Design And Implementation Of Real-time Acquisition And Monitoring System Of Machine Tool Status Based On Flink

With the gradual transformation of the manufacturing industry to the direction of high-precision technology,the importance of CNC machine tools in the production and operation of enterprises has become more and more prominent.Failure of machine tool components will cause the working conditions of the machine tool to change.If the machine tool cannot be correctly located and resolved in time,it may cause the machine tool to collapse and cause huge losses to the enterprise.The general machine tool monitoring system is often a stand-alone program,and the monitoring of the machine tool is one-to-one monitoring.In current enterprise production,multiple machine tools or even multiple workshops are often used for production.Therefore,one-to-one machine tool monitoring systems often lead to scattered management of machine tools,requiring regular inspections by staff.In addition,due to the development of the Internet of Things technology and the large number of applications of sensor technology in machine tools,the operating state of machine tools has gradually shown the "4V" big data characteristics,which gradually poses a challenge to the processing capabilities of single machines.In response to the above problems,this paper implements a real-time monitoring system for machine tool running status based on the Flink stream processing engine design.First of all,according to the component structure of the machine tool,analyze the location that is prone to failure,and put forward the status collection plan based on the communication protocol of the CNC system and the status collection plan of the external sensor according to the characteristics of these locations.Aiming at the problem of the mismatch between the actual data collection frequency and the data processing rate,Kafka is introduced as a data buffer.Then,use the Flink stream processing engine to build a data processing and state judgment module.Use Flink to judge the running status of machine tool components in real time,and use machine learning algorithms to judge the running status of machine tools.In addition,according to the characteristics of the machine tool operating data types,different storage schemes have been selected.In addition,based on the front-end and back-end Web development technology,the system management module,the running status monitoring module,the running data statistical analysis module,and the running data real-time display module are realized.Finally,based on the above research,the development of the real-time monitoring system of the machine tool operation status and the construction of the experimental environment are completed.Taking multiple CNC machine tools in the laboratory as experimental objects,the accuracy and real-time performance of the real-time monitoring system are verified.