Current Waveform Optimization for Low Noise Permanent Magnet Motors

Abstract

During torque production, the varying magnetic fields inside an electric motor excite vibration that radiates acoustic noise. In consumer applications, this noise can influence the perceived product quality. Noises from propulsion and auxiliary electric motors on naval vessels create an acoustic signature that increases detectability. The dominant noise occurs at twice the electrical frequency (2E). For permanent magnet (PM) machines, the attraction between the rotor permanent magnets and the stator iron causes a radial force that varies sinusoidally around the stator. The stator coil currents generate a rotating magnetic field that produces rotor torque. This paper develops a new commutation strategy for PM machines that uses higher stator currents to minimize 2E noise by reducing radial force ripple without sacrificing torque. An analytical model is developed that predicts rotor torque and radial force ripple as functions of the stator currents. Based on this model, the phase currents are optimally commutated to maintain constant torque production and reduce force ripple. The optimal commutation is numerically investigated on a small PM motor using ANSYS FEA. The ANSYS results show a 30% reduction in force ripple at no load.