Research On On-machine Automatic Detection System Of Tool Wear Based On Machine Vision

With the development of the manufacturing industry,the utilization rate of machine tools is increasing.As a machine tool,tool wear condition will directly affect the machining accuracy and efficiency of the machine tool.Therefore,it is of important practical significance to detect the wear state of the cutting tools during the machining process.At present,the tool wear detection methods are mainly based on traditional detection methods,but traditional detection methods have problems such as unstable accuracy and complex sensor installation.Based on these problems,this paper studies a tool wear detection method based on machine vision.This method can not only obtain accurate wear,but also is not susceptible to environmental noise.The purpose of this paper is to use machine vision technology to automatically detect tool wear on the machine,and carry out the software and hardware research of the machine vision-based tool wear automatic detection system.The research content is as follows:(1)The tool wear mechanism and visual inspection method are introduced,the maximum wear width of the tool flank is determined as the tool wear evaluation index,and the tool wear visual inspection principle and system composition are analyzed.In order to realize the automatic acquisition of tool images on the machine,an image acquisition scheme that acquires a continuous image sequence under the rotation of the spindle is proposed,which promotes the application of tool wear detection system based on machine vision in industry.(2)A visual experiment for tool wear detection is designed.Taking the face milling cutter as the experimental object,the cutting processing equipment and hardware selection are introduced,and the image acquisition system is built.The experimental parameters are determined according to the actual application scenarios,and specific experimental schemes are designed.The data collected in the experiment is analyzed in detail,and the characteristics of continuous image sequence data are summarized.This experiment proves the feasibility of the image acquisition scheme,and also provides data support for subsequent image processing algorithm testing.(3)An image processing algorithm based on continuous image sequence is designed.First,through image cropping and enhancement,the efficiency and accuracy of the algorithm are improved.Secondly,a particle swarm optimization algorithm(PSO)optimizated maximum between-class vriance(OTSU)segmentation method based on edge information is proposed to extract the image target area,and a binary image optimization algorithm based on the region area is designed to fill in holes and filter out isolated noise points.Next,an image selection algorithm based on regional shape features is proposed to select the best tool wear image.Finally,an algorithm for measuring the flank wear of the face milling cutter is designed.And the effectiveness of the proposed image processing algorithm is verified through experiments.(4)A set of tool wear detection software system has been developed.Use the Py Qt interface development framework to design the software system interface,and use the lightweight database SQLite for data management.The software system can realize real-time detection,offline detection and detection data management of tool wear,and has broad engineering application value.