直接应用基于自噪声激励的模态参数识别方法识别桥梁模态参数精度较低。原因包括:运营桥梁在自由车辆激励和环境激励共同作用下的振动响应不能被近似为白噪声激励下的响应;车辆过桥过程中相当于对桥梁附加移动质量,车桥耦合体系为时变体系;大型桥梁结构响应具有不稳定成分;振动测试及分析系统精度有限。针对以上问题提出:利用短时的无车上桥时段的加速度响应数据段进行识别;利用时频分析中的小波分析技术提取响应中各阶模态对应的频带并去除不稳定数据,进而对稳定数据段直接在时域中利用波形分析法识别频率和振型;对长期连续的识别结果利用小渡分析和统计分析等技术进行综合后处理。利用上述方法对滨州黄河公路大桥连续近4天的振动监测数据进行识别,识别结果可信、稳定,可应用于实际工程。
Directly applying the modal parameters identification method based on white noise excitation to identify bridge modal parameters will lead to low precision. The reasons are as follows. The vibration responses of bridge in serve under free traffic and ambient excitation cannot be assumed as the responses under white noise excitations approximately. The vehicles on bridge will attach additional moving mass to bridge, as a result the coupling system of vehicle and bridge will become a time-dependent system. The vibration responses of large bridge structure include unsteady-going part ; the precision of vibration test and analysis system is limited. Aiming at above problems, the presented method are provided as follows: using the short time acceleration data under no traffic condition to identify modal parameters; applying wavelet analysis technique subordinated to time-frequency analysis to extract frequency band of each mode and to eliminate unsteady-going data, and applying waveform analysis method to identify the frequencies and mode shapes in time domain directly from steady-going data; applying wavelet analysis and statistics analysis to postprocess the long term identification results. The above method is applied to identify modal parameters from 4 days vibration data of BinZhou Yellow River Highway Bridge. The result is reliable and stable and this method can be adopted in real project.