Rolling bearing digital twin model oriented toward rotational accuracy prediction

Abstract To address the low qualification rate of high-precision rolling bearing products and the need for repeated assembly and selective matching, this paper proposes a digital twin modeling method for rolling bearings oriented toward rotational accuracy prediction. Morphological feature data of the inner and outer raceways are obtained through multi-section measurements using a roundness and waviness measuring instrument. After polar-to-Cartesian transformation and adaptive parameterization, cubic non-uniform rational B-splines (NURBS) curves and bicubic NURBS surfaces are used to reconstruct the raceway surface morphology features. A solid model is constructed through isogeometric transformation and geometric post-processing, and a digital twin simulation model for rotational accuracy is established based on ANSYS–ADAMS. Experimental validation is carried out using a B7008AC angular contact ball bearing as the research object. The experimental results show that, compared with the traditional smooth model, the proposed model can effectively improve the prediction accuracy of bearing rotational accuracy, providing a predictive model basis for intelligent grouping of bearing rings based on error complementarity.