Development Of System Identification Module For Three Axis Engraving And Milling Machine Based On PMAC

The dynamic characteristics of the machine tool feed system directly affect the motion control performance and processing accuracy of the CNC machine tool.The dynamic characteristics of the machine tool feed system are determined by the dynamic characteristics of the mechanical transmission mechanism of the feed system and the structure and parameters of the controller.The control parameters must Matching with the dynamic characteristics of the mechanical transmission mechanism can obtain the ideal motion control performance.At present,the numerical control system is unable to determine the dynamic parameters of the mechanical system,adjust the control performance,and monitor the operating status of the mechanical components.This research takes the three-axis engraving and milling machine as the research object,and develops the system identification module based on the PMAC open CNC system,which is used to obtain the dynamic parameters of the mechanical transmission mechanism conveniently,quickly and accurately,and analyze the linear characteristics and nonlinear characteristics of the identification parameters.Characteristics and time-varying characteristics,tuning controller parameters,predicting machine failures,etc.This research has important theoretical research significance and engineering application value for expanding the functions of the CNC system and improving the intelligent level of CNC machine tools.Aiming at the problem of low accuracy and low efficiency in the off-line identification of dynamic parameters of mechanical transmission,an off-line identification method of dynamic parameters based on dual excitation signals is proposed and implemented,which simplifies the mechanical transmission to a dual inertia model,and each component corresponds to an equivalent According to mechanical parameters,first use pseudo-random binary excitation signal to identify the frequency domain characteristics of the mechanical transmission mechanism,design the sweep signal according to the frequency domain characteristics of the mechanical transmission mechanism,and then use the sweep frequency signal designed based on the frequency domain characteristics to identify the power of the mechanical transmission mechanism The simulation and experiment of off-line identification of dynamic parameters are carried out,which improves the accuracy of off-line identification of dynamic parameters of mechanical transmission mechanism,simplifies the design steps of frequency sweep signal,and improves the efficiency of off-line identification of dynamic parameters of mechanical transmission mechanism.Aiming at the problem that the online identification of dynamic parameters of mechanical transmission is greatly interfered by disturbance signals and the identification accuracy is low,an offline identification method of dynamic parameters based on the extended closed-loop output error method of "object + object correlation disturbance" is proposed and implemented.Establish the control structure of the mechanical transmission mechanism and the controller,adopt the closed-loop direct online identification method,determine the structure of the disturbance item according to the to-be-identified matrix of the mechanical transmission mechanism,reduce the influence of the disturbance signal on the identification accuracy,and carry out the dynamic parameter online identification experiment,The experimental results show that the online identification method in this paper can achieve accurate identification in both straight and circular processing conditions,and improve the accuracy of online identification of dynamic parameters.Aiming at the problems of low accuracy and slow convergence speed in the online identification of nonlinear friction,an online identification method of nonlinear friction based on an improved stochastic gradient descent method is proposed and implemented.An RBF neural network model is established to describe the nonlinear friction and reduce The influence of the integration link in the traditional Lu Gre friction model on the running time of the algorithm.Based on the posterior error,the stochastic gradient descent method is improved,so that the weight update function of the neural network is not affected by the weight adjustment gain,and the sensitivity of the algorithm to speed noise is reduced.,The simulation and experiment of online identification of nonlinear friction force were carried out.The experimental results show that the method of online identification of nonlinear friction force in this paper can effectively compensate for the nonlinear influence caused by friction force,and improve the accuracy and convergence speed of online identification of nonlinear friction force.,Improve the low-speed performance of the feed axis.Aiming at the lack of recognition ability of the CNC system,the CNC system is developed based on the PMAC open motion controller.First,the basic processing module of the CNC system is realized,so that the CNC system has the ability to normally recognize the processing code for automatic processing and manual processing.,Developed the system identification module of the CNC system,and applied the algorithm of this paper to the CNC system,so that the system can quickly and accurately identify the dynamic characteristics and nonlinear friction of the mechanical transmission mechanism,and then quickly tune the controller to improve the dynamic characteristics of the machine tool feed system Finally,the trajectory error test module of the numerical control system was developed,which enables the machine tool to judge the dynamic characteristics of the feed system better.