An efficient approach for the optimization of simply supported steel-concrete composite I-girder bridges. (October 2017)

• Record Type: Journal Article
• Authors: Pedro, R.L.; Demarche, J.; Miguel, L.F.F.; Lopez, R.H.
• Is Part Of: Advances in engineering software. Volume 112(2017)
• Page Start: 31
• Page End: 45
• Publication Date: 2017-10
• Subjects: Composite structure -- Bridge -- Optimization -- Finite element method; Computer-aided engineering -- Periodicals; Engineering -- Computer programs -- Periodicals; Engineering -- Software -- Periodicals; Periodicals; 620.0028553
• DOI: 10.1016/j.advengsoft.2017.06.009
• Languages: English
• ISSNs: 0965-9978

Abstract

Highlights: Two stage steel-concrete composite I-girder bridge structural optimization procedure. Simplified and finite element models employed. Structural design constraints based on AASHTO 2002 recommendations and compared with AASHTO 2012. Backtracking Search Algorithm (BSA), Firefly Algorithm (FA), Genetic Algorithm (GA), Imperialist Competitive Algorithm (ICA) and Search Group Algorithm (SGA) employed and compared. Final stage optimization achieved a cost reduction of 9.17%. Abstract: This paper presents an efficient two-stage optimization approach to the design of steel-concrete composite I-girder bridges. In the first step, a simplified structural model, usually adopted by bridge designers, is employed aiming to locate the global optimum region and provide a starting point to the local search. Then, a finite element model (FEM) is used to refine and improve the optimization. Through this procedure, it is possible to combine the low computational cost required on the first stage with the accuracy provided on the second one. For illustration purposes, a numerical example of a composite bridge designed by Pinho and Bellei (2007) and studied by Leitao et al. (2011) is assessed. The objective function is based on the material cost of the structure. Due to the non-convex nature of the problem and to the presence of discrete variables, the first stage optimization is conducted through five well-known meta-heuristic algorithms: Backtracking Search Algorithm (BSA), Firefly Algorithm (FA), Genetic Algorithm (GA), Imperialist Competitive Algorithm (ICA) and Search Group Algorithm (SGA). The SGA is chosen to pursue the second stage because a statistical analysis has shown that it achieved the best performance. It is shown that the proposed scheme is able to reduce the structural cost in up to 7.43% already in the first stage and can reach up to 9.17% of saving costs in the end of the optimization procedure.