| Today, the gear transmission has been applied widely to the mechanical device in various trades of industry, agriculture, national defense, spaceflight and aviation as well as communications and transport. It can be said that gear transmission is mechanical system which is most widely used and gear is also the mechanical part which is damaged easiest. Therefore, it is important how to detect the fault of geared system as early as possible to make the repair schedule reasonably and economically and to avoid the body hurt and device damage.When the mechanical device is running, motion and force is transmitted through engagement between the driving gear and driven gear in geared system, and the mechanical vibration must be generated. The state of mechanical vibration can be changed by wear, pit, the error of manufacture and installation as well as various defaults of gear. Therefore, vibration signal of gear transmission is information carrier that displays the condition of gear. It contains ample information on normal and abnormal of gear transmission. Various faults of gear and geared system can be diagnosed naturally by monitoring and analyzing vibrational signal.During operating device, operator hopes that the information on initial fault can be obtained as early as possible. But, the vibration motivation from initial fault is much weaker than normal vibration motivation. The path of vibrational spread is complicated (It includes stronger interfering noises and nonlinearity of vibrational spread). The information in vibration signal is ample and frequency component of signal is complicated, useful information of faultbeing submerged in other frequency components and measurement noises, so it is very difficult to diagnose initial fault of gear. Moreover, the problems of multi-fault signal being separated in the multi-fault diagnosis, fault localization and estimation of damaged degree are related to signal analysis. But present information-process technology of mechanical fault diagnosis is only suitable to distributing fault of gear (e.g. severe surface wear), not to initial fault and local fault of gear. The problem "How can characteristic information of gear fault is separated and extracted effectively?" is the task of this study. Some of new methods that are used to diagnose gear fault are developed based on digital signal process technology for random variable. It is emphasized in the thesis that study on moving cepstrum analysis based on torsional vibration measurement, cyclic spectral density of cyclostationary signal, neural networks based on energy analysis of wavelet packet , independent componentdecompose as well as empirical mode decomposition and Hilbert--Huangtransform are applied to gear fault diagnosis.In chapter one, it is clarified that significance of the study, present information-process technology of mechanical fault diagnosis at home and abroad and main task which is completed in the thesis.In the chapter two, first, mathematical models of two kinds of gear faultsin one-stage gearbox (the fatigue crack on root of tooth and the local spalling on the flank of tooth) are established on the basis of gear dynamics. Then, the boundary value problems of ordinary differential equations that are governed by the mathematical models are solved by the numerical method. Last, two kinds of gear fault are numerically simulated so that influence of gear fault to change of vibrational state is studied in theory, and the symptom generated by faults is found.The path of vibration spread is from gear to shaft, then to bearing and finally to the surface of box. Therefore, vibration signal of gear transmission is information carrier that displays the condition of gear. It has a function of magnifying or attenuating the frequency content of signal. The influence ofpath of vibration spread on vibrational signal measured is studied by theory analysis and measuring transfer function in chapter three. The vibration signal on the measured point of gearbox surface is formulated with impulse r...
|
PDF Full Text Request