Research on Extraction Method of Fatigue State Magneto Acoustic Emission Characteristic Parameters Based on CEEMD

Abstract

In view of the strong background interference of (MAE) signal and the unique advantage of complementary ensemble empirical mode decomposition (CEEMD), CEEMD is introduced into magneto acoustic emission signal processing. A feature extraction method of MAE signal based on CEEMD algorithm is proposed. In the proposed method, the signal is decomposed by the method of CEEMD, the correlation coefficients of the decomposed eigenmode function (IMF) components and the original signal are analyzed, and the IMF component with a large correlation coefficient with the original signal is retained. Then the components with large correlation coefficients are reconstructed, and the feature parameters are extracted, so that the interference can be effectively suppressed. The CEEMD algorithm can reduce the energy difference between the reconstructed signal and the original signal while maintaining the signal characteristics of the original signal. Finally, when the characteristic parameter curve obtained from the reconstructed signal reaches the yield strength of the material, the trend of the whole graph becomes smooth and stable which accords with the actual situation and verifies the effectiveness of the method.