Self-Adaptive-Selected-Harmonic Strategy for Radial Vibration and Acoustic Noise Attenuation of PMSM With Accelerometer Feedback

Radial vibration suppression in motor control can be realized by off-line calibration methods such as multi-physical co-simulation or traversal searching to find the best injected current harmonics. However, these methods are often prone to mechanical installation, temperature changes, and other uncertainties in practice. By contrast, this paper presents a method utilizing the real-time radial acceleration signals to suppress the motor radial vibration and achieve further noise attenuation under real-world uncertainties. To realize fast convergence of the optimal injected harmonics, a self-adaptive-selected-harmonic (SASH) radial vibration suppression strategy of permanent magnet synchronous motor (PMSM) is developed. Based on the pole-slot combination, the harmonic order of the most influential vibration is theoretically calculated. Then, with the radial acceleration signal acquired by a motor-housing-installed accelerometer, the real-time amplitude of the target vibration order is extracted, followed by a fast simulated annealing algorithm to search the optimal amplitude and phase of injected voltage harmonics. Experiments have verified the effectiveness of this proposed strategy, which achieves the suppression of selected-harmonic vibration from 0.03g to below 0.005g, as well as more than 8 dBA average attenuation of the selected-harmonic radial acoustic noise in a wide speed range.