Buckling Load Estimation Using Multiple Linear Regression Analysis and Multigene Genetic Programming Method in Cantilever Beams with Transverse Stiffeners

Analytical methods cannot find the exact solution for inelastic lateral torsional buckling. This study aims to develop innovative solutions by creating closed-form equations. A series of numerical studies (ANSYS) have been conducted for buckling load calculation on European I-section cantilever beams reinforced with transverse stiffener plates at different intervals. Formulations were developed using two methods to estimate the found load values more practically. Multiple linear regression analysis and multigene genetic programming methods were used to obtain these formulations. According to the error statistics, the multigene genetic programming method gave more accurate results than the multiple linear regression analysis methods in buckling load estimation. The estimates obtained from the multigene genetic programming method and the numerical results calculated in the ANSYS program were found to be compatible with each other. The scientific novelty brought by the research is to develop more original formulations for cantilever beams instead of using the buckling load calculation described for simply supported beams in the specifications. The scientific difference is that the developed formulations can calculate in a way that can consider the contribution of transverse stiffeners to the buckling load. This study will show that formulations designed with computer technologies can be an alternative calculation method for estimating the lateral buckling load according to the transverse stiffener plate spacing for European I-section cantilever steel beams.