基于无标记视觉测量算法的核电厂管道振动测量方法研究

Research on Vibration Measurement Method of Nuclear Power Plant Pipeline Based on Unmarked Vision Algorithm

为改善接触式测量手段难以对薄壁管、小支管等结构振动响应进行有效测量的问题,本文提出了基于视觉测量条件下,通过相机标定算法和光流算法计算不同时刻的相邻帧光流,实现对管道二维方向的无标记视觉结构运动测量。以悬臂梁和核电管道两种典型结构为对象进行试验验证,对测量对象随机取点的计算数据与激光位移传感器、加速度传感器等测量结果进行了对比。结果表明,管道振动的视觉测量计算结果与激光位移传感器和加速度传感器测量结果基本一致,相对误差小于4.9%,因此,本文所提无标记视觉测量算法可作为管道振动测量的一种非接触式测量选择方案。

In order to improve the problem that the vibration response of thin-walled pipes and small branch pipes is difficult to be effectively measured by contact measurement method,this paper proposes to calculate the optical flow of adjacent frames at different times based on the camera calibration algorithm and optical flow algorithm under the condition of visual measurement,so as to realize the unmarked visual structure motion measurement in the two-dimensional direction of the pipeline.Experimental verification was conducted on two typical structures,cantilever beam and nuclear power pipeline,and the measurement results of random points were compared with those of laser displacement sensor and acceleration sensor.The results indicate that the results of visual measurement of pipeline vibration are basically consistent with those of laser displacement sensor and acceleration sensor,and the relative error is less than 4.9%.Therefore,the unmarked visual structure motion measurement algorithm proposed in this paper can be used as a non-contact measurement option for pipeline vibration measurement.