Research On Surface Identification And Weld Torch Position Control Method Based On Rotating Arc

Surfacing is a kind of welding technology which is used to increase or restore the size of parts or make the surface of weldments have special properties.As an ideal processing method,it occupies a large proportion in the development of modern industry and is widely used.Almost all departments including aviation,transportation,energy,metallurgy and electronics are inseparable from surfacing technology.However,the current surface overlay technology is limited to manual operation,and there are many problems in the research of automatic surface overlay,such as the inaccurate extraction of welding signals,the inability to identify the surface of overlay,and the difficulty in controlling the real-time attitude of welding torch.In order to ensure the quality and efficiency of surfacing,it is necessary to realize the surface identification of surfacing layer in the process of surfacing.In this paper,the theory of La position surface recognition is put forward on the basis of magnetic control rotating arc sampling.Based on this theory,the point cloud of rotating arc sampling is obtained by space normal vector and isosurface recognition.After space coordinate inversion,the adjustment of torch space angle and distance is obtained,and the adaptive surfacing repair of irregular surface is realized.The specific contents are as follows:1.The design of rotating arc sensor for surfacingIn order to adapt to the special situation of data sampling for irregular surface in surfacing,an ellipse rotating arc sensing model is proposed.Based on this model,a complex function compensation model is proposed to ensure the real-time supplement of weld deviation discrimination information in the process of surfacing,that is,to ensure the accuracy and effectiveness of the obtained surfacing surface data.2.Extraction of spatial point cloud from welding signals acquired by rotating arcBased on the analysis of the actual welding situation of surfacing and the dynamic model of surfacing,the acquisition model of surfacing layer surface point cloud is established.The electric signal data obtained by rotating arc can be extracted from the surface point cloud coordinates,and the point cloud can be pre interpolated to obtain more comprehensive surfacing surface point cloud information,so that the data range of subsequent surface identification is more complete and Accurate.3.Carry out La position surface recognition on the above verified space point cloudAiming at the way of obtaining the surface information of hardfacing layer by rotating arc scanning,this paper puts forward the LA-possion surface recognition theory to analyze the above spatial point clouds.The input information of LA-possion equation can be obtained by obtaining the normal vector by the least square method of partition,and then the isosurface extraction operation is carried out for the point cloud data containing the normal vector information Adjustment of torch position.4.A full state feedback nonlinear sliding film control model is proposed to guide the position and orientation of welding torch in real timeThe normal vector and isosurface information obtained from La position surface recognition are used as the input of the control system.After the space attitude control of the surfacing controller and the transformation of the space coordinate matrix by the converter,the welding robot performs the adjustment of the space attitude of the welding torch.The results show that the proposed surface recognition theory and torch pose control model have good real-time performance,high recognition accuracy,and can significantly improve the efficiency and quality of surfacing production.