Research On Highlight Machining Tool Condition Monitoring Based On Clustering By Fast Search And Find Of Density Peaks

In the CNC machining tool will inevitably wear,which have a great impact on the product processing quality and efficiency.In the metal mobile phone shell high-light mass production tool life is poorly consistent.tool wear will make the product surface blue line,white line,fog,drawing and other defects,if you can not find the state of the tool and timely replacement of the tool,it will lead to degradation of processed products or even scrapped.At present,the identification of the state of the highlight chamfering tool mainly depends on the manual observation of the workpiece processing quality to determine the state of the tool,which is time-consuming and laborious and largely affected by the staff level of operation and work experience.Combined with the actual production situation and demand of metal mobile phone highlight processing,this paper carried out high light tool condition monitoring research,based on machine tool spindle current,vibration.Based on the research and development of tool condition monitoring technology at home and abroad,this paper summarizes the current tool condition monitoring technology,analyzes various signal processing and state feature extraction techniques,and sums up the shortcomings of the current tool condition monitoring technology.On the basis of detailed understanding of the batch processing flow and process characteristics of the metal mobile phone highlight processing,the processing characteristics of the high light tool are analyzed and the main factors influencing the quality of the high chamfering are listed.The general idea of high temperature chamfering tool condition monitoring research is developed,and the data acquisition experiment system is set up.The signal data of the five states of the tool are collected and processed,and the signal of the machining section is analyzed.The characteristics of the signal are analyzed from the perspective of time domain,frequency domain and similarity,and the difference of signal characteristics is compared with each other.The time domain,frequency domain and similarity feature are extracted to construct the eigenvector.Clustering by fast search and find of density peaks is used to establish various tool state classification models to realize the classification of various types of signals.The surface quality of the defective workpiece when the state.The differences in the surface morphology of the rake face and the flank face of the new,life span,blue line,white line and fog are analyzed.Finally achieve based on the actual state of the tool to replace the tool,improve product yield,extend tool life and reduce business costs.