基于RQA与GG聚类的滚动轴承故障识别

提出递归定量分析与GG聚类相结合的滚动轴承故障识别方法。利用能够表征信号发散程度的RQA参数--确定率和分层率组成轴承故障识别的特征向量，结合GG模糊聚类实现滚动轴承故障模式识别。对实际故障数据进行分析，结果表明，该方法不仅能够识别滚动轴承的不同程度损伤，而且能够实现不同部位的轴承故障诊断。研究结果为滚动轴承故障识别提供了一种高效、直观的新方法。

Fault Diagnosis of Rolling Bearings Based on RQA and GG Clustering

A fault diagnosis method of rolling bearings based on RQA and GG clustering was put forward. The parameters of the RQA which were able to characterize the degree of divergence of the signal-determinism and laminarity were used to consist the fault feature vector. Combined with the GG fuzzy clustering, it could achieve the fault pattern recognition of rolling bearings. The analyses of the actual fault data show that the method is able to identify different degrees of damage of rolling bearing faults and to complete different parts of the bearing fault diagnosis. It provides an efficient and intuitionistic new way for the identification of rolling bearing faults.