Research On The Fault Diagnosis Of Gearbox With Compound Faults

Gearbox is an indispensible power transmission component, whose working condition will directly affect the performance of the whole mechanical equipment. Therefore, the research on the fault diagnosis technology for gearbox is of great significance to the normal operation of mechanical equipment. Extracting fault feature information from the gearbox’s vibration signal by using various kinds of signal processing methods has been the key to the fault diagnosis for gearbox.Hundreds of engineering practices show that there is usually more than one fault in a mechanical unit, which demonstrate as compound faults. The compound faults in different positions,, with different failure modes and degrees have different effects onto the mechanical equipment. Besides, the interaction and the mutual interference among the different fault components, especially the overlap of fault features under changing rotating speed situation, bring great challenge to the comprehensive fault diagnosis of mechanical equipment, thus, the compound fault diagnosis has become a hard problem in the fault diagnosis of mechanical equipments. Aiming at the problems above, sponsored by National Natural Science Foundation (project number:51275161) and Hunan Science and Technology Plan (project number:2012SK3184), taking gearbox as the research object, modern signal processing methods as the research tool, and compound fault diagnosis as the research target, the dissertation has carried on a profound and systematic research mainly on the fault features separation and the fault characteristics extraction from the vibration signal of a rotating speed changing gearbox with compound faults.The main research work and innovative achievements of the dissertation are as follows(1) On the basis of failure ratio analysis among the components in gearbox, the failure modes, reasons, appearances and vibration mechanism of the main failure components-gears and rolling bearings are analyzed, and the local fault vibration signal model of gears and rolling bearings have been set up. The study shows that, when a local fault occurs in a gear, there will be AM-FM component in the vibration signal, whereas, when a local fault occurs in rolling bearings, there will be periodic impulse component with damped oscillation.(2) Aiming at the separation of gear fault component and bearing fault component and the extraction of fault modulation information from the vibration signal of a gearbox with compound faults, a compound fault diagnosis method for gearbox based on morphological component analysis (MCA) and energy operator demodulation is proposed. Firstly, the fault components of gears and bearings are separated by MCA; then, energy operator demodulation is carried onto the separated fault components of gears and bearings to extract the fault modulation information. Algorithm simulation and application examples show that the fault component separation, followed by the energy operator demodulating, can effectively highlight the fault characteristic of each fault component.(3) Aiming at the separation and extraction of fault characteristics from the vibration signal of a rotation speed changing gearbox with compound faults, a compound fault diagnosis method for gearbox based on MCA and order tracking is proposed. Firstly, each fault component is separated by MCA; then, the separated fault components are transformed to angular domain signals through even angle resampling; finally, Hilbert envelop demodulation will be carried out onto each resampled fault component to extract the fault modulation information. By algorithm simulation and application examples, the local faults of gears and rolling bearings are analyzed. The results show that the proposed method can effectively separate the fault characteristics of gears and rolling bearings from the vibration signal of a gearbox under the condition of rotating speed changing.(4) Aiming at the unsuitableness of cyclostationary demodulating method in extracting fault modulation information from the vibration signal of a rotating speed changing gearbox with compound faults, a compound fault diagnosis method for gearbox based on chirplet path pursuit (CPP) and cyclostationary demodulating is proposed. Firstly, the rotating speed is estimated adaptively from the vibration signal by using CPP; then, with the estimated rotating speed, even angle resampling is carried on to the vibration signal; finally, the resampled angular domain signal is analyzed with cyclostationary demodulating to extract the fault modulation information. Algorithm simulation and application examples show that the proposed method, without the presence of a tachometer, can effectively extract the fault modulation information from the vibration signal of a rotating speed changing gearbox with compound faults.(5) When rotating speed is changing dramatically, analyzing gearbox vibration signal with ensemble empirical mode decomposition (EEMD) will result in mode confusion. Aiming at this problem, a compound fault diagnosis method for gearbox based on CPP and EEMD is proposed. Firstly, rotating speed information is extracted from the vibration signal of a rotating speed changing gearbox with compound faults using CPP; secondly, according to the extracted rotating speed, the vibration signal is resampled with even angle to obtain its angular domain signal; then, the intrinsic mode function (IMF) components can be acquired by the EEMD analysis to the angular domain signal, and the IMF component containing fault information is selected in terms of the correlation coefficient for IMF component and angular signal; finally, Hilbert envelop spectrum analysis is carried on to the IMF component selected to extract each fault modulation information. Algorithm simulation and application examples show that, under the condition of rotating speed changing, the proposed method can extract fault characteristics from the vibration signal of a gearbox with compound faults effectively.Currently, the compound fault diagnosis has been a hard problem in the fault diagnosis of mechanical equipments. This dissertation, taking gearbox as the research object, makes an exhaustive study on the fault diagnosis of gearbox with compound faults consisting of gear and rolling bearing local failures under constant rotating speed and changing rotating speed. Algorithm simulation and application examples indicate that the combination methods, obtained by combining MCA with energy operator demodulating, CPP, order tracking, cyclostationary demodulating, etc, take complementary advantage of each single signal processing method, and can effectively extract and separate the compound faults characteristics from the gearbox vibration signal. Therefore, the combination methods have a brilliant application prospect in the fault diagnosis of mechanical equipments.