基于基本尺度熵与GG模糊聚类的轴承性能退化状态识别

针对轴承性能退化状态的识别问题,提出一种基于基本尺度熵与GG聚类的退化状态识别方法。首先分析轴承性能退化过程中的基本尺度熵演化规律,并分析该参数的单调性与敏感性。考虑到轴承退化状态在时间尺度的连续性,构建基本尺度熵、有效值以及退化时间的三维退化特征向量,并采用GG模糊聚类方法对轴承性能退化状态的不同阶段进行划分,实现对性能退化状态的识别。采用来自IEEE PHM 2012的轴承全寿命试验数据进行实例分析,并与FCM、GK算法进行对比,结果表明本文所提出的方法聚类效果更优,同一退化状态内的时间聚集度更高,能够为轴承性能退化状态的识别提供一种有效的途径。

Rolling bearing performance degradation state recognition based on basic scale entropy and GG fuzzy clustering

A method based on basic scale entropy and GG clustering was proposed to solve the problem of bearing performance degradation state recognition.The evolution law of the basic scale entropy in bearing performance degradation process was analyzed firstly,and its monotonicity and sensitivity were emphasized.Considering the continuity of bearing degradation state on time scale,a 3D degradation feature vector was constructed with basic scale entropy,its root mean square and degradation time,and GG fuzzy clustering method was used to divide different stages of bearing performance degradation state to realize bearing performance degradation state recognition.Bearing full lifetime test data of IEEE PHM 2012 was adopted to do example analysis,and the results were compared with those using FCM and GK algorithms.The results showed that the proposed method’s clustering effect is better and its time aggregation degree is higher in the same degradation state;the method can provide an effective way for bearing performance degradation state recognition.