Abstract: After analyzing the relationship between the warelet ridge and the instantaneous amplitude and frequency of the signal, a new ridge tracking algorithm which based on energy centrobaric ridge point defection method is proposed to estimate the instantaneous parameters of multi-component signal. A new parameter optimization method which changes the centre frequency of complex Morlet wavelet adaptively is brought forward to overcome the limitations that exist in the parameter optimization method with fixed center frequency, then the factors of complex Morlet wavelet are optimized through the new parameter optimization method and normalized wavelet scalogram is obtained. Cembining with local direction computation algorithm, energy centrobaric ridge point detection method is employed to track ridges and compute instantaneous parameters of signal. Simulation results show that the algorithm is insensible to noise, the result of fault diagnosis in gearbox and rolling bearing show that the features of fault vibration signals of rotating machinery can be extracted effectively by using the proposed method.

Key words: parameter optimization;ridge point detection;ridge tracking;instantaneous parameter