MECHANICAL FAULT DIAGNOSIS BASED ON FUZZY CLUSTERING ALGORITHM

2018 International Journal of Mechatronics and Applied Mechanics  
To improve the accuracy of mechanical fault diagnosis, based on the full vector spectrum analysis and fuzzy clustering algorithm, two kinds of improved FCM algorithms were proposed, namely vector -DKFCM (fuzzy kernel clustering based on density function) and vector -UGAFCM (uniform genetic fuzzy clustering) intelligent fault diagnosis method. Meanwhile, the new mathematical model and its concrete realization flow were given, and the effectiveness of the new method was verified. Moreover,
more » ... the cluster validity index is domain correlation, the validity index for classification of rotating machinery common fault was compared and analyzed. The experimental method was and the results showed that three fuzzy clustering validity indexes VXB, VKwon and VCWB applicable to the field of rotating machinery fault diagnosis were obtained. In summary, the fuzzy clustering algorithm, compared with traditional method, has higher classification accuracy.
doi:10.17683/ijomam/issue3.10 fatcat:butxlwrrtbf4zki7rlhs5lenca