Abstract
This paper proposed a digital twin modeling method based on digital twin technology to improve the operational stability of rolling bearings and the accuracy of fault diagnosis methods. A comprehensive digital twin model for the entire lifecycle of rolling bearings was constructed using Modelica language. This model included a multi-state rolling bearing digital twin and integrated twin models for both the bearing drive and load ends. The model employed hybrid noise component to simulate the bearing's actual operating state and degradation process with high fidelity. Based on experimental lifecycle data from the laboratory, the rolling bearing full-life digital twin integrated model parameters were updated. Through the degradation components of the digital twin, the twin data of the rolling bearing was generated. By combining the twin data with actual measurement data, this approach addresses the limitations of traditional methods in the absence of data for bearings, providing reliable technical support for intelligent maintenance and fault diagnosis methods for rolling bearings.