Key Technology Of Machine Vision Equipment Welding Spot Inspection

Resistance spot welding is the most important connection between the metal parts of the car body.The welding quality directly affects the mechanical properties,safety characteristics and service life of the automobile.Ultrasonic detection method has the advantages of cost saving,safety and reliability,and it is the most widely used to detect the quality of solder joints for the domestic and foreign manufacturers.The welding spot ultrasonic testing still rely on manual sampling approach,has been unable to meet the demand of mass production car automation.So it is necessary for the ultrasonic propagation attenuation characteristics and welding spot automatic detection orientation for further research,developed to ultrasonic testing equipment which can real-time monitoring the quality of welding spot,to realize the automatic welding spot quality inspection,and improve the quality level and production efficiency of automobile as a whole.Firstly,this paper studies the basic theory and the general technical route of the machine vision welding spot inspection equipment,the basic theory mainly includes ultrasonic propagation mathematical model and camera calibration technology.Provides theoretical and technical support for the follow-up researchment.Secondly,established the finite element simulation model of welding spot,analyze the corresponding relation of the various types of quality welding spot with the ultrasonic echo signal characteristics,and using signal processing method for effective separation of welding spot signals.Firstly described the finite element modeling method,using the simulation model got the signal transmission attenuation curve of each type of welds;then analyzed the ultrasonic probe's detection stability when angle positioning in the body surface detection conditions;Finally,this paper presents an ultrasonic signal processing method based on the minimum blind entropy deconvolution,which can suppress the weak noise effectively and realize the extraction of effective characteristic information of the ultrasonic signal and quality evaluation of welding spot.Then obtained the center position coordinates of the welding spot.First of all,based on the research of the image processing method,got the appropriate spot image processing methods,then after image preprocessing,edge contour extraction,Huff circle transformation and finally the center position coordinates of the welding spot are obtained.Finally,the design and development of hardware and software for visual inspection equipment of white body machine are introduced.First of all,logic frame construction and related software development work are carried out for the ultrasonic signal acquisition and processing and robot motion control and other functions.At the same time,the hardware used in the detection system is designed and selected,and the hand-eye calibration method suitable for this subject is studied to realize the machine vision guidance and positioning.At the last,through the visual equipment and foreign hand-held testing equipment on the quality of welding spot evaluation results,verify the effectiveness and feasibility of the development of testing equipment.The results show that the machine vision ultrasonic welding spot quality testing equipment can accurately identify the location of the welding spot,and can effectively evaluate the quality of welding spot.On the basis of this study,it is expected to realize the on-line inspection automation of welding spot quality.