Background: The AISC prequalifies several rigid connections that ensure acceptable seismic performance for steel moment frames. Some of these connections are included in the Iranian National Building Code, Part 10. In conventional design practice, the connection type is typically ignored in numerical modeling, and all rigid connections are modeled similarly. However, the actual behavior of each connection can create significant differences in lateral stiffness, energy dissipation, and ductility. The aim of this study is to investigate the effect of three rigid connection types, including WUF-W, WFP, and BFP, on the lateral stiffness of steel moment frames through a combined experimental and numerical approach.
Methods: Three full-scale steel moment frames with a span of 2.8 m and height of 1.8 m were constructed. Beam and column sections were considered equivalent to IPE270 and IPB180, respectively, and the strong column-weak beam principle was controlled. The specimens were subjected to the ATC-24 cyclic loading protocol under displacement control. The yield displacement of the frame was 20 mm, and loading continued up to 100 mm. Numerical modeling was performed using Abaqus. C3D8R elements, combined isotropic-kinematic hardening model, and von-Mises yield criterion were used. Welds were simulated using tie constraints, and bolts were modeled using solid elements with prestressing. Numerical results were validated against experimental data. Key stiffness parameters including initial, post-yield, and secant stiffness, as well as cumulative plastic energy and damping, were extracted from hysteresis curves.
Results: The use of top and bottom flange plates in the WFP connection compared to WUF-W increases the initial stiffness by 83% and ultimate strength by 89%. However, it reduces ductility by 9% and damping by 17%. The BFP connection with an initial stiffness of 0.46 and ultimate strength of 21.5 tons shows intermediate performance. In the WUF-W connection, the plastic hinge forms in the panel zone, while in the other two connections, it is transferred to the adjacent column. Comparison of numerical and experimental results shows good agreement with an error of 6% to 12%.
Conclusion: The choice of connection should be based on design priorities. In seismic regions, the WUF-W connection with the highest ductility and damping, and plastic hinge formation in the beam, is recommended. For high stiffness and strength, the WFP connection is more suitable. For repairability and balanced behavior, the BFP connection is suggested.