Research On Overcomplete Rational Dilation Discrete Wavelet Transform And Application For Rolling Bear

Feature extraction, as the core technology of pattern recognition and machine fault diagnosis, has been a widely concerned topic. How to extract the fault features contained in the vibration signals has become a challenging problem, due to the vibration signal of rolling bearing’s non-stationary and the influence of strong noise. In this paper, we first study the overcomplete rational wavelet transform and then the fault feature extraction method based on overcomplete rational wavelet transform is analyzed. The research focus on the fault feature extraction method. In the end, we put forward two methods of fault feature extraction, one is based on signal-adapted overcomplete rational dilation discrete wavelet transform and another one is based on dual-tree rational-dilation complex wavelet transform. The main results are as follows:① A fault feature extraction technique based on signal-adapted overcomplete rational dilation discrete wavelet transform is proposed in this paper which allows us to construct a wavelet directly from the statistics of a given signal and then decompose the Input signal to various high frequency band signals by the wavelet bases. Subsequently compute the kurtosis values of the all the high frequency band signals. Then select the optimal signal bands based on maximization of kurtosis value. Finally, the fault features of the optimal signal band is detected through its Hilbert instantaneous frequency spectrum. The experimental results demonstrate that the feature extraction technique successfully identifies the incipient fault features.② In order to improve the recognition accuracy of SVM classification, a fault diagnosis method was proposed based on dual-tree rational dilation complex wavelet transform and support vector machine(SVM), according to the characteristics of rolling bearing fault vibration signal. Firstly, decompose the fault signal into several different frequency band components through dual-tree rational-dilation complex wavelet transform. Secondly, normalization processing was made from the energy of each component. Finally, the energy characteristics parameters of each frequency band component were taken as input of the SVM to identify the fault type of rolling bearing.The experimental results prove that the proposed method can identify the fault type accurately and effectively.