In this paper， a wavelet- variational mode decomposition （VMD） joint de-noising method is proposed for the measurement of rotational speed and angular acceleration signal de-noising， which is used to process angular velocity and angular acceleration signals. Firstly， vehicle test and signal acquisition are carried out to obtain the first mass ， the second mass angular velocity and acceleration signals of the dual-mass flywheel. Then the parameters of wavelet and VMD are optimized respectively. The peak-to-peak value of signal-to-noise ratio is used to evaluate the sensitivity of wavelet parameters to the de-noising performance. On this basis， wavelet parameters are obtained. Finally， several modal components （IMF） are obtained by using VMD decomposition. The main IMF is selected by cross correlation coefficient and the signal is reconstructed to obtain the final de-noising signal， and the de-noising results are analyzed. The results show that the proposed method can effectively remove the noise components in the angular velocity and angular acceleration signals. Correlation coefficients between the de-noising signal and the original signal are 0.998 5， 0.997 5， 0.835 4 and 0.683 6， respectively. Besides， the angular velocity and angular acceleration under the dual mass flywheel acceleration condition are calculated. The attenuation rate is above 80% which meets the design performance requirements.