抄録
This paper proposes a method to identify the damage of bridges based on coupling vibration with moving vehicles. The identification problem is formulated as a probabilistic inverse problem where an objective function and its minimization are incorporated. The objective function for the identification consists of observation and prior information terms. DFP method is used in order to minimize the objective function. To overcome the ill-posedness when the observation data is insufficient, prior information term is useful. Gauss and Laplace distributions are compared as prior information for the damage identification. Numerical simulation of the identification is demonstrated with a three-dimensional bridge model. The simulation result shows the possibility of damage identification from limited number of observation points.
