3rd Croatian Conference on Earthquake Engineering 3CroCEE

In many countries, such as Montenegro, numerous existing reinforced concrete (RC) buildings were designed and built before the adoption of modern seismic codes. These buildings make up a significant portion of Montenegro's RC building stock, making their assessment, retrofitting, and strengthening an urgent research topic. This shortcoming because of insufficient transverse reinforcement, poor energy dissipation, and plastic hinges in critical areas, making shear failures in RC beam-column joints one of the most common structural failure types.
Examining and modelling these RC joints and comparing results with experimental data is essential for advancing seismic resilience. The primary goal of this paper was to improve understanding of RC joints, focusing on their vulnerability and critical role in structural integrity. Although these joints represent a local failure point, their breakdown can lead to the collapse of the entire structure. Another aim is to create a foundation for further research, focusing on numerical and experimental analyses of strengthened/retrofitted RC joints.
A nonlinear static analysis of an RC joint was conducted using ANSYS software (version 24.2) with selected material models (CPT215 and REINF264) representing the nonlinear behaviour of concrete and reinforcement, respectively. The load was applied as displacement at the beam end in two steps, with a constant axial force in the column. Vertical force-displacement diagrams were compared with experimental results and those in the ANSYS guide (version 19.1) to verify the convergence and accuracy of results.
Mesh density was varied, using finite elements (FE) of different sizes and shapes, including hexahedral and prismatic forms. Results showed that element size significantly affects numerical accuracy. Additionally, prismatic elements with triangular base achieving faster convergence. The runtime analysis using different mesh densities and finite element types underscored how mesh and element selection impact the balance between modeling accuracy and computational efficiency.