Research On Intelligent Manufacturing Technology Of Robot Welding Based On Image Processing

Welding is an important process technology in manufacturing,and is widely used in industrial manufacturing,military aerospace,household appliances and other fields.During manual welding,harmful gases,arc flash,smoke and so on will cause some damage to the body.In addition,workers need to maintain a fixed welding posture for a long time,and their work intensity is large,which makes it difficult to maintain the stability and consistency of welding quality.Therefore,in order to realize welding automation,improve the production efficiency and quality of welding products,and improve the harsh working environment of workers,a robot welding intelligent manufacturing technology based on image processing was studied.In this paper,ABB IRB2400 welding robot is selected,using offline programming software Robot Studio and three-dimensional design software Solidworks to build a welding robot workstation.By querying the welding manual and conducting welding tests,the welding process parameters were determined.Commonly used welding I/O signals are created,signal correlation is performed,and welding procedures are completed.When performing welding simulation,using the teaching pendant to produce the screen,continuously optimizing the welding parameters,the ideal welding seam was obtained.Welding robot kinematics analysis is a prerequisite for studying robot technology.It provides a theoretical basis for welding robots to achieve trajectory planning and motion control.This article uses the D-H parameter method to determine the parameters of each link of the robot,and uses the Matlab Robotics Toolbox to complete the modeling of the ABB IRB2400 robot.The forward and inverse kinematics of the robot are solved to verify the correctness of the robot modeling.Using polynomial interpolation algorithm for joint space trajectory planning,the fifth degree polynomial interpolation algorithm is determined to be an ideal algorithm.Given the initial and end poses of the robot,the displacement curve,motion curve,angular velocity curve and angular acceleration curve of the end trajectory are generated,which verifies the rationality of the robot design.In order to improve the welding accuracy and realize the real-time tracking of the weld seam,the visual sensor is used to collect the weld seam image and transmit it to the computer for image processing in real time.In image processing,a variety of algorithms such as filter processing,gray-scale processing,and binarization processing are used to process the weld image,and a black and white image with clear edge contour is obtained.At the same time,the center line of the weld is generated by the image skeleton method.During welding,the position deviation between the welding torch and the center line of the welding seam is adjusted to ensure that the welding gun is always aligned with the center line of the welding seam,so that the welding has real-time and high accuracy.The change law of welding temperature field is an important basis for controlling welding residual stress and improving welding quality.Using ANSYS finite element analysis software,a reasonable weldment model was established,and the grid division method was used to divide the grid.Using the life and death unit technology,the filling process of the weld is simulated.By writing parameterized programs in APDL language,the welding temperature field clouds at different moments are obtained.By analyzing the change law of the temperature at different nodes of the weld with time,the welding process parameters are optimized to provide a reference for improving the welding quality.The intelligent manufacturing technology studied in this paper mainly includes offline programming technology,industrial robot technology,weld sensor technology and finite element analysis technology.A set of reasonable welding seam tracking system is designed to solve the problem of the interference of arc light and spatter.It can realize the extraction of the welding seam centerline under various working conditions such as straight line,curve and broken line.The welding seam tracking system cooperates with the collaborative work of welding robot workstations,which is of great significance to the development of intelligent welding technology.