摘要
蒸汽发生器的工作环境复杂、危险,施工人员在完成堵板螺栓紧固时存在一定的风险。提出一种机器人自主作业平台和基于图像识别的螺栓紧固方法,采用图像处理软件Halcon,实现基于灰度值的图像阈值分割,利用圆形曲线算法得到圆后,最终得到圆的几何中心,验证了采用扭矩和时间控制方法的紧固螺栓方法,最后通过螺栓实验和精度实验验证了图像识别和螺栓紧固的过程。结果表明,针对堵板机器人螺栓定位问题采用的基于阈值分割的定位算法识别率为100%,定位最大误差为1.10 mm,最小误差为0.92 mm,满足堵板机器人螺栓自主定位的精度要求。
The working environment of the steam generator is complex and dangerous,and the construction personnel have certain risks when completing the fastening of the blocking plate bolts.The robot autonomous work platform and a method of bolt fastening based on image recognition were proposed.The image processing software Halcon was used to realize image threshold segmentation based on gray value.After the circle was obtained by the circular curve algorithm,the geometric center of the circle was finally obtained.In addition,the method of tightening bolts using torque and time control methods was also shown.Finally,the process of image recognition and bolt tightening was verified through bolt experiments and accuracy.The results show that the recognition rate of the positioning algorithm based on threshold segmentation used for the bolt positioning of the plate blocking robot is 100%,the maximum positioning error is 1.10 mm,and the minimum error is 0.92 mm,which meets the accuracy requirements of the autonomous positioning of the plate blocking robot bolts.
作者
吴玉
董超群
陈嘉杰
董俊杰
温浩
段星光
WU Yu;DONG Chao-qun;CHEN Jia-jie;DONG Jun-jie;WEN Hao;DUAN Xing-guang(China General Nuclear Power Group,Shenzhen 518000,China;College of Mechanical Engineering,Beijing Institute of Technology,Beijing 100081,China)
出处
《科学技术与工程》
北大核心
2020年第28期11626-11631,共6页
Science Technology and Engineering
基金
国家重点研发计划(2017YFB1302501)
山东省重大科技创新工程(2019JZZY010435)。
