -- Numerical simulation; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2016.01.055
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: An advanced FE non-linear modelling of X-Lam timber buildings is presented. Shaking-table tests of X-Lam timber buildings were simulated with good results. Parametric analyses on friction effect and strength domain in fasteners are shown. Abstract: This paper presents an advanced FE modelling of cross-laminated (X-Lam) timber buildings for non-linear dynamic analyses. The model has been used to reproduce the experimental results of the shaking table tests carried out in Japan within the SOFIE project on the 3- and 7-storey full-scale timber buildings. The X-Lam timber panels have been schematized with linear-elastic shell elements, whereas all metal connectors (hold-downs, angle brackets, screws) have been described with 3-DOFs non-linear hysteretic springs. The hysteretic law has a trilinear backbone curve, and is characterised by pinching, post-peak softening, strength and stiffness degradation. The approximating hysteretic laws of the springs have been calibrated on the experimental cyclic tests carried out on each single metal connector. Additional features of the model are the possibilities to account for friction at the interface between upper and lower X-Lam panels, and for a strength domain between shear and tensile force in the metal connectors. Due to the lack of experimental results, these variables have been identified via parametric study so as to reduce the difference between the numerical prediction and the experimental result of X-Lam single walls loaded with cyclic horizontal load. The experimental–numerical comparisons of the shaking table tests demonstrate the capacity of the model to capture the seismic responses of both buildings with errors within 20% in relative acceleration and 7% in roof displacement. Friction has been found to significantly affect the seismic response as it reduces the peak top displacement up to 31%.