Research On Chatter Detection Technology For Turning

Chatter,which is a form of common machine tool vibration,restricts the development of high precision and efficiency of machine tool industry seriously.At the same time,the chatter severely accelerates the tool wear,shortens the life of the machine and brings the damage.Supported by the National Natural Science Foundation of China,the problems of chatter detection technology are studied,and a new method for chatter detection based on time-frequency entropy and frequency ratio of HHT is proposed.The specific contents are as follows:(1)First of all,the significance,background and current situation of the chatter detection for machine tools are introduced,and the improvement and innovation of the chatter detection technology in recent years are summarized.The selection of multiple sensor signals and the feature extraction based on different signal analysis are two key techniques in the chatter detection technology.(2)The basic problem of the research on the field of cutting regenerative chatter is introduced next.Such as the mechanism of cutting regenerative chatter,the dynamic model,the inherent properties,the response analysis etc.At the same time,the main direction of the research field,such as the prediction of cutting stability and the technology of chatter control,is introduced.Moreover,the demand for the development of intelligent chatter control,which makes the real-time feedback of the chatter state become very important,is imperative for the chatter detection.(3)The theory and practice of the vibration testing technology are introduced in detail in this paper.According to the overall process of the chatter test,the selection of hardware,the setting of the software parameter,the installation and placement of the sensor,the technical guidance and experience of the key link are given.Among the signal analysis methods,the defects of the FFT,the STFT and WDT are discussed,and the special application of Hilbert-Huang Transform in signal analysis is introduced in detail.In the end,the experimental data of ETC lathe are combined with the different analysis methods,and the advantages and disadvantages are summarized in the table.(4)An effective identification method of turning chatter is proposed in the core of this paper.The signal after the filtering and the Hilbert-Huang transform is describe in the time-frequency chart and the extreme energy distribution chart.Extract the chatter discrimination parameters-time-frequency entropy and frequency ratio from three characteristics,the increase of energy,the stability of frequency and the continuity of time,and it solves the problems of the traditional chatter discriminant of mean and variance,skewness and kurtosis.Firstly,the discriminant method of the variance of the Hilbert Huang transform spectrum is optimized.Using the time-frequency entropy instead of the statistical variance to solve the problem of the difficulty in the setting of the variance threshold,and distinguishing chatter from the characteristic of the concentrated and strong in progress.Secondly,since the chatter erroneous judgements caused by some periodic jamming signals,the frequency ratio,which is the parameter in the time,is presented,and it improves the accuracy and reliability of the results.After verification of the simulation experiment,the experiment of the ordinary lathe and ETC lathe,it can be proved that the method can effectively identify the stability,chatter and periodic jamming signal.The algorithm is simple and the result is obvious.As a consequence,the method can apply to chatter detection with practical value.(5)Finally,an innovative chatter testing system based on LabVIEW and NI data acquisition equipment is developed.Prediction method of chatter stability based on Tlusty(Stability Lobes Diagram)and the chatter detection algorithm based on time-frequency entropy and frequency ratio proposed in this paper are integrated into the system.The predictive interface of the chatter stability and the real-time monitoring interface for chatter is designed and developed.The results are consistent with the results of the analysis and processing of the signal in real processing.