A Hybrid Genetic Algorithm for Cable Forces Optimization of CFST Arch Bridge

In traditional simulation calculation of concrete filled steel tubular (CFST) arch bridge, to find out the initial state of the backward analysis is very difficult due to the force-bearing characteristics of CFST arch bridge. Genetic algorithm (GA), as a general-purposed global optimization algorithm, has the disadvantages of the premature phenomenon and poor performance in local optimization. In the present paper, a hybrid genetic algorithm (HGA), which combines the conjugate gradient method (CGM) with GA, is proposed to improve the performance of GA for cable forces optimization in a CFST arch bridge. The advantages of GA in global optimization and CGM in local searching ability are both included in the HGA method. Numerical example indicates that the results based on the method may be used to the backward analysis of the initial state, and that the proposed HGA has excellent features of quick convergence rate and best global performance.