摘要
Wigner-Ville分布(WVD)凭借优良的数学性质被广泛应用于信号处理等领域,但因其不满足可加性而引起的交叉项却影响信号分析的准确性。针对此问题提出利用二次聚合经验模态分解(Ensemble Empirical Mode Decompo-sition,EEMD)消除Wigner-Ville分布交叉项的方法。该方法首先用EEMD将原信号分解成为若干本征模函数(IntrinsicMode Function,IMF),再利用EEMD对获得的IMF进行二次处理,得到一组新的IMF,然后对新得到的若干模态成分分别进行WVD计算,最后将各项结果求和得到信号的WVD分布。该方法在改善模态混叠的同时有效地消除Wigner-Ville分布交叉项,并保留Wigner-Ville分布的各种优良特性。将该方法应用于转子碰摩故障诊断,效果很好。
Wigner-Ville distribution （WVD） is widely used in the filed of signal processing due to its multiple excellent mathematical characteristics. However, the cross-terms in WVD often bring us lot of problems, because the WVD is of no additive property in calculation. Aiming at this problem, a cross-terms erasing method based on dual ensemble empirical mode decomposition （EEMD） was put forward. In the method, the original signal was decomposed into several intrinsic mode functions （IMFs） by EEMD, which were then reprocessed to gain a group of new IMFs. After calculating the WVD of each IMF and adding them together, the WVD of original signal was achieved. The method presented effectively avoids the cross-terms of WVD with less mode mixing while keeps all the advantages of WVD. The method was used in the signal processing for rub fault diagnosis and has gained a perfect effect.
出处
《振动与冲击》
EI
CSCD
北大核心
2012年第13期129-133,153,共6页
Journal of Vibration and Shock
