| Summary: | 博士 === 國立臺灣大學 === 土木工程學研究所 === 103 === Using the vibration data recorded by a moving vehicle during its passage over a bridge to detect the frequencies of the bridge is a technique proposed by Yang and co-workers in 2004. This technique has been referred to as the “indirect approach” for measuring the bridge frequencies, which this requires the vibration sensors to be mounted only on the test vehicle. In contrast, the conventional approach relies on vibration sensors directly mounted on the bridge for measuring the bridge respone, for which the name “direct approach” is given. The objective of this study is focused on the dynamic properties of the test vehicle for use in the indirect bridge measurement. Also investigated are the effects of pavement roughness and existing random traffic flows on the identification of bridge frequencies.
First, a brief review will be given of the researches on measurement of bridge dynamic properties using the direct approach. From this review, techniques that have been used in the direct approach, especially those for signal processing, will be adopted in developing the indirect approach. Next, a comprehensive review will be given of the previous studies on the indirect approach. After this, for a test vehicle equipped with internal damping, a parametric study based on a two-stage analysis technique will be carried out to process the vehicle response collected, aimed at evaluating the effect of damping and frequencies of the test vehicle on bridge frequency identification. Furthermore, the stochastic subspace identification (SSI) algorithm is modified to include the vehicle-bridge interaction (VBI) effect, resulting in the proposed so-called VBI-SSI approach, by which the noises associated with the vehicle and roughness are filtered out and the identifiability of the bridge frequencies is enhanced.
Then, a detailed review will be given of the progress by Yang’s research group in designing the test vehicle and in conducting the field test, with attention paid to the mechanics properties of the tires of the test vehicle. Two types of tires will be adopted in the field measurement, i.e., the inflatable wheel and the rigid wheel with elastic surface. Meanwhile, a modified sprung mass model is proposed for the simulation of elastic-surface rigid wheels running over rough pavements. Finally, the test cart equipped with elastic-surface wheels will be used in the field test considering various control factors to analyze its applicability. From the numerical studies and field tests, it is understood that pavement roughenss is the key factor determining the efficacy of the indirect approach for identifying the bridge frequencies. The existence of pavement roughenss may amplify the vehicle-frequency peaks in the acceleration spectrum of the vehicle response, making it difficult to identify the bridge frequency peaks. Thus, in order to enhance the workability of the indirect approach under higher moving speeds, it is necessary to increase the stability of the test vehicle, say, by adding the vehicle weight, while enlarging the vibration amplitude of the bridge (rather than the vehicle), say, by allowing lagre traffic flows to exist on the bridge at the moment when the measurement is taken.
|
|---|
