The Design Of Tool Status Intelligent Monitoring System For CNC Machine

Tool condition monitoring is an important part of the urgent need for modern CNC machine tools.With the advancement of machining technology,tool condition monitoring plays an increasingly important role in CNC machining.Promptly finding tools in the excessive wear phase and then stopping the tool change is crucial for improving the productivity of the industrial site and ensuring the safety of the process.In this paper,the glass finishing machine is used as the experimental platform.According to the industry background that the current CNC machine tools are widely used in the repeated processing of single workpieces,the traditional intelligent monitoring of the traditional tool state often has misjudgment,and the spindle vibration and spindle are studied.The tool status monitoring of the inverter current signal proposes a tool condition monitoring system with higher judgment accuracy.Since each sensor signal has certain advantages and disadvantages in the actual monitoring process,in order to avoid the problem of inaccurate judgment caused by single sensor monitoring,this subject will comprehensively judge the spindle vibration signal and current signal during the machining process of the machine tool main tasks as follows.The tool wear mechanism is introduced,and the tool wear process is divided into three stages to facilitate the decision of the tool wear state.Based on the hardware experimental equipment to build the experimental platform and data acquisition system,for the design of the cutting experiment,select the appropriate position to install the vibration accelerometer and Hall current sensor to achieve the best data acquisition effect.The original signal collected by the sensor contains many feature information independent of the tool state,and denoising processing is needed to improve the signal-to-noise ratio.Then,the mathematical analysis method can be used to obtain the denoised signal and the tool state.Eigenvalues.In this paper,BP neural network and SVM multi-classification are used to learn the eigenvalues of each state of the tool,so that the decision-making algorithm has the ability to judge the state of the tool.Finally,by comparing the decisionmaking precision of the two algorithms,the most appropriate decision-making algorithm is selected.