摘要
针对齿轮多重故障特征提取困难的问题,提出了利用形态小波变换与经验模态分解(Empirical Mode Decomposition,EMD)相结合的齿轮故障特征提取方法。在利用形态小波变换对非平稳信号进行降噪的基础上,运用EMD方法对降噪后信号进行分解,得到贡献率大的固有模态分量。工程试验结果表明:该方法能准确地提取齿轮多重故障特征,在不同工况下均有较好的效果。该方法解决了噪声干扰对早期故障信号引入导致诊断效果不明显的问题,提高了EMD诊断的识别率和效率。
Aiming at feature extraction difficulties of gear multiple failures,feature extraction method of gear failures based on morphological wavelet transform and Empirical Mode Decomposition(EMD) is proposed.On the basis of non-stationary signal noise reduction by using morphological wavelet transform,EMD is used to decompose the reduced noise signal,and intrinsic mode components is got,whose contribution is large.Engineering test result shows that the method can accurately extract gear feature of multiple failures under different conditions.This method solves the problem of the noise interference introduced to the early fault signal which causes obscure diagnosis and improves the recognition rate and efficiency of EMD diagnosis.
出处
《装甲兵工程学院学报》
2014年第2期50-53,共4页
Journal of Academy of Armored Force Engineering
关键词
齿轮传动
形态小波
经验模态分解
故障诊断
特征提取
gear transmission
morphological wavelet
Empirical Mode Decomposition(EMD)
fault diagnosis
feature extraction
