In order to eliminate the influence of noise on adaptive ensamble empirical mode decomposition（AEEMD） in sampling signal

an AEEMD harmonic detection algorithm based on improved wavelet threshold denoising is proposed. The method uses new formula to deal with the effective wavelet coefficients

and uses the method of soft and hard threshold combination to eliminate the noise

and then carries out AEEMD decomposition.The simulation results show that the proposed method can effectively eliminate the random noise and improve the decomposition efficiency by 2. 06 times. The correlation between the effective component and the original signal is improved by 19. 36 %. The improved method can improve the decomposition efficiency well

so as to verify that the proposed algorithm is feasible and effective.