Research On Fault Diagnosis Of Rolling Bearing Of Gearbox Of Machine Tools

Gearbox is an important component of the machine tool transmission system,due to the complex installation space of most of the machine tool gearbox,many structural components,while the gearbox is very susceptible to temperature,lubrication and other external factors when working,especially when the machine tool is running for a long time,the gearbox rolling bearing is very prone to failure,but due to the characteristics of rolling bearing installation can not be found in the early stage of the failure,the accumulation of time will affect the operating state of the entire machine tool and even cause harm to the personal safety of the machine tool operator.Therefore,how to monitor the operating state in the rolling bearings of the gearbox of the machine tool equipment and determine whether the bearing is in a normal operating state is of great significance to the performance and processing accuracy of the machine tool and the personal safety of the operator.Taking the machine tool gearbox as the research object,this thesis proposes a gearbox fault diagnosis method based on Empirical Mode Decomposition(EMD)and Variational Mode Decomposition(VMD),focusing on noise reduction and fault signal recognition in the process of gearbox rolling bearing fault signal extraction.First of all,the research background and significance of gearbox fault diagnosis are introduced,the common failure forms of rolling bearings in machine tool gearboxes and their fault frequency step-by-step frequency domain are introduced,and the commonly used rolling bearing fault detection methods are introduced.The application and detection method of empirical modal decomposition(EMD)in the fault detection of rolling bearings in the machine tool gearbox are mainly proposed,and the application of empirical modal decomposition in the actual production of machine tool gearbox rolling bearings is verified by simulated signals and fault signals,which proves the characteristics of empirical modal decomposition in the fault diagnosis of rolling bearings in machine tool gearbox.Secondly,the application and detection method of the variational modal decomposition(VMD)algorithm based on parameter optimization in the detection of rolling bearing failure in the machine tool gearbox are proposed,and the machine tool gearbox with rolling bearing failure is detected,and the motion status and fault information of the faulty bearing are monitored by the actual monitoring of the faulty bearing,and at the same time,the application of the experience modal decomposition is verified by the simulation signal and the fault signal in the actual production of the fault diagnosis of the rolling bearing of the machine tool gearbox.Further study and understand the methods and characteristics of fault extraction based on parameter optimization variational modal decomposition in the fault diagnosis of rolling bearings in machine tool gearboxes.Finally,the application of the machine tool gearbox rolling bearing fault diagnosis method based on convolutional neural network(CNN)in the actual production equipment is proposed.Since the convolutional neural network itself has a good ability to calculate and extract fault features,the extracted rolling bearing vibration signal can be directly processed,this chapter first introduces the composition of CNN,algorithm and other theoretical basis,and then uses the modal components decomposed by the previous algorithm to develop the test process,and the four models of CNN,EMD-CNN,EEMD-CNNN,VMD-CNN are compared and analyzed in practical applications of the machine tool gearbox rolling bearing fault detection model.