| Autor: |
Dejian Sun, Bing Wang, Xiong Hu, Wei Wang |
| Předmět: |
|
| Zdroj: |
International Journal of Acoustics & Vibration; Jun2019, Vol. 24 Issue 2, p312-309, 8p |
| Abstrakt: |
A fault diagnosis method using improved pattern spectrum (IPS) and FOASVM is proposed. Improved pattern spectrum is introduced for feature extraction by employing morphological erosion operator, and this feature is able to present fault information for roller bearing on different scales. Simulation analysis is processed and shows that, the value of IPS has a steady distinction among different fault types and the calculating amount is less than traditional method. After feature extraction, SVM with FOA, which can help with seeking optimal parameters, is employed for pattern recognition. Experiments were conducted, and the proposed method is verified by roller bearing vibration data including different fault types. The classification accuracy of the proposed approach on training is 87.5% (21 24) and reaches 91.7% (44 48) in a testing data set. The analysis shows that the method has a good diagnosis effect and an acceptable recognition result. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
