中文摘要：

针对变速器故障齿轮振动信号特征和在现实条件下难以获得大量故障样本的实际情况,提出一种复调制细化包络谱和支持向量机相结合的故障诊断方法.通过变速器齿轮故障模拟试验模拟齿轮磨损故障,采用Hilbert变换对齿轮振动信号进行解调得到振动信号的包络,对包络信号进行复调制细化谱分析,提取齿轮轴转频基波及其谐波幅值.分析发现基波及其谐波幅值随着齿轮磨损程度的增加明显增大,故将其作为支持向量机的输入特征向量,以判断齿轮故障.试验结果表明：该方法在小样本的情况下能有效地诊断齿轮故障.

英文摘要：

According to the characteristics of the vibration signals of faulty transmission gears and the practical difficulties in obtaining large enough fault samples,a fault diagnosis method of transmission gear based on complex modulation zooming envelope spectrum and support vector machine （SVM） is proposed.Firstly,the faults of worn gears are simulated by a simulation test on transmission gears and the vibration signals of gears are demodulated by using Hilbert transform to obtain the envelope of vibration signals,and the amplitudes of fundamental wave and its harmonics of gear shaft are extracted by ZFFT analysis on envelope signals.It turns out that the amplitudes of fundamental wave and its harmonics obviously rise with the increase in the degree of gear wear,which is then taken as the input eigenvectors of SVM to judge the gear faults.The results show that the proposed method can effectively diagnose the gear faults with small samples.