The Research On Fault Diagnosis Of Servo System On CNC Milling Machine

CNC machine tool as the cornerstone of modern industrial and level of development level has become a national manufacturing industry modernization degree of central symbol, its development level directly determines the machinery, aerospace, automobiles, ships, buildings and other pillar industries and energy, transportation and other basic industries. CNC machine tool servo system is a key part of CNC machine tool, its performance has a direct impact on the accuracy of the numerical control machine tool operation and processing, workbench moving speed and tracking precision of the technical indicators. Faults of the servo system of CNC machine tool, small influence machining precision, damage to the machine, will endanger the personal safety, restricting the development of national industrial base. Under the support of the Jilin Provincial Department of education project "Research on the fault diagnosis and prediction technology of precision mechanical and electrical systems", the fault diagnosis system of the CNC milling machine servo system was studied.Firstly, the servo system of CNC milling machine is studied as the research object, and the working principle of the servo system, the common failure mode and the cause of the fault are studied deeply. Mainly from the servo system drive system, transmission mechanism, detection device three most of the analysis of its failure mechanism.Secondly, the servo system of CNC milling machine fault mechanism analysis based on fault diagnosis system. The key technology of fault signal nonlinear interference problem, select the fault sensitive parts of the installation of the acceleration sensor signal acquisition, using time-frequency signal processing method of the original signal adaptive denoising and feature extraction and processing. And considering the complex nonlinear relationship between the fault and the fault mode of the neural network, a large amount of data, high quality requirements, lack of explanation mechanism, the establishment of fault location and fault type identification of two neural networks, combined with the servo system related working state, fault diagnosis method of the regular input products rough, build the fault the diagnosis system of CNC milling machine servo system. using a two-stage neural network and the theory of BRB, the EEMD nonlinear signal analysis and feature extraction, combination of two neural networks with BRB theory and neural network to solve the explanation mechanism, the effective realization of fault diagnosis to many parts of the servo system of CNC milling machine, and to provide a reference for other complex electromechanical system fault diagnosis system.Finally, the overall design of the data acquisition, processing, diagnosis and identify the servo system fault diagnosis system, with CNC milling machine servo system of experimental verification. Based on LabVIEW servo system fault diagnosis system man-machine interface, to bearing fault of servo system as an example of fault diagnosis example analysis. Simulation and experimental results show that based on two level neural network with confidence rule base servo system fault diagnosis system can effectively of the servo system of the CNC milling machine for fault diagnosis.