Research On Welding Seam Tracking Method And Obstacle Identification Technology Based On Laser Displacement Sensor

With the continuous development and depth promotion of modern automatic welding technology,the upgrading and transformation of traditional manufacturing industry are facing great opportunities and challenges.As an indispensable material processing method in traditional manufacturing industry,welding automation is an inevitable trend in the development of welding industry.The emphasis of modern welding automation lies in the instant processing of welding seam tracking.Therefore the welding seam tracking is the key prerequisite of modern automatic welding.The key point of welding seam tracking is to identify and correct the curve track of complex welding seam.The obstacles with random distribution,variable size and structure can be classified and identified and intelligently avoided.In this paper,the laser light intensity control system are designed and the laser intensity adaptive driver circuit is adapted to all kinds of plates,and analyses the process of automatic welding seam tracking of six kinds of deviations.In view of the girder artifacts on the random distribution of launder,triangle,circular plate,obstacles such as reinforcing plate study a girder obstacle recognition method based on bayesian contour classification.Design of laser intensity control system and laser intensity adaptive drive circuit are designed.According to the principle and characteristics of laser displacement sensor,seam tracking model is established,six deviations in the process of welding seam are analysed,measured by the laser displacement sensor principle and mathematical methods,six kinds of deviations in welding seam were studied.In the process of solving the seam tracking about left and right deviation and the up and down deviation value,can through the actuator instantly and accurately adjust welding-torch gesture has paved the way for the intelligent seam tracking,and lay the foundation for the intelligent seam tracking accurately and instantly,which provides a new thinking for automatic welding seam tracking.This paper introduces the principle of welding seam obstacle recognition and proposes a method of girder obstacle recognition based on bayesian contour classification.Using the left and right difference method to predict and extract the obstacles contour data,using multistage area of girth than the outline of the describing function combined with bayesian classification to the girder obstacles for effective classification,bayesian classification outline of mathematical model is established by using MATLAB software to obstacles scan data has carried on the simulation analysis,and finally on the basis of scanning laser displacement sensor was tested,the test results show that the four types of obstacles in the probability of test sample required length difference than are large and algorithm classification completely satisfy the real-time demand of welding,Therefore,the proposed bayesian contour classification method has high accuracy and real-time performance in the prediction of girder obstacles.The experimental platform of welding seam tracking based on laser displacement sensor is built to verify the welding seam tracking mentioned above the feasibility of the method,according to the sensor tracking experiment shows that the welding seam tracking effect is very well,stable performance,with high precision,in line with the requirements of automatic welding process tracking.