نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
Background: Extended end-plate beam-to-column connections are widely used in steel moment frames due to their constructability and reliable structural performance. Their design in codes like Eurocode 3 is based on the component method; however, the accuracy of this approach in predicting the actual connection behavior, particularly in the nonlinear range and under varying geometric parameters, requires systematic evaluation.
Methods: In this study, a comprehensive parametric sensitivity analysis was performed, and the accuracy of component-based methods in predicting the flexural behavior of these connections was evaluated. The connection response was modeled using three independent approaches: the standard component method (EC3), the advanced component method (ACM), and three-dimensional nonlinear finite element analysis in ABAQUS. The numerical model was validated against experimental data, with prediction errors of less than 5% for initial stiffness, design flexural strength, and ultimate moment capacity. Four key geometric parameters were investigated: end-plate thickness (tp), horizontal bolt distance from the plate edge (e₁), column flange thickness (tfc), and vertical bolt arrangement parameters (P₁, Z₁, Z₂).
Results: The standard EC3 method underestimated the ultimate moment capacity by approximately 66%, whereas the ACM reduced this error to about 14%. The parameter e₁ was the most influential, such that increasing e₁ from 15 to 35 mm improved both the initial stiffness and the ultimate moment by about 50% and reduced the stress concentration at the plate edge. Increasing the end-plate thickness from 6 to 14 mm raised the initial stiffness by about 90%. Increasing the column flange thickness beyond a threshold of about 10 mm provided no further improvement due to a shift in the governing mechanism. Moreover, increasing the bolt row spacing in the tensile zone (P₁ from 80 to 170 mm, with corresponding changes in Z₁ and Z₂) reduced the effective internal lever arm and decreased the initial stiffness by up to 54%.
Conclusion: Based on these findings, a two-stage design strategy is proposed, consisting of preliminary parameter selection using the advanced component method followed by final validation through nonlinear numerical analysis. The proposed framework can reduce code-based design conservatism and enable more efficient utilization of the actual connection capacity.
کلیدواژهها English