Design of bending-active tied arches by using a multi-objective optimization method

The design of bending-active structures is a challenging problem, due to the high non-linearity of the activation process, the coupling between member sizing, structural shape and the deformability and buckling sensitivity inherent in the resulting lightweight configurations. Due to the large number of form-finding variables, the choice of member sizing is one of the main difficulties at the conceptual phase. In this paper, authors propose a design tool to generate efficient structural configurations for braced bending-active tied arches using multi-objective optimization strategies. Initially, a non-linear FE analysis is performed for each plausible configuration and at each generation of the optimization algorithm. In a second step, a genetic algorithm classifies the solutions and establishes new structural configurations according to best performance. Solutions are evaluated in terms of stresses in the active member and cables, and maximum deflections, as required by design codes for pedestrian bridges. Results are given in terms of non-dimensional parameters, in order to make them applicable to a wide variety of scales.