Moving forces identification based on structure health monitoring for cable-stayed bridge with regularizations

A new approach suitable for cable-stayed bridge is presented in this paper. Moving force identification method for cablestayed bridge was seldom reported. In previous research bridges were modeled as Timoshenko beams or orthotropic plates, while cable-stayed bridge can't be modeled as beams or plates. In this paper, the deck of bridge was separated from other parts, by replacing the cables with equivalent forces based on measurements. And according to D'Alembert principle the equation of motion of deck can be established, from which the moving external forces can be obtained, given measured time-varying cable forces and accelerations. However this is typically an inverse and ill-posed problem, because the measured global variables is not so sensitive to local loads such as vehicle loads for a large-span bridge meanwhile the noise plays an essential role in measured signals. The Tikhonov regularization methods are adopted for solving the ill-posed problem. Simulation studies are used to verify the method. The main contributions of this paper are: a. to propose a method for identifying moving forces on cable-stayed bridge; b. both the location and the magnitude of the force can be identified; c. the vehicles in different lanes can be identified.