摘要: To meet the requirements of improving work efficiency, the rotating machinery represented by gas turbines reduces the gap between rotor and stator, making rub-impact occur frequently. However, a few researchers combined with the current gas turbine condition monitoring concluded that it could diagnose the rub-impact fault. In this study, the vibration features that can help diagnose the compound fault of rotor unbalance and rub-impact are researched. Taking the Jeffcott rotor system as the research object, the mathematical model is constructed considering the coupling of radial and torsion, and the Runge-Kutta method is used to solve the differential equation. Through the analysis of the model results, it is concluded that the compound fault can be diagnosed by the vibration features-reduction of the fundamental frequency amplitude. Finally, the correctness of the model is verified by experiments. At the same time, the vibration features of the rotor and stator are compared to show the consistency of vibration features, which shows that the proposed features can diagnose this compound fault.