基于振动时频图像全局和局部特征融合的柴油机故障诊断

Diesel engine fault diagnosis based on the global and local features fusion of time-frequency image

提出了一种基于振动时频图像全局和局部特征融合的柴油机故障诊断方法。采用平滑伪维格纳分布(SPWVD)方法生成柴油机振动时频图像,分别用核主元分析(KPCA)和局部非负矩阵分解(LNMF)方法提取时频图像的全局和局部特征进行融合,并用独立分量(ICA)分析方法对融合后的特征进行降维,对降维后的融合特征进行分类完成对柴油机的故障诊断。试验结果表明,基于振动时频图像全局和局部特征融合的柴油机故障诊断方法,能够准确诊断柴油机的气门故障。

The global and local features fusion of a time-frequency image was introduced into the diesel engine fault diagnosis. Time-frequency images of a diesel engine were generated by the method of smoothed pseudo wigner-ville distribution（ SPWVD）. Then, the kernel principal component analysis（ KPCA） and local nonnegative matrix factorization（ LNMF） method were used to extract its global and local features,and the independent component analysis（ ICA） method was used for the dimension reduction of the characteristics after fusion. Finally,the fused features were classified to complete the diesel engine fault diagnosis.