Study on integrated computer aided design method for series of steel roof truss based on Genetic Algorithm

Abstract

Copyright © 2013 by Research Publishing Services. Steel roof trusses of industrial buildings usually have repeating patterns. The automatic selection of size, shape and topology variables will reduce environmental impacts and provide additional cost savings. In this study, an integrated design method incorporating a simple Genetic Algorithm (GA) is developed for discrete optimization of two dimensional repeating pattern steel trusses. The objective function, considered solely the total weight, is minimized subjected to serviceability and strength requirements. The GA-FEM based design approach is developed as a module in MATLAB. Special features include seven design configuration variables, an open format for prescribing loading conditions, design checking criterion using AISC specification, and a comprehensive database of available commercial steel member sizes. The results found in the optimization process help the designer providing valuable preliminary conclusions guiding to good final solutions while avoiding the repetitive evaluation of trialand-error designs in a time consuming process.