Residual strength estimation of damaged steel tubular columns using digital image correlation

Tubular structural steel members are extensively used in modern constructions due to their outstanding structural efficiency, improved capacity against torsion and buckling, and aesthetic appearance. Steel tubular members may deform due to unanticipated loading conditions and accidental damage by wrong handling while transporting to the construction site, which induces damage leading to large permanent deformations. The ultimate capacity of the steel members is reduced due to permanent deformations. In such a case, the residual strength assessment of the member becomes critical before using it in structural applications. This study proposes a methodology by incorporating the Digital Image Correlation (DIC) technology to capture the whole field deformations and deformed shapes and model the 3D damaged shape of the specimen. Initially, the surface details in cartesian coordinates of the deformed specimen are obtained by scanning using a set of two cameras in a 3D DIC setup. The scanned images are stitched to obtain the complete 3D surface coordinates of the specimen. Further, the data is processed by regularising nodes, surface smoothening using 2D Inverse Fourier Transform, and converting the topographical information into the surface point cloud data or nodes, which is used to generate a mesh. The mesh is imported into Finite Element Analysis (FEA) software like Abaqus, and the damaged specimen's residual strength is predicted. The residual strengths of two (one circular and one square) damaged tubes were predicted using the proposed methodology and validated with the experimental results.