参数优化变分模态分解和多域特征融合的行星变速箱故障诊断

Planetary Gearbox Fault Diagnosis Based on Variational Mode Decomposition With Parameter Optimization and Multi-domain Feature Fusion

为提高行星变速箱的诊断精度,提出了基于参数优化变分模态分解和多域特征融合的行星变速箱故障诊断方法。首先,为提高变分模态分解(VMD)对信号的分解效果,以天牛须搜索(BAS)为优化方法进行VMD参数的优化。其次,以参数优化VMD完成信号的分解,并筛选出分解所得本征模态函数(IMF)中包络熵最小的IMF。再次,计算包络熵最小IMF的时城、频城和尺度城特征组成高维多城故障特征集,而后通过监督局部切空间排列(SLTSA)完成特征集的降维。最后,通过支持向量机对故障特征进行识别,得到诊断结果。行星变速箱故障诊断实例验证了方法的有效性。

In order to improve the fault diagnosis accuracy of planetary gearbox,a fault diagnosis method based on variational mode decomposition with parameter optimization and multi-domain feature fusion is proposed.Firstly,in.order to improve decomposition effect of variational mode decomposition(VMD),beetle antennae search(BAS)is used to optimize V MD decomposition parameters,and the intrinsic mode function(IMF)with minimum envelope entropy is selected from the decomposed IMFs.Thirdly,the time-domain,frequency-domain and scale-domain features of IMF with minimum envelope entropy are calculated to form a high-dimensional multi-domain hybrid fault feature set,and then dimension reduction of the feature set is accomplished by supervised local tangent space alignment(SLTSA).Finally,the fault features are identified by support vector machine and the diagnosis results are obtained.An example of fault diagnosis of planetary gearbox verifies the efectiveness of the method.