Study Of Vibration Fault Diagnosis On Low Speed And Heavy-duty Gear

Introducing the meaning, purpose, research status, and developing trend of gear fault diagnosis. Analyze the common main fault type and reason. Study some kinds of typical fault signal character of low speed and heavy-duty gear and common gear fault diagnosis signal processing method. Vibration and fault mechanics of low speed and heavy-duty gearbox are reviewed, in the thesis. Combined with traditional vibration signal processing method in time and frequency domain which are FFT, Cepstrum analysis, Zooming analysis and Sideband analysis of gear and soon on, vibration signal demodulation is used in fault diagnosis of coal mine elevator gearbox. Fault character of elevator gearbox are extracted. To the modulate phenomena of gear vibration signal, study square demodulation, Hilbert transform demodulation method and their application in low speed and heavy-duty gear fault diagnosis.Demodulating the demodulation information of fault gear from complicated vibration signal. Finally study the support vector machine in detail, it is a machine learning arithmetic especially aim at small samples. discuss the basic principle of support vector machine classifying and several multi-class support vector machine arithmetic, and use'one against one'arithmetic which is adapt to machine system multi–fault classifying realize elevator gearbox multi–fault classifying, the result shows good classifying effect, provides us with a new method to develop the intelligent fault diagnosis of gear. The research shows that support vector machine applying to mine machine fault diagnosis field has important meaning.