Identification Of Nonlinear Rotor System Unbalance

Identification of unbalance in linear rotor system has been well developed. Methods of balancing can be categorized into two groups: the influence coefficient method and modal balancing. However, these methods are not effective when the rotor system become nonlinear. Although the nonlinearity can be reduced by balancing the rotor with several test runs, sometimes, the downtime is very expensive.It is necessary to find out a technique to identify unbalance of nonlinear rotor system. Researchers have done some work with identifying parameters and residual unbalance simultaneously. However, these identifications are very complicated. It would take too much time to identify the unbalance by using these identifications. In present study, a new harmonic balancing method, without parameter estimation, is proposed to identify unbalance of nonlinear rotor system.In this paper, first of all, the motion equation of rotor system is introduced. The DOF (degrees of freedom) of rotor system has been reduced by mode method. So, the solving of motion equation becomes easier.The nonlinearity of the rotor system is mostly produced by the journal bearing. Previous work paid too much attention to the bearing parameters estimation, which is not necessary for unbalance identification. In the new algorithm, those bearing nonlinear parameters are all coupled into one term, the oil film force. This force is nonlinear but it is closed to periodic. It can be expanded by Fourier transform. Unbalances only produce work-frequency forces. According to this, work-frequency amplitudes are extracted from the response of the rotor and they are used to identify unbalance with the help of the prior knowledge of the system. In this algorithm, even though the work-frequency oil film forces are included in the result of identification, the location difference of bearing and balancing plane make it can be subtracted from the result. The process of calculation and the condition of the equation solving have been presented.In order to test this new algorithm, numerical simulation has been taken. Before the numerical simulation, the nonlinearity of rotor system supported by journal bearing has been studied. Simulation shows that he new algorithm is effective, stable, and robust.Finally, the method has been applied in experiment. Rotor system response after balancing has been reduced to a low level. The proposed method is testified to be applicable.