I. Welding Robot Welding Process
Welding industrial robots are primarily used in industries requiring high productivity. Typically, spot welding and arc welding can be performed with the assistance of robots. Besides resistance spot welding and arc welding, two common welding processes used for production purposes are metal inert gas (MIG) welding and tungsten inert gas (TIG) welding.
Welding robots automate processes, ensuring higher precision, less waste, and faster operation. With the widespread application of machinery, welding robots can adapt to various welding processes, including arc welding, resistance welding, and spot welding. One of the more common welding processes used by welding robots is arc welding. Arc welding is a welding process in which metals are fused together electrically. A direct current (DC) or alternating current (AC) is used to create an arc between electrodes (consumable or non-consumable) and the metal, causing them to melt and bond together.
Resistance spot welding is a welding process in which two shaped copper alloy electrodes are used to concentrate welding current into a single point, simultaneously joining sheet metal together. The high current forced into a single point causes the metal to melt and form a weld. By using high current at a specific location, the rest of the sheet metal is not heated during the welding process.
Spot welding: Some materials resist electric current, making other forms of welding impossible. This often occurs in the automotive industry, for parts used to assemble car bodies. To overcome this problem, welding robots use a variant of resistance welding to join a pair of thin metal plates at a single point.
II. Visual Weld Tracking System for Pipeline Welding Robots
A weld seam tracking system based on the generation of visible light has been proposed for application in pipe welding robots. First, a vision sensor was designed based on analysis of laser reflection from the welding surface, the camera's position, the plane of the laser beam, and the effect of the laser stripe image after welding. To prevent severe reflection interference in the weld seam image, image processing and feature extraction algorithms were developed. An image vision control system was then employed for weld seam tracking in pipe welding. The system's performance was verified through experiments controlling a pipe welding robot for weld seam tracking.
In robotic welding, weld seam tracking is a crucial issue and the foundation for high-quality automated welding. Most industrial welding robots are used for training, repeating a path to meet the beam positioning requirements during welding. This approach presents several problems, such as inaccurate welding position and deformation/torsion at the weld joint due to heat diffusion. These issues cause the beam to deviate from its theoretical welding path, making beam control of the weld seam trajectory essential during welding. Furthermore, pipe welding robots cannot pre-define the weld seam because it may shift from its internal position when the pipe changes direction. The weld seam trajectory can change with pipe movement along its axis. In such cases, this approach is unsuitable for pipe welding, and the welding robot needs to correct for beam misalignment and timely weld seam adjustments during welding.
To prevent weld misalignment during pipe movement, a solution is to use a three-degree-of-freedom multi-robot to raise and lower the pipe, adjust its position, and correct its orientation. When the pipe changes direction, the weld will deviate from its original position, necessitating a weld tracking system for high-quality welding.
