Ultrasonic rolling surface roughness study of shaft parts based on three-dimensional morphology reconstruction

The ultrasonic surface rolling process (USRP) can effectively reduce surface roughness, reduce the stress concentration and extend the service life of a workpiece. Surface roughness prediction by finite element simulation has been a hot research topic of late. Most of the previous studies are based on plane models, but the influence of initial surface roughness on the prediction results of shaft parts has not been considered. In this paper, wavelet transform is used to reconstruct the measured cylindrical sample surface data, and the shaft model with initial surface roughness is established. Then the finite element simulation model of USRP is established to predict the surface roughness. The results show that compared with the experimental results, the maximum error of the proposed method is 9 %, which is better than that of the finite element simulation model without considering the initial surface roughness, which verifies the effectiveness of the proposed method.