论文部分内容阅读
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 signal processing and other fields due to its excellent mathematical properties. However, the cross term caused by its unsatisfactory additivity affects the accuracy of signal analysis. Aiming at this problem, a method of eliminating the Wigner-Ville distribution cross term by using Ensemble Empirical Mode Decomposition (EEMD) is proposed. In this method, the original signal is decomposed into several intrinsic mode functions (IMFs) by EEMD, and then the obtained IMFs are secondarily processed by using EEMD to obtain a new set of IMFs. Then, some new modal components WVD calculation, respectively, and finally the sum of the results obtained WVD distribution of the signal. This method effectively eliminates the Wigner-Ville distribution cross term while improving modal aliasing and preserves various excellent features of the Wigner-Ville distribution. The method is applied to the rubbing fault diagnosis of rotors and works well.