Design Of Automatic Screw Repair System Based On Machine Vision

Single screw extruder is mainly used for extrusion of PVC,polyethylene and other thermoplastic plastics,it can be processed a variety of plastic products,also can be used for melting granulation.The extruder screw is the key part of the extruder,so the quality and precision of the screw is also improved.But a long time exposed to high temperature and strong chemical environment,and continuous operation in the barrel,resulting in the screw thread surface corrosion,wear,peeling and other irregular surface defects.The screw thread surface wear will not only reduce the service life of extrusion machine,even in the process of production will pr oduce safe hidden trouble.In recent years,our country usually adopt the traditional repair technology to repair the damaged screw thread surface,however,these traditional repair technology on the size of the screw,screw materials and high precision requirements of the screw,will be limited,it is difficult to achieve a certain degree of accuracy requirements.In order to make up for the inadequacy of the traditional restoration techniques,the machine vision technology and the mechanical arm control system are used to detect the screw thread surface of the extruder and determine the size of repair in this paper.Finally realizes the screw thread surface repair.This paper based on the machine vision technology of extruder screw thread surface wear automatic detection accuracy,choose lighting de vice,image acquisition card,linear CCD camera,computer and a series of hardware equipment,screw test of machine vision system has been established.In this process,the camera calibration,and the image preprocessing by MATLAB,and then using the SIFT algorithm to pretreatment of image feature extraction and featu re matching,to reconstruct the 3D model of the screw by point cloud data,comparing the reconstruct 3D model with standard model,it is concluded that the condition of the screw thread surface wear and the size of repair was determined.Then,the worn surface was repaired by laser cladding,and the structure of the repairing device was introduced.The control system of the mechanical arm is controlled by the motion control card,According to the control system of the mechanical arm,the IRB7600 series six axis intelligent robot produced by ABB company is selected.From the analysis of the kinematics of the mechanical arm,draw the 3D model of the mechanical arm and the structure diagram of the connecting rod.Solving the end coordinates relative to the base coordinate system transformation matrix,thus the position coordinates and attitude coordinates of the mechanical arm are obtained.And from the analysis of the inverse kinematics of the mechanical arm,the solution position and attitude of the mechanical arm is solved,selected the closest a set of solution.Using the cubic spline interpolation algorithm and the S curve deceleration con trol algorithm are combined to determine the trajectory of the mechanical arm.