Structural optimization of two-girder composite cable-stayed bridges under dead and live loads

A large number of variables are involved in the optimization of cable-stayed bridges, which makes the optimization impractical when many load cases are considered. To reduce the number of variables to be optimized, a discrete phases approach for structural optimization is developed in this study. The approach couples the finite element method with the genetic algorithm optimization approach. The design variables are divided into two categories: (i) main variables: number of stay cables, I-girder inertia, concrete slab thickness, and tower dimensions; and (ii) secondary variables: I-girder dimensions, stay-cable areas, and pre-tensioning forces. Two design objectives are tested: (i) lightest deck mass; and (ii) lowest material cost. Three load cases are considered: (i) dead and truck plus lane live loads; (ii) dead and lane live loads; and (iii) dead load. The results show the importance of considering the truck loads in structural optimization and the efficacy of the phases approach for different objectives.