基于ICEEMDAN和EMDE的滚动轴承故障诊断

Fault diagnosis of rolling bearing based on ICEEMDAN and EMDE

针对滚动轴承振动信号的非线性和非平稳特点,开发基于改进自适应噪声完备集成经验模态分解(ICEEMDAN)和增强多尺度分布熵的故障识别模型。首先,利用ICEEMDAN分解滚动轴承振动信号,得到1组内禀模态函数(IMF),根据相关系数筛选出其中反映故障状态关键特征的IMF分量;然后,利用增强多尺度分布熵对各敏感IMF分量进行复杂性评估,得到滚动轴承的故障特征向量;最后,为识别滚动轴承的不同故障类型,使用支持向量机作为故障识别分类器。实验结果表明:所提故障诊断方法具有可观的故障识别精度和稳定性,相比于其他故障诊断方法,该方法具有明显的优势。

According to the nonlinear and non-stationary characteristics of rolling bearing vibration signals,a fault identification model based on improved complete ensemble empirical mode decomposition with adaptive noise(ICEEMDAN),enhanced multi-scale distributed entropy and support vector machine is developed.Firstly,ICEEMDAN is used to decompose the vibration signal of rolling bearing to obtain a set of intrinsic mode functions(IMF),and the IMF components reflecting the key characteristics of fault state are selected according to the correlation coefficient.Then,the complexity of each sensitive IMF component is evaluated by using enhanced multi-scale distribution entropy,and the fault feature vector of rolling bearing is obtained.Finally,in order to identify different fault types of rolling bearings,support vector machine is used as a fault recognition classifier.The experimental results of rolling bearing data sets show that the proposed fault diagnosis method has considerable fault identification accuracy and stability,and compared with other fault diagnosis methods,this method has obvious advantages.