Research On Multi-stage Rotor Balancing Method Based On Unbalance Parameters Equivalent Model

In high-speed rotating machinery,vibration is the main factor that leads to fatigue damage,failure,and shortened lifespan.The main technical means of controlling machine vibration is to balance the rotor.Improving balance efficien cy and quality can effectively improve the economic benefits and high-speed stability of rotating machinery.In the field of aviation engines,with the development trend of rotor structure refinement and flexibility,and the increasing driving force of operating time and decreasing unplanned maintenance,rotor balancing technology faces increasingly greater challenges.How to improve balance efficiency and quality has become a key issue for engine manufacturers,users,and maintenance units.The balance method based on a dynamic model can obtain the unbalance parameters required for balancing through a single trial run,and has the advantage of high balance efficiency.However,this method has problems such as inaccurate identification of unbalance parameters,inability to balance multi-stage rotors,and inability to update the dynamic model in real time,making it difficult to achieve high-quality balance of multi-stage rotors in aviation engines.As a result,the study concentrates on the need for balancing quality and efficiency of multi-stage rotor of aero-engine,and starts the research around balancing method based on dynamics model,the primary research contents are as follows:In order to address the problem that the current balancing method based on dynamic models is only applicable to rotors of a few stages,the research on multi-stage rotor balancing method for aero engines is carried out.Starting from the Euler-Bernoulli continuous beam model,an equivalent model of the unbalance parameters is established based on the vibration response under unbalance excitation,and the complexly distributed unbalance on the multi-stage rotor is equated to the equivalent unbalance on the fixed correction planes.The research shows that when the rotational speed of the rotor is lower than the Nth critical speed of the system,the original complex unbalance distribution can be transformed into N equivalent unbalances distributed on correction planes,allowing for the addition of balance weights on fixed correction planes to balance the multi-stage rotor.To deal with the problem of inaccuracy of current unbalance parameter identification methods,research on highly accurate unbalance parameter identification methods is carried out.A direct mapping model describing the relationship between the vibration response of the measurement point and the unbalanced force is established by using a reduced-order model that takes gyroscopic effects into account.The steady-state vibration response is used as input,and regularization techniques are employed to solve the direct mapping model,obtaining the unbalance forces of the system.Spectral correction techniques are then used to extract the amplitude and phase of the unbalance forces,achieving the goal of improving the accuracy of unbalance parameter identification.To address the problem that the current model updating method is not able to update model under the operation condition,the research on the dynamics model updating method under the operation condition is carried out.Analyze the distribution properties of the operating state modal parameters of the system,improve the criteria for identifying stable modes,achieve the rejection of erroneous and harmonic modes and the identification of structural modes,and accomplis h the automatic extraction of structural modes of the system in the operating state by fusing with hierarchical cluster analysis technology.Then,by analyzing the influence of structural parameters on the system modal parameters to determine the parameters to be updated and establishing an objective function that minimizes the feature frequency based on operational modal parameters,the differential evolution algorithm is employed to solve the updating parameters,achieving the updating of dynamic models and providing an effective solution for model updating under operating conditions.Build a rotor-bearing system and carry out experimental validation studies.First,the effectiveness of the model updating method is verified.The results show that the updated dynamic model significantly improves the prediction accuracy of unbalance response.In the simulated rotor of an aero-engine,the method improves the amplitude and phase prediction errors at low and high balance speeds by 27.7 %,4.1°,and 7.4 %,-1.4°,respectively.Next,the feasibility of the parameter identification method is verified.The results show that the proposed unbalance parameter identification method can identify unbalance parameters with higher accuracy compared to classical unbalance parameter identification methods.For the three-disk rotor-bearing system,the proposed method reduces the identification errors of amplitude and phase by 5.3 %,3.5° and 34.1 %,5.9° at low-speed and high-speed balancing speed,respectively.Finally,the effectiveness of the multi-stage rotor balancing method is verified.The results demonstrate that when the rotational speeds are low-speed balancing speed,first-order critical speed,high-speed balancing speed,and second-order critical speed,respectively,the average vibration amplitude at four measurement points of a simulated rotor of an aero-engine is reduced by 69.0 %,70.7 %,54.3 %,and 47.8 % after balancing.