Research on CEEMD-AGA Denoising Method and Its Application in Feed Mixer

Abstract

Rotating shaft is the key part of rotating machinery, which directly affects the performance of the whole machine. Field test is an easy and quick way to obtain the load data in engineering practice. However, because of various reasons, the load data are often mixed with many noise components. Based on the autocorrelation function, the CEEMD (complementary ensemble empirical mode decomposition) denoising method is proposed in this paper. The AGA (adaptive genetic algorithm) is adopted to solve parameter optimization problems in CEEMD. A new similarity function is proposed as the fitness function. Lastly, the proposed denoising method is applied to a feed mixer’s load which is obtained by field test. The result shows that the CEEMD-AGA method has good robustness, noise components of small stress amplitude and large stress mean are removed, and there is a high correlation between the original data and the reconstructed data, which demonstrate that the CEEMD-AGA method can reduce the influence of noise components effectively.