Weld identification technology of pressure steel pipe for wall-climbing robot operation

Junjie Zhu, Jun Hu, Lianwei Liu, Yanzheng Zhao

Abstract


With the continuous development of hydropower industry in China, the test and maintenance task of hydropower generator
set is increasingly heavy. As a key component of the generator set, the water inlet pressure steel pipe, because of its large diameter and long
distance characteristics, makes the past by manual detection and maintenance of the operation mode has a long period, high labor intensity
and other problems. With the progress of robot technology, mobile robots are considered to solve the above problems instead of manual
work. In the inspection and maintenance of pressure steel pipe, there are many tasks around welding seams. In order to improve the quality
and effi ciency of robot operation, it is necessary to realize high-precision welding seam identifi cation. This paper takes the welding seam
identifi cation of pressure steel pipe as the research content, adopts the semantic segmentation method based on improved DeepLabv3+,
takes the two-track wall-climbing mobile robot as the test platform, and takes the actual pressure steel pipe as the test object to carry out the
algorithm instantiation test and analysis. Field tests show that the pressure steel pipe weld identifi cation technology can identify the weld
effi ciently and accurately, which is helpful to improve the working effi ciency and accuracy of the wall-climbing robot.

Keywords


hydropower pressure steel pipe, wall-climbing robot, weld identifi cation

Full Text:

PDF

Included Database


References


[1] Lichao Cao,Xiaoguang Liu,Xiaoming Jiang, etal. Optimization Design of Crawler Wall-climbing Robot [J]. Automation and Information Engineering,

2019, 40(2):9-13.

[2] Simin Cheng,Weiyu Chen,Peijie Cong. Research Status of Wall-Climbing Robot [J]. Mechanical and Electrical Engineering Technology.2019,48(09):6-10.

[3] Jie Wang,Zhiheng Mai,Yuenong Fei. Research status and development trend of Welding Seam Inspection Robot [J]. Transducer & Microsystems, 2020,

39(02): 1-3+10.

[4] Huanwu Cong,Fujuan Guo,Fei Lv, etal. Research on weld identification technology based on CCD image processing [J]. Electronic Measurement

Technology, 2012, 35(03): 73-7.

[5] Yufei Zhang,Yishun Zhang. Research on weld image recognition Algorithm based on OpenCV [J]. Welding Technology, 2020, 49(05): 12-4.

[6] Jiajie Yu,Jianping Zhou,Ruilei Xue, etal. Weld Surface quality detection based on structured light vision and illumination model [J]. Chinese Journal of

Lasers: 1-15.

[7]LI X, LI X, KHYAM M O, et al. Robust Welding Seam Tracking and Recognition. IEEE Sensors Journal [J], 2017: 5609-5617.

[8]DONG S, SUN X, XIE S, et al. Automatic defect identifi cation technology of digital image of pipeline weld [J]. Natural Gas Industry B, 2019, 6(4): 399-

403.

[9] Shikuan Zhang,Qingxiao Wu,Zhiyuan Lin. Detection and segmentation of structured light fringe in weld image [J]. Chinese Journal of

Optics,2021,41(05):88-96.

[10]Chen L C, Zhu Y, Papandreou G, et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]//Proceedings of the

European conference on computer vision (ECCV). 2018: 801-818.




DOI: https://doi.org/10.18686/esta.v10i4.608

Refbacks

  • There are currently no refbacks.


Copyright (c) 2023 Junjie Zhu,Jun Hu,Lianwei Liu,Yanzheng Zhao