| Tube-sheet exchanger has important application in boiler, petroleum, chemical and other industries. Welding of tube and sheet according to the design requirements has vital significance in improving labor conditions and increasing the welding quality and productivity. The tube plate automatic welding equipment can satisfy the requirement of welding for tube plate heat exchanger when the positional deviation of the seam was small.But it has great difficulties in welding uneven tube plate workpiece or welding workpiece with large deviation from the welding position.In this paper, COMS industry camera and laser generator were used to establish vision-based tube and sheet robot welding system. The COMS industry camera was calibrated by pinhole camera model, and the geometrical optical model of laser was constructed using laser triangulation principle. The eye-hand matrix was calculated out. According to the experiment results, it could be found that the accurate of calibration based on BP neural network was higher.Industry camera was used to capture laser stripe, of which the characteristic was extracted by image processing algorithm. The center of the circular weld and the plane of circle could be obtained. The information about circular weld was transformed into a coordinate that could be read by robot for directing the spin welding device, which was installed in the end of sixth axis, to complete the weld task. The experimental results show that the trajectory deviation is 0.49 mm, which can satisfy the production requirements.The D-H parameter of SR10C robot was figured out and the robot mode was established by solving its forward kinematics and inverse kinematics equations. The circular track of welding torch was planned from the practical situation of tube sheet welding. In this paper, the 3-dimensional model of robot and its operation environment was established by VC6.0 and OpenGL software and the moving track of tool was simulated. |
PDF Full Text Request