Optimization of Bolted Steel T-Stub Connection Based on Nonlinear Finite Element Analysis Using Genetic Algorithm

Abstract

The equivalent T-stub method is frequently employed in infrastructure projects, including bridge engineering, to simplify bolted connection analysis. However, steel connections remain inherently complex due to nonlinear behavior, cost considerations, and code compliance, framing the design process as a discrete structural optimization problem. This research addresses these challenges by presenting a comprehensive calculation framework that combines the finite element method (FEM) and genetic algorithm (GA) to accurately evaluate the structural performance of bolted T-stub configurations. The proposed approach accounts for nonlinear behavior, thereby reflecting realistic structural responses. To enhance the simulation efficiency and reduce the computational time without significantly compromising accuracy, the study introduces a simplified modeling methodology. The effectiveness of the approach is demonstrated through the development and experimental validation of a selected T-stub connection. Furthermore, a parameter sensitivity analysis is conducted to showcase the range of possible outcomes, emphasizing the potential for optimization. Finally, the proposed connections were optimized using GA, highlighting the benefits of structural optimization in achieving efficient and precise designs for steel connections.

Keywords

• T-stub steel connection
• finite element method
• nonlinear analysis
• genetic algorithm