Fault Diagnosis Of Resonance Demodulation And Wavelet Analysis

The impulse signal is contained in the fault signals of some pivotal components such as gears and axletrees. Extracting weensy impact information is an important method to diagnose equipment. Resonance demodulation and wavelet analysis provide theories basis for extracting impulse signal form vibration signal.The math model of resonance demodulation is established in the paper and its impulse response is worked out. This provides theories basis for extracting impulse by hardware. This paper introduces basic theories of wavelet analysis and analyses the merits of continuous wavelet transform which is used fault diagnosis. Three methods of extracting impulse (continuous wavelet transform, discrete wavelet transform and envelope analysis based on maximum of wavelet modulus) are introduced. Every method is gave an applied example and analytic results are consistent.Using Winbond MCU (W77E58) and Sunplus MCU (SPCE061A) develops fault diagnosis apparatus based on resonance demodulation and wavelet analysis. W77E58 controls LCD, which constitute man-machine interaction interface. SPCE061A acquires data and algorithm of FFT and wavelet packet decomposition is implemented in it. The fault diagnosis apparatus effectively extracts impulse signal and its accuracy is proved by experiment.Resonance demodulation technology has succeeded in application on product line of rolling mill of high-speed wire in Wu Han and An Yang Steel Corporation and tremendous economic benefit has been gained. The fault diagnosis apparatus provides multifold diagnosis methods. These methods include resonance demodulation analysis, FFT analysis, wavelet decomposition and extracting impulse signal by wavelet. These functions enhance reliability of fault diagnosis. Hu Bei Education Department has identified the apparatus.