Research on an Analysis method of Engineering Mechanics in Bridge Field based on Machine learning

Abstract

The identification of bridge cable force has a great influence on the safety and reliability of arch bridge structure, which is a difficult problem in bridge construction. According to a certain tension sequence, the required sample library of suspender cable force is obtained by establishing the finite element model by MIDAS software. The designed cable force of the bridge is taken as the input vector of the radial basis function method (RBF method), and the initial cable force obtained is the output vector. The solution accuracy is controlled by distributed density parameters, the nonlinear mapping relationship between them is approached, and the initial cable force of suspender is calculated directly. The experimental results show that the error between the calculated value and the design value of the cable force can be controlled within 5% by the BRF method, which meets the engineering requirements.