An Analysis of Signals De-noising by Using CEEMDAN

Abstract

Dynamical properties of mechanical systems can be obtained with the vibration signals from the systems. However, noises makes it difficult to accurately acquire the features of systems. Therefore, de-noising operation is significant step for analysis of vibration signal in the practical engineering. In order to resolve this problem, the Complementary Ensemble Empirical Mode Decomposition (CEEMDAN) method is introduced to try to eliminate noises from the analyzed signal. At first, we illustrate the theory of the method. On this base, the signal, which contains several harmonic components with white noise, is constructed, and processed by CEEMDAN. As the result shown, the method can effectively remove noise; moreover, the harmonic components can be accurately separated. And these indicate that the CEEMDAN is an effective method for the de-noising.