Noise and vibration of electric machines are a major source of machine failures and performance degradation. For example, the vibration of spindle motors in disk drives directly impact the storage density. The vibrations of electric machines are an obstacle toward using them in ultrahigh-precision motion control applications.
A common source of noise and vibration in a permanent-magnet synchronous motor is the radial forces due to rotor eccentricity [or unbalance magnetic pull (UMP)]. If the rotor is displaced from the center of the stator, the air-gap magnetic field becomes distorted, resulting in destabilizing radial forces.
Eccentricity may arise in electric machines due to bearing wear, rotor imbalance, or manufacturing tolerance, and is therefore unavoidable. Efforts to mathematically model the eccentricity and the resulting vibrations in electric machines have been undertaken for a long time and are still an active topic for research. Prior force model are either too complex to implement or not entirely analytical. In order to fast calculate the eccentricity-induced radial forces or to compensate the UMP in real time, we need simple but accurate closed-form force models, which are not yet reported in the literature.
This study proposed a method to actively reduce the vibrations of permanent-magnet synchronous motors by independently controlling the phase currents based on the feedback of the rotor vibrations. The efficacy of the method is validated on a test rig that is equipped with a non-contact exciter and linear hall-type vibration sensors. The theroy is applied to a two-pole toroidally wound brushless dc (BLDC) motor, but the resulting force model can be used for any two-pole uniform-gap permanent-magnet synchronous motor.
In this paper, we have presented an analytical model of the radial forces resulting from the eccentricity present in a permanent-magnet synchronous motor, using the theory of perturbation. The predictions by the model are very accurate(maximum 2% mismatch with FEA). The force model is synchronous with respect to the eccentricity and is independent of the rotor angle. To validate the algorithm for vibration reduction, we used dSPACE and Matlab/Simulink for validation and data acquisition. Proposed algorithm depending on static eccentricity (0.2 mm at X-direction) reduce vibration about ?5 dB withing 100 Hz.