Research On Unbalance Identification Based On Finite Element Model Using Robust Estimate

Rotor unbalancing is one of the main causes of vibration and noise in aero-engine, which not onlyaffects work performance and fatigue life directly but also can cause other kinds of fault. So it is ofengineering value to research flexible rotor dynamic balancing.Modal balance and influence coefficient methods are the two classic rotor balancing approaches forflexible rotor. The former requires to know the modal characteristics of the rotor and the latter, that iswidely used in dynamic balance due to its easy to implement, needs to plus trial mass repeatedly tomeasure the influence coefficient. At present, Identification unbalance and realization dynamic balanceusing the vibration data measured in the actual work condition of aircraft engine are key technologies ofdynamic balance. Considering the above problems, the paper focuses on the dynamic balance methodwithout test Weights. Identification of unbalance only use the vibration signals obtained from themeasurement. Firstly, a robust algorithm based on Huber-M estimation is proposed in the paper in thatit can effectively determine not only the amplitude and phase but also the position of unbalance in therotor. Besides, it has the advantage of insensitiveness to noise and therefore outlier data can beeliminated. The robustness of the method is demonstrated by an example. Secondly, based on modelupdating technique, an accurate finite element model of the rotor was built using the modal test data.A simulation analysis of the unbalance identification was carried on the model. Finally, theexperimental verification of the proposed algorithm was conducted by the measured rotor vibrationdata and finite element model. The simulation and experimental results show that, when the test datacontains an amount of noise, the method can accurately identify the unbalance of the rotor systemwith a small number of measuring data.