基于EMD和HHT的内燃机噪声信号时频特性研究  

Study on the Time-Frequency Characteristics of Engine Noise Signal Based on EMD and HHT

作　　者：徐红梅[1] 郝志勇[1] 杨陈[1] 景国玺[1] 

机构地区：[1]浙江大学机械与能源工程学院,杭州310027

出　　处：《内燃机工程》2008年第6期60-64,69,共6页Chinese Internal Combustion Engine Engineering

基　　金：国家自然科学基金资助项目(50575203)

摘　　要：简述了经验模态分解(EMD)和Hilbert-Huang变换(HHT)的基本原理与方法。设计仿真试验,验证经验模态分解方法和Hilbert-Huang变换技术对复杂信号的分解和时频分析能力。以某6缸发动机为研究对象,采用小波包分解方法对噪声信号进行滤波预处理,并对滤波后信号进行经验模态分解,得到多个具有不同频率的本征模态函数分量,分别对各分量进行希尔伯特黄变换,分析其幅值和频率随时间变化的特性,结合内燃机结构声辐射机理,分析各分量产生原因,识别噪声源。研究结果表明:经验模态分解方法和Hilbert-Huang变换技术适于分析内燃机噪声信号。The basic principles and methods of empirical mode decomposition （EMD） and Hilbert- Huang transform （HHT） were briefly discussed. The simulation experiment was conducted to verify the decomposition and time-frequency analysis ability of EMD and HHT for complex signals. Taking a six cylinder engine as example, the wavelet package decomposition was first used as a filter to .preprocess the noise signal, and then the filtered signal was decomposed into a collection of intrinsic mode functions （IMFs） with different frequency by the method of EMD. HHT was employed to analyze the amplitude and frequency of each IMF varying with time. In terms of the structural acoustic radiation mechanism of engine, the origin of each IMF was deduced, and the noise sources can be identified. The research has shown that the method of EMD and HHT is suitable for the analysis of engine noise signal.

关 键 词：内燃机 噪声 经验模态分解 HILBERT-HUANG变换 时频分析 

分 类 号：TK421.6[动力工程及工程热物理—动力机械及工程]