摘要
希尔伯特黄变换(Hilbert-HuangTransformation,HHT),是先把一列时间序列数据通过经验模态分解(EmpiricalModeDecomposition,EMD),然后经过希尔伯特变换获得频谱的信号处理新方法。详细地介绍了HHT方法的理论和算法。首先,通过仿真信号把该方法与小波变换(WaveletTransformation,WT)方法进行了比较研究,验证了方法的优越性;然后,把该方法用于旋转机械油膜涡动故障诊断中,研究结果表明:该方法相对传统的分析方法在较低转速区能更早发现油膜涡动故障,说明把基于HHT的时频分析方法用于旋转机械故障诊断是有效的。
The Hilbert-Huang Transformation (HHT) consists of the following steps: First, decomposition of time sequenced data by empirical modes (EMD) followed by a Hilbert transformation to obtain a new method of spectrum treatment. In this paper, the theory and algorithm of the HHT is explained in detail. A comparison is performed between this method and the Wavelet Transformation (WT) method with the help of simulated signals, which shows that this method is superior to the WT's. The method has been applied to diagnose oil film whirling fault signals. Results show that this method is able to detect oil film whirling earlier and at a lower rotor speed than the traditional method. It can be concluded that the HHT-based method can effectively be applied to the fault diagnosis of rotating machinery. Figs 11 and refs 8.
出处
《动力工程》
CSCD
北大核心
2004年第6期845-851,共7页
Power Engineering
基金
国家自然科学基金(50205025)
浙江省自然科学基金项目(5001004)
