Rough neural networks-based fault diagnostic system for rotary machinery
Shulin Liu · 2002
This paper presents a rough neural networks-based approach for fault diagnosis of rotary machinery. This approach can directly extract diagnosis parameters for rotary machinery from the vibration signals processed by the wavelet packet transform. These data are used to construct rough neural networks based fault diagnosis system for rotary machinery. The accurate diagnosis results can be obtained directly from the set of complete fault signals, and the satisfactory diagnosis results can also be derived from the set of incomplete fault signals by using the approach. The mean accuracy rate of diagnosis exceeds 90%.