Modelagem numérica de uma ligação viga-pilar semirrígida em concreto pré-moldado utilizando cantoneiras soldadas na viga e no pilar

Abstract

In precast concrete structures, the proper design and execution of the beam-to-column connection are essential to ensure overall structural performance. The determination of the stiffness of this connection can be carried out through experimental, analytical, or numerical methods. The use of numerical methods to analyze the performance of beam-to-column connections is highly efficient for understanding the behavior of the connection components. The beam-to-column connection experimentally investigated by Silva (2025) presents several particular characteristics, such as the use of welded angle plates and a relatively low beam height. Therefore, the main objective of this study is to perform a numerical modeling of a semi-rigid beam-to-column connection in precast concrete using welded angles at both the beams and the column. With the aid of the Abaqus software and employing the finite element method, a numerical model was developed and validated through comparison with the results of the reference experimental test. The analysis involved comparing the load–displacement curve, the load–strain curves for the reinforcing bars and angle plates, and the moment–rotation curve of the connection. Regarding the final displacements, the percentage difference was 11.44%, while for the secant stiffness the difference was 18.56%. All curves exhibited similar behavior to those obtained experimentally. Aparametric study was conducted by varying the diameter of the continuity reinforcement to 20 mm,16mm,and12.5 mm. It was observed that smaller bar diameters led to more flexible connections, resulting in reduced stiffness, with a decrease of 39.29% in secant stiffness between the largest and smallest bar diameters. The equation from NBR 9062:2017 was calibrated to determine the secant stiffness under negative bending in beam-to-column connections with continuity reinforcement. Using the experimental model as a reference, the parameters k and Led were obtained for the connection. These values were then applied to calculate the stiffness for the reinforcement diameters used in the parametric study, and the results were compared with those from the numerical method. The differences between both methods were small, ranging from a maximum of 20.41% to a minimum of 3.52%. Therefore, the numerical modeling assumptions proved to be effective in representing the behavior of this connection, the parametric analysis was consistent, and the equation calibration yielded satisfactory results.