Study On Fault Diagnosis Correlation Of Gear System Based On Tribology And Dynamics

Gear is widely used in the agricultural and industrial equipments as the general element of transmitting power and motion in the machinery. For the tribological system of a gearbox, many scholars attempted to solve the consistency and accuracy through the relationship of fault diagnosis based on wear and vibration information while the wear and vibration analysis were used in the condition monitoring and fault diagnosis. However, there is no positive progress in how to correlate wear and vibration information and what extend of correlation can be suitable for the fault diagnosis of gearboxes. Therefore, the gearbox is selected as the research object to study the relationship of tribological information and dynamics information, as well as the wear and vibration characteristic information and fault diagnosis correlation based on wear information and vibration information in this thesis.Firstly, from the point of tribological and dynamic information during the process of gear meshing motion, the theoretical correlation of tribological and dynamic information from the gear system was represented in detail and the implement process of the correlation investigation was proposed. Based on the previous research, considering the different friction information in the dynamics simulation method, physical models of gear meshing simulation were established as the object of the automobile gear train, and the dynamics characteristics were simulated totally under the condition of the different friction coefficients. By analyzing the simulation results, the tribological and dynamic information of the gear system showed the strong coupling relationship when both were mutual influenced during the process of the output status characteristics of the gear tribological system.Secondly, from the point that generation of wear debris was the product of wear energy loss, the fault diagnosis model of the gear tribological system was established using the parameters of energy loss. The experimental study showed the gear tribological system energy consumption decreased with the increase of load, and increased with the increase of the rotating speed, but the energy loss still increased with the increase of load when the tooth surface fault was very severe, such as the severe spalling fault tooth surface.Next, on the basis of the common characteristics that both wear and vibration of gears consume the system energy, the fault diagnosis correlation model was established based on wear and vibration of the gear tribological system. Two characteristic parameters of fault diagnosis information were proposed, which wear cumulative relative scale (WCRS) of wear element relative mass fraction and vibration cumulative relative scale (VCRS) of root mean square of time-dome speed signals. The proposed model was verified using the fatigue wear and vibration testing data of new gears and WCRS and VCRS parameters. The results show that there is the high correlation of the faults of a gear tribological system based on wear information and vibration information, which indicates that the proposed theory and methods in this thesis are feasible.Finally, through designing the method of wear and vibration tests of the simulated fault gears under the different conditions and fault types, wear and vibration characteristics, and the testing relationship of fault diagnosis were investigated in detail based on wear and vibration information when the gears broke down. The characteristic information obtained from the tests showed that the changes of gear vibration signal characteristics were similar with those of the wear characteristics. The interrelationship of gear wear and vibration information during fault diagnosis shows that the increase of wear quantity was not only related with the pitting degree of tooth surface but also with the number of pitting teeth while the occurrence of pitting tooth surface. However, the vibration intensity of the gear was the most relevant with the pitting degree of tooth surfaces, but not close related with number of the pitting gear teeth. When the spall occurred in the tooth surface, the more serious the spall of tooth surface was, the more severe wear was, and the more intensive vibration was. Accordingly, the fault diagnosis results are consistent using both characteristics information in the gear. Further study on the correlation was provided that the changes of vibration signal were more sensitive than the changes of wear when the tooth surfaces of the gear broke down.