Global optimization of design parameters of network arch bridges

Abstract

Structural optimization of network arch bridges is performed in this study. Optimization has been achieved through execution of a simulator, evaluation of a constrained objective function and adjustment of system parameters in an iterative and directed way. Objective is to minimize the total material cost required for hanger, arch concrete and arch reinforcement of network arch bridges. The optimization problem is characterized by having mixed design variables. Structural analysis of network arch bridges is performed by a finite element simulator, ANSYS. Optimization of structural design of the bridge is performed through a global optimization algorithm EVOP. An interfacing environment has been developed by integrating ANSYS with EVOP. The interfacing has been verified through some benchmark problems of optimization. Finally parameters of a tied arch bridge are invoked in the optimization process as the initial design and after completion of optimization process, optimal design variables i.e. hanger arrangement, no of hangers, cross section required for hanger, arch section and rise to span ratio of arch, are obtained within a range of design constant parameters. Response parameters of the arch bridge with optimum design variables are analyzed regarding hanger stress, bending moment, axial force and influence line for bending moment in arch and amount of reinforcement required in the arch. Results are compared with the initially designed arch. Results show that arch bridges with optimal design variables using global optimization technique shows significant improvement over the arch bridges designed initially. Optimal design variables confirm significant reduction of bending moment in arch than the arch bridges with vertical hangers. Based on optimum design criteria it is found that circular arch geometry requires shallower arch than that required for parabolic arch. In addition, within the range of design constant parameters it is observed that 36% to 40% of total cost can be saved for circular and parabolic arches if design is optimized. Results also show that parabolic arch with optimum design variables is more economic than the optimal arch bridges with circular arch geometry. The interfacing environment developed in the study opens the door for simulation driven most economical design of any type of structure.