Research On All-position Welding Control Technology Of Pipeline

With the continuous advancement of the national "2025" strategy,various industries have achieved good development.As an industrial connection method,welding plays an important role in large-scale metal equipment and pipeline welding infrastructure projects.In the field of pipeline welding,both traditional manual welding and semi-automatic welding are faced with the problems of high labor intensity and low welding efficiency.In response to the above problems,an all-position welding control system based on machine vision and nonlinear regression was developed,aiming to improve welding quality,improve welding efficiency and reduce manual workload.First,the overall scheme of the system including vision module,wireless module,control module and execution module is designed,the hardware model is determined and the design of the software flow chart is completed.Comprehensive consideration,the system framework selected for the project is "industrial computer" + "PLC" + "touch screen",using industrial computer to process image information,to realize the calculation of welding seam tracking and rectification,and control execution modules such as welding robots and digital welding machines through PLC Action,using the touch screen as the carrier of data display and command transmission,to achieve human-computer interaction function.Secondly,a mathematical model is designed to analyze the relationship between welding influencing factors and welding process parameters.Through the partition test,it is verified that when the welding parameters are fixed,the molten pool flow presents different shapes at different positions of the same pipe under the action of gravity,and shows similar shapes at the same position of different pipes.In this paper,the whole pipeline is subdivided into 16 regions according to the characteristics of different welding positions.It is proved by experiments that the welding current I,welding voltage U and the crawling speed V of the welding machine robot have a great influence on the shape of the molten pool and the welding seam formation.The function equations between the above three parameters and the width d of the welding layer to be welded,the height of the welding layer and the crawling angle of the welding robot are obtained.Thirdly,an image processing algorithm to obtain the coordinates of the inflection point of the weld groove is designed.The processing flow is as follows: the first step,according to the characteristics of the interference noise of the weld image,the Wiener filtering method is used to remove the high-frequency noise signal without destroying the edge information of the image,and then the average filtering method is used to smooth the image to process the Gaussian noise.And use the shock filtering method to obtain clear image edges;the second step,design a clustering algorithm to further obtain the effective feature information in the image and improve the accuracy;the third step,use the fitting method to fit the image feature points extracted in the previous step.The straight line segment is synthesized,and the height and width of the weld in the image are calculated according to the calibration equation.Finally,the prototype of the pipeline all-position welding robot is designed,and the welding test platform is built.The vision module,wireless module,control module and execution module coordinate and cooperate to realize all-position autonomous welding.The results of the welding experiment show that the deviation of the height and width of the weld is less than 0.5mm and 0.2mm,respectively,which meets the welding accuracy requirements.This subject studies the control technology of all-position welding of pipelines,and develops an all-position welding control system.The system realizes the flexible control of the process parameters in the whole welding process.The experimental results show that it performs well in optimizing the welding quality of the pipeline and improving the welding efficiency.