Analysis And Identification Of Oil Film Instability State In Rotor Bearing System

In recent years,catastrophic accidents caused by low-frequency vibrations in rotating machinery systems have made people aware that the analysis and identification of oil film instability has become crucial.Among the many factors affecting the instability of the oil film in the rotor bearing system,the unbalanced force and the nonlinear oil film force are particularly prominent.Although the research results on the nonlinear oil film force are fruitful,the influences of the unbalanced force on the accumulation and expansion process of the oil whirl have not been clearly studied.The influence of unbalanced force on the occurrence of oil film instability and identification of oil film instability faults are mainly studied in this paper.With regard to the influence of unbalanced forces on the process of oil film instability,a nonlinear dynamic model of a single-span double-rotor rotor bearing system is established,and the equations under different eccentricity ratios is performed numerical simulations by the fourth-order Runge-Kutta methods.The bifurcation diagram,Poincare map,waterfall plot,and axis trajectory image are obtained.It is observed that the eccentricity has a significant impact on the occurrence of oil film instability.Without the eccentricity,the system’s movement state does not appear chaotic.With the increase of the eccentricity,the movement state of the system becomes more complex,and the phenomenon of transition between multiple states of motion occurs,and the frequency of transition from single-cycle to other-state transitions increases.The eccentricity also has a certain influence on the stability of the system.When the eccentricity is large,the effect of unbalanced force is stronger than the effect of nonlinear oil film.So that the rotor system is not prone to oil whirl,and it will disappear within a certain speed range.In order to verify the model,the experiments of the oil film instability are carried out.Compared with the numerical analysis results,the nonlinear dynamic model established in this paper can truly reflect the impact of the unbalanced quantity on the oil film instability process.In order to realize the fault identification,the other four faults are also simulated in the experiments including the normal rotation test,the misalignment test,the rubbing test and the misalignment unbalanced experiment.The recognition effect of a single learning machine is influenced by the initial parameters,and the BP-FI model is introduced.The BP-FI fusion technology is applied to the fault identification of the rotor bearing system,which can identify oil film instability failure with different eccentric ratios and different types of rotor system failure.Moreover,the BP-FI model has a higher recognition rate and better stability than a single classifier.