A numerical-informational approach for characterising the ductile behaviour of the T-stub component. Part 1: Refined finite element model and test validation. (1st January 2015)

• Record Type: Journal Article
• Authors: Fernandez-Ceniceros, J.; Sanz-Garcia, A.; Antoñanzas-Torres, F.; Martinez-de-Pison, F.J.
• Is Part Of: Engineering structures. Volume 82(2015:Jan. 01)
• Page Start: 236
• Page End: 248
• Publication Date: 2015-01-01
• Subjects: Steel -- Bolted connections -- Finite element analysis -- Damage -- Triaxiality; 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.2014.06.048
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: 3D numerical model of a bolted T-stub component performed by ABAQUS. Non-linear damage model for the ductility assessment of the bolted connection. FE model calibration and experimental validation against 18 test specimens. Influence of the damage parameters on the T-stub component's response. Abstract: In the component-based method, the ductility characterisation of the tension component plays a primary role in accurately predicting the rotation capacity of the whole connection. In this paper, a refined finite element (FE) model is presented to assess the complete force–displacement response of the T-stub component, from the initial stiffness up to the fracture point. The refined FE model includes a non-linear continuum damage mechanics model that considers the onset of the damage, the damage evolution law, and the component failure. A parametric study was conducted to evaluate the influence of the damage parameters in the overall response of the T-stub component. The numerical results obtained from the refined FE model strongly agree with the experimental results of 18 test specimens. The comparison reveals the proposed FE model's satisfactory accuracy, with an average fitting error below 9%. This paper constitutes the first part of a comprehensive methodology based on combining FE analysis and soft computing techniques. The ultimate goal of this methodology is to predict the key parameters that define the force–displacement response of the T-stub component.