Towards optimal strength design of cold roll formed steel structural members considering manufacturing process effects

Abstract

Cold-formed steel (CFS) materials are increasingly used in a wide range of applications in building construction. CFS structural members possess many advantages, most notably high load capacity-to-weight ratio and high speed of construction. To design CFS members accurately and gain optimal structural performance, the two factors, namely ‘geometry effect’ and ‘manufacturing effect’, must be included in design procedures. However, whilst the first has been widely investigated and implemented in design, the latter has not been much studied and considered in the current design practices. This research aims to develop and validate an optimal strength design approach that takes into consideration key ‘geometry’ and ‘manufacturing process’ effects on the material and structural properties into the design of CFS structural members. The cross-sectional shapes considered in this research were channel and zed sections as they are widely used in building construction applications. Numerical modelling and physical testing of the material during the manufacturing process were carried out and implemented to examine the behaviour and design of CFS structural members subjected to load in building applications. Different Finite Element (FE) model arrangements and methods to predict the buckling and nonlinear buckling analyses were tested. The FE models were assessed and validated against experimental data from literature studies, with an excellent degree of comparability. The models were then utilised to perform comprehensive parametric studies and optimise CFS sections. A comprehensive parametric study for longitudinally stiffened channel and zed beam sections under distortional bending was conducted to investigate the effects of a stiffener’s properties on the section strength including its position, shape, size and material properties by the cold work at bends. Limits for optimal design of the sections were suggested. The suitability of a design method, the Direct Strength Method (DSM), in predicting the ultimate moment capacity for CFS beam sections was assessed using the FE analyses results. The DSM predictions were found significantly cross-sectional dependent, especially in the sections where the tip of web stiffeners shifted away from the web in horizontal direction failed by distortional-global buckling (D-G), providing more accurate predictions for certain cross-sections than for others. Shortcomings were confirmed and suggestions for improvements were given, especially the inclusion of the D-G in the DSM design guideline. This may confirm that the design methods previously used for optimisation are somewhat simplistic and the reported optimisation results not correct in their predictions. A new, more sophisticated practical design approach has been developed by combining detailed nonlinear FE modelling and optimisation. It accounted for all possible buckling failures considering the effects of key ‘geometry’ and ‘manufacturing process’ factors, such as initial geometric imperfections and work-hardening introduced by the cold-rolling process. The validated FE model innovatively combined with the optimisation algorithms using integrated Design Of Experiment, response surface methodology and multi-objective genetic algorithm. The optimisation approach was devised to achieve the optimal design shape of the channel and zed sections under distortional bending. Some significant improvements have been obtained in distortional buckling and ultimate bending strength of the optimal cross-section shapes, compared to the original sections, without increasing the amount of the material used. Experimental measurements and testing of the materials were carried out to investigate the change of material properties during the manufacturing process, and their results were used for the validation of numerical simulations. The experimental programme has been fully described within this research, including techniques implemented, data generated, and analysis methods adopted. The results of the current test programme were used to investigate the cold work effects on the corner and stiffener bend regions of CFS sections and the accuracy of existing predictive models was evaluated. The results were then used to accurately quantify the cold work effects on the bending strength of the CFS sections. It has been revealed that the optimised sections achieved were less prone to distortional buckling failure and have gained significant section strength benefit from the cold work effects. It was also shown that the effects of geometry and the manufacturing process had to be carefully considered to design CFS members accurately and gain optimal structural performance. Recommendations for further research are also proposed.