Chatter Monitoring On-line Of Milling Thin-walled Parts Based On Computer Vision

Thin-walled parts are widely used in aerospace,shipbuilding,and automotive fields.And chatter is widespread and affected adversely during thin-walled parts processing.Chatter on the one hand makes the machining process unstable,affects the workpiece processing accuracy,and reduces the machining efficiency.On the other hand,it aggravates the wear of machine tool components,reduces the life of the tools and machine tools,and causes noise pollution.In order to improve the surface quality and production efficiency of thin-walled workpieces,accurate identification and effective monitoring of chatter in the machining process has become an important research in the field of intelligent manufacturing.Due to the noise and vibration interference in the cutting environment,the accuracy of chatter monitoring is generally low and the cost is high,making it difficult to guide actual industrial production.This paper presents an online monitoring method of milling chatter on thin-walled workpieces based on computer vision,and realizes on-line monitoring of chatter based on image texture information.The specific research content is as follows:Firstly,the composition and working principle of the computer vision system are analyzed.The hardware composition of the thin-walled flutter image acquisition system based on computer vision is determined.The chatter image acquisition scheme for milling thin-walled parts is designed to obtain the chatter image of the milling of thin-walled parts.Analyze the image preprocessing algorithm,design the chatter image preprocessing process and algorithm,and provide high-quality image support for chatter image recognition.Secondly,the dynamic cutting model of thin-walled parts is established to analyze the characteristics of chatter.From the basic characteristics of the processed surface image(color,shape,texture,spatial relationship,etc.),the characteristics of optimal expression of chatter are selected.The features of the vibration image are extracted and characterized;the chatter image identification algorithm for milling thin-walled workpieces meeting the requirements of precision and real-time performance is studied,and the chatter image is recognized with high accuracy.Finally,based on the image acquisition system for thin-walled parts and the identification algorithm for milling chatter images based on texture features,an on-line monitoring system for milling chatter on thin-walled parts is established to realize on-line monitoring of chatter on thin-walled parts.Firstly,the design requirements and objectives of the online monitoring system are clarified;secondly,the system hardware components and work flow are analyzed;and the online monitoring system is set up;finally,the thin-walled part milling experiment is designed;based on the application examples,the verification algorithm and the scientific conception of this article are correct and reasonable.