基于桥梁实测应变大数据的超重车信号识别方法

Identification Method of Overload Vehicles Based on Monitored Big Data of Strain in Bridge

以珠江黄埔大桥北汊斜拉桥为研究对象,初步尝试利用实测应变大数据对超重车荷载这类异常信号进行识别。通过在桥梁有限元实体模型上施加临界荷载,得到判别超重车信号的应变阈值;将小波变换得到的实测应变数据的高频分量与应变阈值进行对比,从而提出基于应变阈值的超重车信号识别方法。将该方法与文献中基于小波临界系数的纯信号处理识别方法进行对比,证明了该方法的合理有效性。结合两种方法对多个监测点实测数据进行识别,不仅可以初步确定超重车信号发生的高频时间段;而且可以提高超重车信号识别的概率。该结果可以为桥梁管理部门提供更有价值的指导。

Based on the monitored data from the long-termed health monitoring system on the North Branch of Huangpu Bridge of Pearl River in Guangzhou,the abnormal signals was attempted to identify caused by overload vehicles. A method is proposed for identification of overload vehicles,by comparing the high-frequency component of the measured strain data obtained by wavelet transform with the strain threshold of overweight vehicle signals obtained by applying the critical load on the finite element model of the bridge. The method is compared with the pure signal-processing method based on the wavelet critical coefficient in the literature. Results prove the proposed method to be reasonable and effective. The combination of the two methods can not only determine the high frequency occurring time periods of overweight vehicles,but also improve the acquisition probability of overweight vehicle signal,which provides more valuable guidance for the bridge management.