Research On Fault Diagnosis Method Of Cnc Machine Tools Based On Second Generation Vold-kalman Filter Order Analysis

CNC machine tools are the foundation of modern manufacturing,which directly affects the country’s economic development and national comprehensive strength,reflects a country’s manufacturing level,and plays an important role in machinery,hydropower,aerospace and other industries.Once the CNC machine tool fails,it will not only affect the production efficiency,but also cause the equipment to stop running or cause huge economic losses in severe cases.Therefore,it is necessary to perform condition monitoring and fault diagnosis on key components such as CNC machine tool bearings and ball screw pairs.In this paper,based on the second-generation Kalman filter order tracking method,the calculation method of the speed curve,the drawing of the three-dimensional order spectrum,the extraction of the fault feature order components are studied.Development.The specific research content is as follows:Firstly,the speed calculation based on the current signal of the CNC machine tool is realized.Speed calculation is the basis of order analysis and the key to realize order tracking.For the case of a speed pulse signal,a speed curve calculation method based on first-order digital differentiation is implemented.For the case where there is no tachometer and it is inconvenient to install the tachometer,the first-order digital differential and sine function fitting are used to realize the calculation of the speed based on the current signal of the CNC machine tool.For the case of only vibration signal,the calculation of speed curve based on time-frequency fusion is realized.The results show that the accuracy of the speed curve calculated by the current signal is better than the accuracy of the speed curve calculated by the tachometer calculation and instantaneous frequency estimation,and it is not limited by the structure of the measured object.Secondly,based on the second-generation Kalman filter order tracking method,the time-domain extraction of order components and the decoupling of multi-axis cross orders are realized.The results show that this method can directly separate and extract each order component from the time-domain signal.The coupling order generated by the multi-axis linkage of CNC machine tools can be accurately decoupled and extracted.Thirdly,through the speed curve calculated from the current signal,the threedimensional order spectrum drawing without tachometer and the fault feature order extraction are realized.The results show that the efficiency and order resolution of the three-dimensional order spectrum based on the second-generation Kalman filter order analysis method are higher than those of the first-generation Kalman filter order analysis method.In addition,the three-dimensional order spectrum can accurately find the fault feature order and natural frequency of CNC machine tools.Identify the degradation of the preload of the ball screw by identifying the change of the order of the balls.Finally,developed a CNC machine tool fault diagnosis system based on order analysis.Its main functions include: parameter setting,data collection,characteristic order calculation,speed-up and down order analysis,real-time order analysis,fault feature order component extraction and other functions.The system interface is friendly and easy to operate.