中文摘要：

在旋转机械故障智能诊断中,收集大量的样本比较容易,而要对所有的样本进行类别标记却较为困难。针对这一问题,提出了一种基于拉普拉斯支持向量机的旋转机械故障智能诊断方法。滚动轴承故障诊断实例表明,有标记样本的数量较少时,与仅使用有标记样本进行学习的支持向量机相比,基于拉普拉斯支持向量机的诊断方法利用大量的无标记样本进行辅助学习,可以显著提高故障诊断的正确率。

英文摘要：

In the intelligent fault diagnosis of rotating machinery, collecting a large number of data was relatively easy, but giving all collected data a label was often difficult. Aiming at this situation, an intelligent fault diagnosis approach for rotating machinery was proposed based on Laplacian support vector machines（LapSVM）. The diagnosis example of rolling bearings shows that when the number of labeled data is limited, compared with the SVM that uses only labeled data for learning, the fault diagnosis approach based on LapSVM can improve the accuracy of fault diagnosis significantly by using a large amount of unlabeled data together with labeled data for learning.