基于车桥振动的桥梁振型识别及主梁挠度计算方法

根据桥梁间接测量法的基本原理，对一座三跨连续梁桥建立车桥振动有限元模型，提取匀速通过桥梁时车辆竖向加速度时程响应，利用中心差分法计算接触点加速度时程响应，采用快速傅里叶变换得到加速度频谱图，应用峰值拾取法识别出桥梁的前三阶频率；利用带通滤波技术从接触点竖向加速度响应中提取与桥梁频率相关的分量响应，通过希尔伯特变换得到测试桥梁的前三阶振型，并与有限元理论振型进行比较。结果表明：车体质量的改变对于振型识别没有明显影响；车速较低时对于振型的识别不利，但选择合适的驱车速度仍能够保证振型的识别精度。基于桥梁有限元模型，将识别的振型进行质量归一化，计算主梁的测试位移柔度矩阵，设计桥梁标准荷载试验方案，利用柔度矩阵计算主梁在试验荷载下的预测挠度，并与理论挠度进行比较。结果表明，在合适的车速下，预测挠度与理论挠度误差基本满足工程精度要求。

Vibration mode identification of bridge girder and deflection calculation of main girder based on vehicle-bridge vibration

According to the basic principle of bridge indirect measurement, the finite element model of vehicle-bridge coupled vibration is established for a three-span continuous girder bridge. The time-history response of the vertical acceleration is extracted when the vehicle passes the bridge at a constant speed. The acceleration spectrum is obtained by using fast Fourier transform. The acceleration spectrum is obtained by using fast Fourier transform. The central difference method is used to calculate the time-history response of the contact point acceleration. The first three frequencies of the bridge are identified by the peak picking method. The bandpass filtering technology is used to extract the component response related to the bridge frequency from the vertical acceleration response of the contact points, the first three modes of the bridge are obtained by Hilbert transform. The identified mode shapes are compared with the finite element theoretical mode shapes. The results show that the change of vehicle mass has no obvious effect on modal identification. Although the low speed is unfavorable to the mode identification, the accuracy of mode identification can be ensured by selecting the appropriate speed. Based on the finite element model of the bridge, the mass of the identified modes is normalized, the test displacement flexibility matrix of the main girder is calculated, and the standard load test scheme of the bridge is designed. The flexibility matrix was used to predict the deflection of the main girder under the test load, and compared with the theoretical deflection. The results show that the errors of predicted deflection and theoretical deflection meet the requirements of engineering accuracy.