基于HHT信号处理方法的水轮发电机组低频压力脉动信号的特征提取

Feature Extraction of Hydraulic Units′ Low Frequency Pressure Fluctuation Signal Based on Hilbert-huang Transform

基于经验模态分解(Empirical mode decomposition,EMD)的希尔伯特黄变换(Hilbert-Huang transform,HHT)方法能够较好地分析非平稳和非线性的信号.提出了基于Pascal程序设计语言的HHT算法的程序实现方法.通过对某水轮发电机组水导轴承的振动信号进行分析发现,EMD方法等够有效地分离出低频压力脉动信号中的各个单一模态的振动分量,分解出来的各振动分量具有物理意义.Hilbert谱相比于传统的时频谱具有更好的时频分辨率,且具有自适应的信号处理能力.

Hilbert-huang transform（HHT） based on empirical mode decomposition is considered more effective in dealing with non-stationary and non-linear signal.In this paper,the computer program of HHT algorithm is designed by Pascal programming language.After vibration signal analysis on turbine guide bearing of a hydraulic unit,we found that EMD method can extract the single mode components from low frequency pressure fluctuation signal effectively,and the single mode components have more physical meaning.Hilbert spectrum has higher resolution in time-frequency distribution and the ability of self adaptive signal processing compared to traditional time-frequency distribution.