Bridge Frequency Identification Using Multiple Sensor Responses of an Ordinary Vehicle

This paper presents a novel strategy for indirectly identifying the natural frequency of a bridge via multiple sensors on an ordinary vehicle. It does not require specifically designed cars or multiple cars as the excitation sources, as many earlier studies have used, but only one normal vehicle, providing practicability. A signal processing method incorporating cross-spectrum estimation can extract the bridge frequency among complex and varied frequency components in the vehicle spectrogram. The bridge natural frequency as a common vibration component among different sensor responses is identified as a clear peak that occurs in the cross-spectrum. The methodology is validated by an experimental study using a steel beam and a scale truck model with an engine to construct the vehicle–bridge interaction system. With this approach, the first mode frequency of the bridge is successfully identified under various bridge conditions. In addition, the performance of the proposed strategy in bridges with and without road roughness is discussed, and the influence of vehicle weight on the success rates of frequency identification is also confirmed.