摘要
针对滚动轴承故障信号非线性、非平稳特征导致的故障频率难以提取的问题,提出一种基于补充总体平均经验模态分解(Complementary EEMD,CEEMD)和奇异值差分谱结合的滚动轴承故障诊断方法。CEEMD分解向原信号成对地添加符号相反的白噪声,几乎消除残留白噪声的影响。首先,对故障信号利用CEEMD算法进行分解,得到若干IMF(Intrinsic Mode Function)分量,然后运用相关系数—峭度准则对IMF分量进行筛选并重构,再对重构信号进行奇异值分解,并求出奇异值差分谱,根据奇异值差分谱理论进行消噪和重构,最后对重构信号进行Hilbert包络谱分析,提取故障频率。实验结果表明,提出的方法,能精确地提取滚动轴承的故障频率。
Due to the non-stationary and non-linear characteristics of motor rolling bearings vibration signal which makes it hard to extract the fault frequency,a new method of fault diagnosis for the rolling bearings based on complementary ensemble empirical mode decomposition( CEEMD) method and difference spectrum of singular value is proposed. By adding the white noise in pairs into a target signal,the method of complementary EEMD( CEEMD)almost eliminates the influence of the white noise. Firstly,bearing fault signals are decomposed into a finite number of IMFs based on the way of CEEMD; Then,filtering the components according to the correlation coefficient-kurtosis criteria,and the selected IMF components are used to reconstruct the signal. A Hankel matrix is constructed by the reconstructed signal and the singular value difference spectrum can be obtained after singular value decomposition. Then,the singular value difference spectrum theory is used to reconstruct signal and eliminate noise.Finally,the reconstructed signal is demodulated by Hilbert transformation to extract the fault features. Results of experiment signals analysis show that the method proposed in this paper can identify gear fault patterns effectively.
出处
《电力科学与工程》
2016年第1期37-42,共6页
Electric Power Science and Engineering
关键词
CEEMD
奇异值差分谱
相关系数—峭度准则
滚动轴承
故障诊断
complementary ensemble empirical mode decomposition(CEEMD)
difference spectrum of singular value
correlation coefficient-kurtosis criteria
motor bearings
fault diagnosis
