Dynamic uncertainty analysis of rod-fastening rotor systems with rubbing effects using a Chebyshev interval method

Abstract

The study investigates the uncertainty quantification and parameter influence of a rod-fastening combined rotor system considering rub-impact effects. A continuous-discrete coupled dynamic model incorporating rub-impact boundaries is formulated based on the Lagrange equation, and the strongly nonlinear dynamic response is solved using the Newmark-β numerical method. An interval propagation analytical framework based on Chebyshev polynomial expansion is proposed to establish an efficient parameter uncertainty quantification system. The accuracy of the model and interval algorithm is validated through experiments and Monte Carlo simulations. The results demonstrate that the Chebyshev interval algorithm effectively characterizes parameter uncertainty, significantly reduces the computational complexity of nonlinear analysis, and enables rapid estimation of the rotor system's interval response. Interval analysis reveals the mapping relationship between rub-impact parameters and nonlinear vibration characteristics. A parameter classification criterion is established based on the difference in influence pathways: Direct Response-Influencing Parameters (DRIP) and Natural Frequency-Influencing Parameters (NFIP). Numerical experiments indicate that NFIP parameters influence vibration characteristics indirectly by altering the system's natural frequency, whereas DRIP parameters directly affect the dynamic response. The parameter sensitivity index of rub-impact clearance is considerably greater than that of contact stiffness and friction coefficient, and rub-impact fault-related uncertainties exert a notable influence only within the rub-impact domain. The proposed quantification framework provides a quantitative engineering criterion for predicting rub-impact fault thresholds and tolerance band design in aero-engine rotor systems.