导　　师： 柳亦兵

学科专业： H1101

授予学位： 硕士

作　　者： ;

机构地区： 华北电力大学

摘　　要： 希尔伯特-黄变换(hilbert-huang transform，简称hht)是美籍华人科学家huang在1998年提出的一种新的信号时频分析方法。本文研究该方法在非平稳信号分析中的实际应用问题，主要完成以下几项工作：1)通过多种典型仿真信号的hht时频分析，并与短时傅立叶变换、小波变换等时频分析方法进行比较，探讨了hht方法的适用性；2)采用周期或对称延拓的方法对抑制hht端点“飞翼”问题进行了分析；3)针对hht的虚假分量问题，提出能量比值进行虚假分量识别的改进方法：4)应用改进的hht对几种类型的工程实测非平稳信号进行时频分析(如机械设备振动信号、风速信号等)，提取信号的特征信息，取得了令人满意的结果。 Hilbert-Huang Transform proposed by scientist Huang in 1998 is a new time-frequency signal analysis method. This paper studies the practical applications of this method in analyzing non-stationary signal and completes the following tasks. 1/) through analyzing a variety of typical simulating signal by HHT method and compared with the other typical time-frequency analysis methods such as the short-time Fourier transform, wavelet transform we discuss the applicability of the HHT method; 2/) Aiming to restrain the endpoint 'flying swing', a cycle or symmetric extension methods is adopted here; 3/) Propose energy comparison improved method to identify the illusive component due to HHT; 4/)Apply the improved HHT method to analyze several types of engineering measured non-stationary signals /(such as mechanical equipment vibration signals, wind speed signals etc/) , extracts the characteristic information of the signal, and satisfactory results are obtained.

关 键 词： 非平稳信号 信号分析 希尔伯特 黄变换 时频分析

分 类 号： [TN911.6]

领　　域： [电子电信] [电子电信]