Research On Gearbox Fault Diagnostics Based On Order Tracking

Gearbox is one of the most important components in transmission systems, gear failures in transmission systems will cause inability of power transmission, and then result in catastrophic accident. Condition monitoring is an effective method to keep track of the condition of the machines, detect fault occurrence and decrease failure rate. Therefore condition monitoring of gearbox must be used. Many machines such as wind turbine and bucket wheel excavator normally operate under fluctuating load and speed during service. In this case, vibration signal is amplitude and frequency modulated to some extent. The fault diagnosis technique becomes more complicated when gearbox is subject to varying operating condition, hence, traditional signal processing technique is no longer capable to solve the problem. Due to this reason, it is urging to find methods for gearbox fault diagnosis under fluctuating load and speed conditions. This dissertation investigates the application of Time Synchronous Average Method in gear fault diagnosis and further studies Phase Demodulation Based Order Tracking to tackle the complexity under fluctuating speed. The main works of this dissertation are listed as follows:1) Major failure modes and mechanisms in gearbox are firstly studied and analyzed. Mathematic models of gearbox vibration signals are set up based on the characteristics of the vibration sources for later simulation. The possible signal signatures related to various failure circumstances are also analyzed and the analysis result is used for gear fault diagnosis.2) Analysis result of studying the principle of Time Synchronous Average Method shows that this method can effectively reduce random noise and eliminate the disturbance from other non-synchronous periodic signal and that there exists optimized synchronous average number can totally eliminate the non-synchronous periodic signal. Signature extraction methods based on Time Synchronous Average are analyzed, the effect of different signatures are verified by real measurement. This shows FM4, NA4and NB4can be used for early gear crack detection, NA4and NB4can monitor the damage progress.3) The high resolution of phase-time relationship based on phase demodulation is proven and therefore, in this dissertation, two phase demodulation based order tracking are investigated. Both simulation results and measured data shows these two methods can be used under a relatively high range of speed fluctuation and offer good order tracking results.4) Finally, the measurement data are analyzed and the result proves that the introduced Order Tracking Method and the Time Synchronous Average Method can provide accurate diagnosis of the tooth fault under fluctuating speed conditions.