论文部分内容阅读
将SVM的分类算法应用于齿轮小样本故障诊断中。选取识别能力好的时域无量纲指标和频域中的9段频谱融合作为支持向量机的特征矢量,对齿轮的三种典型故障进行分类,结果表明:SVM在解决小样本情况下的机械故障诊断的分类问题中具有良好的应用前景。
The classification algorithm of SVM is applied to the fault diagnosis of gear sample. Three time-domain non-dimensional indicators with good recognition ability and nine segments of frequency spectrum in the frequency domain are selected as the feature vectors of support vector machines to classify the three typical faults of the gear. The results show that the mechanical failure of SVM in solving the small sample Diagnosis of the classification problem has a good prospect.