Search

Cold formed steel members are widely applied in light roofing systems, mainly with trussed structural solutions. Developed from thin steel sheets (usually of around one millimeter thickness) stiffened trapezoidal CFS can be considered as an alternative for large span roofing, allowing the assembling of an orthotropic roofing system. In order to improve its flexural buckling behavior, these thin-walled sections must be designed with intermediate stiffeners and quite large possibilities can be achieved in order to (i) improve the cross-section strength under bending moment, (ii) simplify and improve the stiffeners arrangement, (iii) maximize the roof covering width, (iv) develop the roofing system for long span bridging and (v) design with as thin as possible steel sheets. Considering all these conditions, the general cross-section shape can be infinitively modified adopting different stiffener geometries, dimensions, distribution, as well as combining distinct stiffeners along the cross-section. Rational systematic choices and buckling results allowed identifying the “best” solutions, taking advantage of the shape grammar contribution combined with finite strip method buckling analysis. The obtained results revealed the key geometric parameters that must be considered to perform effective improvement of roofing trapezoidal CFS.

This paper formalizes the domain of all and only manufacturable open cold-formed steel (CFS) profiles aimed at improving their strength capability in the development of new products. A Genetic Algorithm is applied over this domain to get a set of best solutions from structural and constructive viewpoints. A basic review is carried out with regard to (i) Shape Grammar, (ii) Finite Strip Method-based elastic buckling analysis, (iii) Direct Strength Method-based structural strength computation and (iv) modified procedure of the Non-dominated Sorting Genetic Algorithm. CFS-manufacturing is defined as a language and its general grammar is presented. This grammar-based model responds to requisites of descriptive power, simplicity of implementation, and customization to specific needs. The CFS grammar has been implemented using scripting languages available in pre-existing CAD and CAE applications. Implementation includes generation, visualization, analysis and improvement capabilities. Results show that innovative shapes with improved strength could be generated from the finite set of allowable manufacturing grammar rules encoded by the grammar.