3d Poınt Cloud-Based Imperfectıon Determınatıon of Cold-Formed Steel Members For Numerıcal Modelıng

Abstract

Cold-formed steel (CFS) has been widely used as a construction material in the last decade. However, the overall behavior of individual CFS members is affected by geometric imperfections significantly. In this study, two main strategies are developed to investigate the impact of the CFS members' geometric imperfections. First, texture-mapped point clouds of C and omega-section CFS members with varying dimensions are collected by a three-dimensional scanner. A novel point cloud-based geometric imperfection detection and quantification method that detects both local and global imperfections on CFS members is developed. The obtained results showed that geometric imperfections on CFS members vary along the length of each member significantly, even for the members with identical dimensions. The detected geometric imperfections are then compared to imperfections values/limits reported in previous studies. Second, a mode shape-based imperfection coefficient computation method. Finite element models of CFS members are then generated by considering the ideal geometries; mode shapes are obtained due to eigenvalue elastic buckling analysis. The outputs of the geometric imperfection detection and quantification method are treated as signals, and these signals are decomposed into the obtained mode shapes to compute geometric imperfection coefficients. These imperfection coefficients are later integrated into the numerical model, and the analysis results are obtained. Finally, axial loading tests are conducted on the investigated CFS members. The numerical results and experimental results are then compared. The results showed that the reconstructed mode shapes could not fully represent the exact geometric imperfections. However, the results obtained using mode shape-based geometric imperfection coefficients are generally closer to the experimental results than the common method used in practice.