中文摘要：

针对涡旋压缩机振动信号的非平稳性和难以获得大量实际故障样本的情况，提出一种奇异值谱和支持向量机（Support Vector Machine，SVM）的故障诊断新方法。该方法首先通过对涡旋压缩机信号进行小波包分解，构建时频系数矩阵；然后对该矩阵进行奇异值分解（Singular Value Decomposition，SVD），获得信号的奇异值谱，并计算奇异谱的分布参数作为故障识别的特征向量；最后将特征向量作为SVM的输入，实现涡旋压缩机故障类型的辨识。试验结果表明：即使在小样本情况下，该方法仍能有效识别涡旋压缩机故障类型。

英文摘要：

In view of the non-stationary features of vibration signals of scroll compressor and the difficulty to obtain a large number of fault samples in practice, a new method of fault diagnosis based on singular value spectrum and support vector machine is proposed. Firstly, through applying wavelet packet decomposition to signals a time-frequency coefficient matrix is constructed, then this time- frequency matrix is decomposed by SVD to attain singular spectrum of signal, and the distribution parameters of the singular spectrum are calculated as the feature vector of fault identification. Finally, the feature vector is used as the input of SVM to identify the fault types of scroll compressor. The experimental results show that even in small samples, the method can effectively identify the fault types of scroll compressor.