Fully automated and non-contact force identification of bridge cables using microwave remote sensing

Abstract

Efficient force identification of bridge cables is important to the performance evaluation of cable-supported bridges. For this purpose, sensors are conventionally installed on the cables to record vibration responses, which is nevertheless time-consuming and labor-intensive. This study develops a fully automated and noncontact cable force identification method using microwave radar sensing, enabling forces of a group of cables to be identified simultaneously and thus largely improving the efficiency. The fast sieve method with a linear time complexity is proposed to find the local maxima of the Fourier spectrum of the displacement measurements. Subsequently, the weighted hash voting is presented to identify the mode orders of the picked frequencies. An appropriate cable vibration model can then be used to compute the tension force. The proposed method has been validated in cable force identification of a newly constructed railway cable-stayed bridge during loading tests. Cable force variations in the loading tests are discussed.