More efficient cold-formed steel elements and bolted connections

Modern society is challenged by economic and environmental issues, requiring engineers to develop
more efficient structures. Using cold-formed steel (CFS) frame in construction industry can lead to more sustainable design, since it requires less material to carry the same load compared with other materials. However, the application of CFS structural systems is limited to low story buildings due to the inherent weaknesses of premature buckling behaviour of members and the low ductility of connections. Consequently, current design guidelines of CFS systems are very conservative especially in the case of seismic design. Furthermore, there is no generic optimisation framework for the CFS elements, capable of taking into account both manufacturing/construction constraints and post-buckling behaviour.
This study aims to better understand, to predict, and to optimise CFS elements based on their strength and post-buckling behaviour. The optimised elements can be then included in full-structure modelling to develop more efficient CFS structural connections with high ductility and energy dissipation capacity, suitable for multi-story buildings in seismic regions.
The geometrical dimensions of manufacturable CFS cross-sections were optimised regarding their maximum compressive and bending strength. All the sections were considered to have a fix coil width and thickness while the optimisation was performed based on effective width method suggested in EC3. The optimised solutions were achieved using Particle Swarm Optimisation (PSO) algorithm. The accuracy of the optimisation procedure was assessed using experimentally validated nonlinear Finite Element (FE) analyses accounting for the effect of imperfections To allow for the development of a new ‘folded-flange’ beam cross-section, the effective width method in EC3 was extended to deal with the presence of multiple distortional buckling modes. Improved strength were achieved for CFS elements by using the proposed optimisation framework.
A non-linear shape optimisation method was presented for the optimum design of CFS beam sections based on their post-buckling behaviour. A developed PSO algorithm was linked to the ABAQUS finite element programme for inelastic post-buckling analysis and optimisation. The results also demonstrate that the optimised sections develop larger plastic area, which is particularly important in seismic design of moment-resisting frames.
An experimental programme was carried out at the University of Sheffield to investigate the design and optimisation, considering interactive buckling in cold-formed steel channels under compression and bending. Both standard and optimised sections were tested. The specimen imperfections were measured using a specially designed set-up with laser displacement. Material tests were also carried out to determine the tensile properties of the flat plate and of the cold-worked corners. A total of 36 columns with three lengths and 6 back-to-back beams were completed. The column specimens were tested under a concentrically applied load and with pin-ended boundary conditions while the beams were tested in a four-point bending configuration. Based on the tests, numerical models were proposed and calibrated and the proposed optimisation framework was verified.
A numerical study on the structural behaviour of CFS bolted beam-to-column connections under cyclic loading was presented. An innovative two node element which can take into account the slippage-bearing effects was proposed and implemented using an ABAQUS user defined subroutine. The connection performance in terms of strength, ductility, energy dissipation capacity and damping coefficient were investigated. The effects of bolt configuration, cross-sectional shapes and thicknesses on the connection performance were therefore examined. It is indicated that the proposed numerical model is robust and computationally efficient to simulate the failure modes and moment-rotation response of CFS bolted moment resisting connections.