基于计算机视觉的焊接固有应变实测方法研究
Research on Measurement Method of Welding Inherent Strain Based on Computer Vision
苟中林 1罗宇 1袁昱超1
作者信息
- 1. 上海交通大学 船舶海洋与建筑工程学院, 上海 200240
- 折叠
摘要
固有应变被用于大型焊接结构应力和变形预测中.本文提出一种基于计算机视觉的固有应变工程应用的简便实测方法.该方法设计了一个特殊的带小孔的定位尺,将定位尺放置在焊缝附近,通过小孔在焊缝附近打上测量标记,在焊接前后用手持拍照设备对测量点进行拍照.采用一系列图像处理算法求解标记点在定位尺坐标系中的坐标,进而计算获取焊缝的固有应变.研究结果表明,该方法有较高的的重复精度,能够满足固有应变高效准确测量的工程要求.
Abstract
Inherent strain is used to predict the stress and deformation of large-scale welded structures.A simple and convenient measurement method for the engineering application of inherent strain was proposed based on computer vision.In this method,a special positioning ruler with small holes was designed.The positioning ruler was placed near the weld,and measured marks were marked near the weld through the small holes.The handheld camera device was used to take pictures of the measured points before and after welding.A series of image processing algorithm was used to solve the coordinates of the marked points in the positioning ruler coordinate system,and then the inherent strain of the weld was calculated and obtained.The research results show that this method has high repeatability and can meet the engineering requirements of high-efficiency and accurate measurement of inherent strain.
关键词
固有应变测量/图像处理/焊接变形预测Key words
inherent strain measurement/image processing/welding deformation prediction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据