Rotating Machinery Diagnosis Using Wavelet Packets-Fractal Technology and Neural Networks
| Autor: | Chin-Hao Chen, 陳志豪 |
|---|---|
| Rok vydání: | 2007 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 95 The faults feature extraction is one of the most important research topics in the field of mechanical fault diagnosis. It is a issue which hinders the mechanical fault diagnosis technique from further improvement. Therefore, in this study, the problem of faults feature extraction and diagnosis was addressed using the signal processing technology. The methods combine signal processing tecdhnique with neural network to present a new fault diagnosis procedure for rotating. In the thesis divides seven chapter, at first, it gives a general view of fault diagnosis for rotating mechanical including developments and present situations. Power cepstrum, bispectrum and wavelet transform methods were discussed, and overview to the fractal dimension and neural network were also included. These methods are then used to perform faults diagnoses of rotational machinery systems. Form the results, it is shown that a combination of the these signal analysis tools give a more reliable condition monitoring method for rotary machinery. When faults occur they usually produce nonstationary vibration signals, by using wavelet packets transform on these signals, the fractal dimension of each frequency bands is extracted and the box dimension is used to depict the failure characteristics of vibration signals. Then the failure modes can be classified by radial basis function neural network. Experiments were conducted and the results shown that the proposed method can detect and recognize different kinds of faults in rotating machinery. Therefore, it is concluded that the wavelet packets-fractal technology combined with neural network method can provide an effective way to diagnosis faults in mechanical systems. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|