青马悬索桥的时域模态识别

MODEL IDENTIFICATION OF TSING MA SUSPENSION BRIDGE IN TIME DOMAIN

文章研究了青马悬索桥模态参数时域识别。青马桥脉动测试信号弱、干扰因素多 ,模拟滤波器的滤波效果又不好 ,而Chebyshev数字滤波器却能精确地确定通带和阻带的边缘频率 ,故本文采用它对脉动加速度进行预处理。然后用RD函数将脉动信号转变为自由衰减信号。为了提高时域法模态参数识别精度 ,本文采用了改进的特征系统实现算法。由于青马桥加劲梁长达 2 160m ,加速度计和仪器通道数有限、测试只能分组在不同时间进行 ,而目前的时域模态技术只能处理同时记录的数据。本文先分组时域识别、然后通过参考点拼装全桥的模态 ,共识别出青马桥的 79个模态 ,包括五个复模态。后者与青马桥典型复阻尼特性相一致。本文还把上述结果与频域法及有限元计算结果进行了比较。

An attempt is made to study the model identification of Tsing Ma Bridge at Hongkong in time domain. The response acceleration signals of the bridge are weak. It is disturbed strongly during testing. Neither analogue filters nor digital filters of window function can do well on such signals. The Chebyshev digital filter has more accurate edge frequencies of passband and stopbond. The acceleration signals can be pre?processed by the digit filter. Then, random decrement functions, as the free response acceleration, are obtained from the ambient signals. The testing is carried out using group by group method because of the longer stiffening deck and limited number of equipments. Seventy?nine modes of the bridge are identified including five complex modes. The identified modes in time domain are compared with that of the identified modes in frequency domain and the analytical results.