基于DT-CWT、粗糙集和神经网络的轴承智能诊断

对一种滚动轴承的智能故障诊断方法进行了研究,包括对偶树复小波变换(DT-CWT)特征提取、用粗糙集理论进行数据约简和神经网络进行模式识别。利用DT-CWT的平移不变性、粗糙集的数据约简以及神经网络的自动识别功能,达到尽量减少人工诊断的效果。理论和实际信号的研究表明:该智能诊断过程能初步达到在线自动诊断的目的。

Intelligent Diagnosis of Roller Bearing Using Dual-Tree Complex Wavelet Transform, Rough Set and Neural Network

An approach to intelligent diagnosis is studied, which is based on feature extraction by the Dual-Tree Complex Wavelet Transform (DT-CWT), then attribute reduction on the basis of rough set theory, and finally pattern recognition by Artificial Neural Network. DT-CWT has the advantage of approximate shift invariance while the decimated Discrete Wavelet Transform does not have. The method can improve the recognition efficiency and reduce the risks involved in manual diagnosis. The experimental results indicate that the proposed method can extract significant feature sets, and can accurately distinguish many fault patterns, and has some practical value for the on-line condition monitoring of modern industrial demands.