Research On Technology Of GMAW Automatic Welding Of Complex Space Trajectory To Flatrack Container Side Rail

With the implementation of the national "The Belt and Road Initiative" strategy,the multimodal transport industry has achieved rapid development.Flatrack container plays an important role in multimodal transport,and side rail assembly is a key step in the manufacture of flatrack container.The welding structure of side rail is mainly composed of complex welds at multiple locations.There are random obstacles such as U-shape tunnel,process plate,auxiliary plate and hole on the welding trajectory,which make the workpiece have the characteristics of variability of welding trajectory and diversification of structure.Moreover,the side rail itself has a large deflection,so the efficiency of automatic welding of side rail by traditional methods,such as robot guiding is low.The bad on-site working environment,complex space weld trajectory and low precision of machining and assembly make it difficult to control the stability of welding quality and production efficiency.In order to solve these problems,an automatic welding system and a complete set of equipment for flatrack container side rail are necessary to be developed.With this solution as it's mentioned above,it will fulfill the goal of automatic tracking welding of complex trajectory welds,improve welding production efficiency and quality,reduce labor costs,meet the production needs of enterprises,and finally promote the reform and upgrade of welding industry.1.The starting point positioning method is based on the fusion sensing of vision and welding wire touch.It works by establishing the two-dimensional coordinate system of laser scanning displacement sensor and using the laser sensor to obtain the weld position for the first time positioning.After that it corrects the deformation deviation of welding wire with the aid of welding wire tactile sensation for the second time positioning.This solution is a simple and practical way to the welding starting point positioning of welding wire length change and welding wire inclination change in the normal range.A rotating mirror device is adding to the front of the laser displacement sensor so that the laser scanning displacement sensor can get a large range of measurement.With this design,the laser sensing system gets two-dimensional measurement capability.It enlarges the measurable range and it is easy to be controlled.A welding wire tactile system is proposed.The system explores the relative position relationship between the welding wire end and the welding seam by tactile mean,and assists in correcting the deviation caused by welding wire deformation.The experimental results showed that the developed laser scanning displacement sensor combined with welding wire tactile sensing system can accurately locate the welding starting point.2.The welding trajectory recognition method is based on laser scanning displacement sensing,which was used for automatic guidance of welding gun of intermittent welds in space.A self-adaptive laser scanning displacement sensor system for measuring angle is developed to detect the corner features of complex structures.Based on the optical design of point laser optical triangulation and rotating mirror,the sensor can change the measuring angle while maintaining the physical installation position and posture of the sensor,so as to adapt to the corner detection of workpiece welds under different product types and complex working conditions.In this paper,a welding trajectory recognition algorithm based on Euclidean distance discrimination is proposed,which accurately classifies four typical spatial trajectories in complex container rail structures.The algorithm extracts shape features by constructing feature triangles of welding trajectories,then uses discrete Fourier analysis to process shape feature sets,and solves the feature vectors used to describe shapes.Finally,the samples classify to different categories according to Euclidean distance between feature vectors of test samples and category matching library.Based on this method,the welding gun is guided to automatically complete “weld beginning-arc extinguishing-obstacle avoidance-arc igniting-welding”,which greatly improves the manufacturing efficiency and quality stability of complex container rails.3.The fast estimation method of three-dimensional pose for high-speed GMAW welding of large container rail and the algorithm of arc welding seam tracking are studied.The vertical plate plane and flat plate plane of large-scale container rail welding are obtained by fast point cloud segmentation,and then the weld pose information is calculated.The experiment of pose information estimation is carried out.Under high welding speed,the low error of weld pose recognition is realized,and the robustness of automatic welding of large-scale box beam to complex working conditions such as positioning weld and low assembly accuracy is enhanced.The observation equation and state equation of swinging arc are established.The welding electric signal is processed by particle filter algorithm,and the deviation value is obtained by unilateral integral comparison method.The contrast experiment of welding seam tracking is carried out.The results show that the particle filter algorithm based on fuzzy PID control can improve the precision of welding seam tracking.4.The law of droplet transfer in automatic welding of high-speed GMAW fillet weld based on swinging arc was studied.The automatic tracking welding experiment of swing arc fillet weld was carried out,and the effects of different welding processes and swing parameters on droplet transfer were studied.The droplet transfer process of swing arc high-speed automatic fillet welding was successfully observed by using industrial highspeed camera technology.Combined with welding parameter analysis system and swing arc seam tracking system,the droplet transfer characteristics of GMAW automatic fillet welding based on swing arc tracking were revealed,and the relationship between the minimum distance from wire end to workpiece and droplet transfer was analyzed.The relationship equation used to qualitatively describe droplet transfer and the minimum distance from wire end to workpiece was deduced.The results show that the change trend of droplet transfer frequency is opposite to that of the minimum distance from wire end to workpiece.It is helpful to determine the appropriate process parameters and estimate the real-time spatial position information of the welding gun in the automatic welding process.Several swing arc automatic tracking welding machines were assembled in Guangdong Xinhui CIMC Special Transportation Equipment Co.,Ltd.And the automatic welding experiment of side rail was carried out.The test results showed that it can achieve the quality demand of customers.Finally,according to the on-site production condition,two automatic welding production lines for flatrack container are designed and developed.