Condition Monitoring Research Of Tools Based On Stochastic Resonance Method

The advanced technology of processing and manufacturing technology will directly affect the quality of a country's economic development.The quality of CNC machine tools is directly related to the quality of products and production efficiency.The diagnosis of machine tool faults has always been a hot topic in the engineering field.The current research direction is how to realize online monitoring of the tool status and accurately determine the cause of the fault,which has important significance for improving product quality and production efficiency.In the field of tool fault diagnosis,the traditional analysis method is to obtain the vibration signal during processing,and analyze the vibration signal characteristics in time domain and frequency domain.It has achieved good results and has been widely used,but the traditional analysis method cannot accurately detect the weak signals of the tool at the initial stage of the fault.Since the stochastic resonance method has a good effect on extracting the characteristics of weak signals,how to extract the weak signals at the initial stage of the fault and timely discover the tool fault information is an important part of this paper.Firstly,the development of tool condition monitoring method and the application of stochastic resonance method in vibration signal analysis are described in detail.In the second chapter,the vibration signals at the early,middle and later stages of tool wear are analyzed by the traditional analysis method.After the vibration signal is filtered and de-noised,spectrum features and power spectrum features are extracted.The third chapter verifies the role of stochastic resonance method in extracting weak signal features through simulation experiments.According to the relation between SNR and noise intensity,when the noise intensity reaches half of the barrier height,SNR can reach its peak.After the scale transformation,the classical SR system successfully detects the spindle base frequency and the endmill rotation frequency.The fourth chapter uses the method that combines frequency-shifted,re-scale and adaptive SR.The appropriate adaptive SR system parameters are found according to highest spectral peak and the variance of the zero-crossing distance.Then the obtained tool vibration signal is inputted into the adaptive frequency-shifted and re-scaling SR system,and the spindle base frequency and endmill rotation frequency characteristic informations are detected accurately.Through the study of this paper,it is shown that the classical bistable SR system is feasible to carry out online monitoring of tool wear state through scale transformation or frequency-shifted and re-scale,which provides a better method to realize on-line monitoring of early tool fault.