Rotating machinery fault diagnosis using signal-adapted lifting scheme

Wavelet transform has been widely used for vibration-based machine fault diagnosis. However, it is a difficult task to choose or design appropriate wavelet or wavelets for a given application. In this paper, a new signal-adapted lifting scheme for rotating machinery fault diagnosis is proposed, which allows us to construct a wavelet directly from the statistics of a given signal. The prediction operator based on genetic algorithms is designed to maximize the kurtosis of detail signal produced by the lifting scheme, and the update operator is designed to minimize a reconstruction error. The signal-adapted lifting scheme is applied to analyze bearing and gearbox vibration signals. The conventional diagnosis techniques and non-adaptive lifting scheme are also used to analyze the same signals for comparison. The results demonstrate that the signal-adapted lifting scheme is more effective in extracting inherent fault features from complex vibration signals.