Bayesian prediction of deflection based on measurement data for cable-stayed bridges

Abstract

For structural management purposes, various sensing techniques have been applied to monitor the structural deflection of cable-stayed bridges, such as girder deflection and pylon incline. However, it is not an easy task to make effective decisions on bridge management based on measurement data. This study proposes a new Bayesian method for the probabilistic prediction of the deflection of cable-stayed bridges. To build a probabilistic prediction model based on monitoring data, the proposed method introduces the Gaussian process regression with a new combination of kernel functions. The proposed method is applied to an actual cable-stayed bridge in the Republic of Korea, and a probabilistic prediction of the pylon incline is made successfully.