Deformation Monitoring of Xijiang Railway Bridge with Ground-based Radar Interferometry

Ground-based interferometric radar (GBIR) is an innovative tool for monitoring bridges deformation. Bridges are usually located over rivers and valleys, and the severe water vapor make it challengeable to extract accurate deformation from atmospheric delay error. Moreover, GBIR can only capture the one-dimensional linear deformation, but hard to retrieve the multi-dimensional deformation. In this study, the Xijiang Railway Bridge is taken as an example to conduct deformation monitoring experiments. The independent component analysis (ICA) is used to remove atmospheric effects errors. The error average optimization rate of ICA is 83.96%, shows excellent performance in GBIR bridge error removal. We also use a data fusion framework based on GBIR and inertial vision-based measurement (IVM) to obtain multi-dimensional dynamic deformation of the bridge with crossing train. The maximum of deformations along the vertical and lateral directions are 211.3 mm and 123.4 mm.