基于AMD-HHT的非平稳信号紧密间隔频率检测  

Closely spaced frequency component detection method of non-stationary signal based on AMD and HHT

作　　者：时培明[1] 苏翠娇 韩东颖[2] 

机构地区：[1]燕山大学电气工程学院,秦皇岛066004 [2]燕山大学车辆与能源学院,秦皇岛066004

出　　处：《仪器仪表学报》2014年第12期2817-2825,共9页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(51475407;51104129);河北省自然科学基金(E2012203194);河北省高等学校科学技术研究项目(YQ2013020)资助

摘　　要：针对希尔伯特-黄变换方法不能识别具有相近频率的非平稳信号(两个频率之间的比值小于1.5)的问题,提出一种解析模态分解法(AMD)与希尔伯特-黄变换(HHT)结合的非平稳信号紧密间隔频率检测新方法。由于AMD方法需要确定信号里的各个频率成分,因此,首先通过EMD对信号进行分解,再利用希尔伯特变换得到信号的频谱并经过处理得到各个频率值。利用AMD方法提取各个不同频率成分的信号,并用该方法对信号进行分解,判断该频率成分是否由于含有多个相近频率成分而没有被分解,如果含有相近频率成分,则用AMD方法将其分解为单频率分量信号。仿真结果表明,该方法解决了希尔伯特-黄变换不能有效分离两个相近频率信号的问题,保证了信号有效数据的正确分解,提高了信号的分解精度。通过对滚动轴承故障信号的分析证明该方法有良好的效果。Aiming at the problem that Hilbert-Huang Transformation method is unable to distinguish the signals with closely spaced frequency components （the two frequency ratio is less than 1.5）, a new frequency detection method of non-stationary signals with closely spaced frequency components based on AMD and HHT is proposed. Because the AMD method needs to know the frequency of each component, firstly EMD is used to decompose the signal; then, Hilbert transform is used to obtain the frequency spectrum of the signal and the frequency value. AMD is used to extract the signals of different frequency components and decompose the signals, and then judge whether the frequency component contains more than one frequency components. If there are two or more frequency components, AMD method is used to separate the signal to single frequency components. The simulation results show that the proposed method solves the problem that HHT cannot effectively separate a signal with closely spaced frequency components. This method ensures the correct decomposition of the signal effective data and improves the accuracy of signal decomposition. The fault signal analysis of rolling bearing proves that the method has good effect.

关 键 词：解析模态分解法 希尔伯特-黄变换 相近频率分量 信号提取 

分 类 号：TH165.3[机械工程—机械制造及自动化] TP277[自动化与计算机技术—检测技术与自动化装置]