基于边缘线追踪的斜拉桥拉索振动频率识别

Identification of cable vibration frequency for cable-stayed bridge based on edge line tracking

针对传统频率法在斜拉桥拉索频率识别中存在传感器布置困难、检测效率低、实现成本高的问题,提出基于拉索边缘线追踪的方法,借助智能手机采集的拉索振动视频,开展了复杂工况下的斜拉桥拉索振动频率识别。首先,引入Otsu算法计算图像感兴趣区域(region of interest,ROI)阈值,利用阈值自适应的Canny-Hough算子进行拉索亚像素边缘检测,定位拉索边缘线的亚像素位置。其次,计算边缘线相对于ROI原点的距离得到拉索振动位移,将位移信号进行逐次变分模态分解(successive variational mode decomposition,SVMD)以减少相机振动和其他噪声的干扰。最后,将降噪后的位移信号进行傅里叶变换识别拉索振动频率。对斜拉桥模型和户外人行桥斜拉索进行试验测试,结果表明即使在复杂工况下,所提方法仍可有效识别拉索的振动频率。

Here,aiming at problems of difficult sensor arrangement,low detection efficiency and high implementation cost of traditional frequency method for cable frequency identification of cable-stayed bridge,a method for cable vibration frequency recognition based on cable edge line tracking was proposed.With the help of cable vibration videos collected with smartphone,cable vibration frequency identification of cable-stayed bridge under complex working conditions was performed.Firstly,Otsu method was introduced to calculate the threshold value of the region of interest(ROI)of images,the threshold adaptive Canny-Hough operator was used to do Lasso sub-pixel edge detection,and locate the sub-pixel position of Lasso edge line.Secondly,the distance between edge line and the ROI origin was calculated to obtain cable vibration displacement,and the successive variational mode decomposition(SVMD)was performed for cable displacement signal to reduce interferences of camera vibration and other noise.Finally,Fourier transform was performed for the denoised displacement signal to identify cable vibration frequency.Tests were conducted for cable-stayed bridge models and outdoor pedestrian bridge cable-stayed cables,and the results showed that the proposed method can effectively identify cable vibration frequency even under complex working conditions.