科研之友
微信
新浪微博
Facebook
分享链接
摘要
Reliable fault detection of bearing faults is crucial to avoid costly machine failures. To this end, many methods have been proposed over the years. However, most of them are dependent on properly selected parameters, such as the center frequency and the bandwidth of the bandpass filter for the pass band in the high-frequency resonance method. Such parameters may also have to be updated in a variable operating environment which may not always be possible without the involvement of domain experts. As such, we propose a new non-parametric and non-filtering method which can perform reasonably well in the presence of vibration interference and noise. This method is based on the derivation of amplitude demodulation of a signal to suppress unwanted periodic interference components in the acquired signal. It has been shown that the proposed method can boost the signal-to-interference ratio by up to 2-12 times compared with the energy operator approach and can handle signals tinted by both noise and multiple interference. It compares favorably with other non-parameter methods, such as the plain envelope method and energy operator method. Its effectiveness has further been demonstrated using both simulated and experimental data.
- 出版日期2013-10
