噪声与振动控制

›› 2012, Vol. 32 ›› Issue (1): 113-118.DOI: 10.3969/j.issn.1006-1355-2012.01.027

• 6.信号处理与故障诊断 • 上一篇    下一篇

HHT统计特征在航空发动机振动分析中的应用

吴娅辉,李新良,洪宝林,张大治   

  1. （ 中国航空工业集团公司 北京长城计量测试技术研究所 计量与校准技术重点实验室， 北京 100095 ）
  • 收稿日期:2011-05-03 修回日期:2011-06-16 出版日期:2012-02-18 发布日期:2012-02-18
  • 通讯作者: 吴娅辉

Statistical Feature of Hilbert - Huang Transform and Its Application in Aeroengine Vibration Analysis

WU Ya-hui,LI Xin-liang,HONG Bao-ling,ZHANG Da-zhi   

  1. （ AVIC Changcheng Institute of Metrology and Measurement, Key Laboratory of Science and Technology on Metrology & Calibration, Beijing 100095, China ）
  • Received:2011-05-03 Revised:2011-06-16 Online:2012-02-18 Published:2012-02-18
  • Contact: WU Ya-hui

摘要： 航空发动机振动信号包含了多种振源的振动信号和大量的噪声分量。通过对其进行Hilbert-Huang变换, 将复杂信号分解为代表不同物理意义的单分量固有模态函数（Intrinsic Mode Functions, IMF），然后对每一个IMF子带信号提取基于能量加权的广义粗糙度特征实现对振动信号的描述。最后将上述特征送入SVM分类器进行训练，根据分类器的输出结果确定航空发动机的工作状态和故障类型。通过对实测航空发动机试车时得到的振动信号的实验分析结果表明，该算法可以有效地识别发动机的振动故障。

关键词: 振动与波, Hilbert - Huang变换, 广义粗糙度特征, SVM, 振动分析, 航空发动机

Abstract: The vibration signals of the aeroengine include many multi-frequency components and noise components. In this paper, the Hilbert-Huang transform is used to decompose the vibration signals into a number of intrinsic mode functions (IMF) which represent different physical meanings of the data. Then, the generalized roughness vector of the IMF weighted by its energy is extracted from the IMF signals. Finally, as the fault characteristic vectors, the derived parameters are input into the support vector machine (SVM) classifier and the working conditions and fault patterns are identified by the output of the classifier. Results from the analysis of aeroengine vibration signals show that this fault diagnosis method can classify the working conditions and the fault patterns effectively.

Key words: vibration and wave, Hilbert - Huang transform, generalized roughness vector, support vector machine, vibration analysis, aeroengine

中图分类号: