Research On Network Monitoring And Energy Consumption Modeling Of NC Machine Tool Based On MTConnect

As the manufacturing industry continues to grow in size,the annual energy consumption of manufacturing industries increases year by year.As the “working machine” of the manufacturing industry,CNC machine tools are not only an important part of the energy consumption in the production process,but also the focus of attention of many scholars.This article carries on the thorough analysis to the numerical control machine tool data monitoring present situation,constructs the machine tool network monitoring platform according to the MTConnect agreement,simultaneously carries on the modeling and the prediction to the machine tool energy consumption,the main research content includes:(1)By studying the current status of numerical control machine tool data monitoring,the network monitoring architecture of CNC machine tools is determined,and the composition and workflow of the MTConnect protocol are analyzed in detail.According to the machine tool information modeling description language,the CNC machine tool for performing experiments is completed.Construction of Extensible Markup Language(XML)Information Model.(2)The overall design of the data monitoring platform for numerical control machine tools was analyzed.The current major techniques for machine tool data collection were analyzed.The main functions of the MTConnect adapter were realized by using the secondary development kit and power sensor of the Fanuc CNC machine tool.The language develops the agent and client part of MTConnect on the PC platform,which realizes the remote monitoring of CNC machine tool data.(3)The energy composition and distribution characteristics of NC machine tools are studied.The power balance equations of NC milling are established by the energy flow of machine tools,and the optimization of support vector machine regression based on improved Particle Swarm Optimization(PSO)is established.Vector Regression(SVR)energy prediction model.The PSO algorithm is used to optimize the penalty factor and kernel function parameters in the SVR parameters to improve the energy prediction accuracy of the SVR model.(4)Using CNC machine tool data monitoring system to monitor the machine tool data,and using the optimized SVM regression model to predict the energy consumption of the monitored cutting data and analyze the prediction error.The result validates the validity of the model.