Stochastic modeling and uncertainty analysis of rod-fastening combined rotor system based on experimental data

The mechanical features of the bolted connection interface significantly influence the dynamic behavior of the rod-fastening combined rotor (RFCR). Key factors such as preload, surface roughness, and polish govern the mechanical characteristics of this connection interface. However, accurately estimating the connection interface parameters proves challenging due to nonlinear behaviors such as loosening and slippage of the rods, which arise from manufacturing and assembly errors, and vibrations during rotor operation. To evaluate the dynamic characteristics of the RFCR effectively, it is crucial to create a stochastic model that accounts for uncertainties and then identify the system parameters accordingly. Based on experimental data, this study thoroughly investigates the inherent characteristics of the RFCR. Initially, the impact of preload magnitude and its detuning on the vibration characteristics of the rotor system is experimentally analyzed. Subsequently, the rotor structure is modeled using a stochastic general node approach, and the uncertainty of the model parameters is addressed through Bayesian identification. The suggested model's efficacy and dependability are evaluated by comparing experimental findings. This research demonstrates that the proposed methodology not only enhances the accuracy of rotating machinery system analysis but also contributes to the advancement of dynamic theory and its application in systems with multiple uncertain parameters.