Torque ripple minimization for interior permanent magnet synchronous motor considering both structural shape and control of current harmonics

Abstract

Interior Permanent Magnet Synchronous Motors (IPMSMs) can provide high efficiency and power density and mainly applied to home appliances (i.e., electric fan and washer) and industrial applications (i.e., electric vehicle/hybrid electric vehicle and generator) as a high-performance electric motor. Since IPMSMs have torque ripple issues, many studies attempt to solve this issue. The torque ripple causes acoustic noise, mechanical vibration, and instability of control. Although torque ripple is generally reduced using either structural shape optimization of the motor or control of current harmonics, the reduction of torque ripple reaches the limit. Therefore, the objective of this research is minimizing the torque ripple considering both structural shape of the motor and control of current harmonics. In this study, the design variables for optimizing the structural shape of the motor (i.e., tooth of stator, length of air barrier and permanent magnet, notch, and angle between permanent magnets), and fifth- and seventh- order current harmonics are chosen for minimizing the torque ripple. In order to consider magnetic saturation, the common finite element method software (i.e., ANSYS MAXWELL) is employed and the performance of electric motor is analyzed. Furthermore, the circuit design and analysis software (i.e., SIMPLORER) and the tool for designing the gate driver and current controller (i.e., MATLAB SIMULINK) are employed simultaneously to consider the real impact occurring in the industrial field (i.e., nonlinearity of Pulse Width Modulation (PWM) inverter). After the optimization process, the post processing is conducted. As a result, our findings confirm that the optimized motor can maintain the average torque and minimize the torque ripple at rated speed and maximum speed. Furthermore, since the performance of the optimized motor is analyzed using the co-simulation system, the torque waveform is extracted by reflecting the real impact of nonlinearity of PWM inverter and PID controller.