Study on Degradation State Recognition for Rotating Machinery Early Fault Based on PCA-FCM

摘要：

Fault diagnosis commonly only carries out the recognition between the fault and the normal states not include the different states classification of the same fault, which is a problem of the fuzzy degradation process. PCA for rotating machinery the early fault feature extraction and the application of FCM for different fault states recognition are mainly introduced to solve the above problem in the paper. Collecting the rotating machinery shaft misalignment signal based on the experiments, through the timedomain analysis, PCA is carried out to obtain PCs which reflect the changes of time domain eigenvalues; and then use FCM algorithm to cluster these eigenvalues. By using fuzzy closeness degree to recognize the different fault states, calculate Euclidean distance between each sample and fault clustering centers in the paper, thus we can obtain the results of diagnosis. The diagnosis results show that the method proposed in the paper can identify the different states of the same fault.

关键词： component Rotating machine early fault diagnosis PCA FCM

作者: Liang Wei Sun Xiaoyu Zhang Laibin

作者单位: College of Mechanical and Transportation Engineering,China University of Petroleum (Beijing) Beijing, China

会议类型: 国际会议

会议名称: 2011 Seventh International Conference on Natural Computation(第七届自然计算国际会议 ICNC 2011)

会议地点: 上海

会议语种:英文

页码: 1712-1716

在线出版日期: 2011-07-26（万方平台首次上网日期，不代表论文的发表时间）