Given the weak noise resistance and low identification efficiency of traditional bridge modal parameter identification methods, this study proposes a data-driven bridge modal parameter identification method based on synchronous clustering theory using the dynamic response of a moving, passing dual-axle vehicle. Firstly, the dual axle vehicle equipped with sensors on both axles traverses the bridge, and the dynamic responses of the vehicle are collected to calculate the dynamic responses of contact points and further obtain the residual response of contact points. Secondly, the residual response of contact points is clustered according to the corresponding statistical moment. Finally, the clustered data of residual response are used to extract the modal parameters of the bridge using the synchronous theory. Thus, the bridge unit stiffness can also be obtained accordingly. The novelty includes the development of an algorithm to eliminate road roughness by changing the axle weight while maintaining the stiffness and the application of synchronous clustering theory to identify bridge modal parameters. Both numerical simulations and field measurements have been conducted to validate the proposed method. The effectiveness of the method has been validated in practical bridge inspections such as simply supported beams and equal span continuous beams. The method mainly focuses on identifying low order modal vibration modes and deflection of simple supported beams and continuous beam bridges with small and medium-sized spans. The results show that compared to the frequently used transfer rate and stochastic subspace identification methods, the proposed method has greater noise resistance and performs better on identifying modal parameters with external excitation and variation of bridge damping ratio.

中文翻译：

---

基于同步聚类理论的超车双轴车辆动力响应桥梁刚度辨识


针对传统桥梁模态参数识别方法抗噪能力弱、识别效率低等问题，该文利用移动、超车双轴车辆的动力响应，提出了一种基于同步聚类理论的数据驱动桥梁模态参数识别方法。首先，双轴均配备传感器的双轴车辆穿越桥梁，收集车辆的动态响应，计算接触点的动态响应，进一步得到接触点的残差响应;其次，根据相应的统计时刻对接触点的残差响应进行聚类。最后，利用残差响应的聚类数据，利用同步理论提取电桥的模态参数。因此，也可以相应地获得桥单元刚度。新颖性包括开发一种算法，通过改变轴重来消除道路粗糙，同时保持刚度，以及应用同步聚类理论来识别桥梁模态参数。已经进行了数值模拟和现场测量以验证所提出的方法。该方法的有效性已在实际桥梁检查中得到验证，例如简支梁和等跨度连续梁。该方法主要侧重于识别简单支承梁和中小跨度连续梁桥的低阶模态振动模式和挠度。 结果表明，与常用的传输速率和随机子空间识别方法相比，所提方法具有更大的抗噪能力，并且在外激励和电桥阻尼比变化的模态参数识别方面表现更好。