A Digital Twin Model of Planetary Gearbox with Bearing Fault

Digital twin modeling technology strives to improve the digitization and intelligence of complex equipments and is widely used in intelligent maintenance of complex planetary gear transmission systems. Modeling and optimization are two main parts of digital twin modeling for planetary gearbox. Different from modeling methods based on big data and neural networks, the methods based on dynamic theory can clearly describe the mechanism of fault generation and changes in dynamic response. However, it is difficult to provide reasonable and accurate dynamic parameter settings while modeling. In this paper, a digital twin modeling method based on multibody dynamic parameter optimization by experiment for planetary gearbox is proposed. Firstly, a multibody model of planetary gearbox with bearing fault is established and the dynamic parameters that need to be updated are analyzed. In order to improve the accuracy of the twin model, reasonable preprocessing is applied on the initial simulated and experimental vibration signals. Finally, 51 dynamic parameters considering fault force, bearing, gear meshing and bolting of the multibody dynamic model are updated by experimental signal and a digital twin model of the planetary gearbox with bearing fault is obtained.