VMD复合熵值法在齿轮故障诊断中的应用

Application of VMD Compound Entropy Method in Gear Fault Diagnosis

针对齿轮实际工况复杂、故障特征难以提取的问题,提出一种基于变分模态分解(Variational Mode Decomposition,VMD)复合熵值法的故障诊断方法。首先,采用VMD方法对不同工况下齿轮振动信号进行分解,并对分解过程中关键参数的选择进行了研究;其次,根据频域互相关系数准则筛选出可有效表征齿轮状态特征的固有模态函数(Intrinsic Mode Function,IMF)进行信号重构,为反映齿轮信号在不同尺度上的时-频细节复杂度特征,计算重构信号的样本熵、奇异值熵、功率谱熵和能量熵,形成高维状态特征向量;最后,将高维状态特征向量作为最小二乘支持向量机(Least Square Support Vector Machine,LS-SVM)的输入,对齿轮的工作状态和故障类型进行识别分类。通过实测齿轮信号的分析,结果表明,该方法能够有效实现齿轮故障的诊断。

Aiming at the problem that working condition is complex and it is difficult to extract the fault feature of the gear,a method of gear fault diagnosis based on variational mode decomposition(VMD)compound entropy was proposed.Firstly,the gear vibration signal was decomposed by VMD and the selection of key parameters in the decomposition process was discussed.Then,the frequency domain cross correlation coefficient was used to choose the corresponding he intrinsic mode functions(IMFs)to be reconstructed,in order to reflect the time-frequency detail complexity characteristics of gear signals on different scales,the sample entropy,singular value entropy,power spectral entropy and energy entropy of the reconstruction signal were calculated and used to compose the state feature vector.Finally,the high-dimensional feature vector was input last squares support vector machine(LS-SVM)for gear fault diagnosis.Through analysis of the gear test signals,the experimental results show that the proposed method can be used to diagnose gear faults effectively.