Double Random SVPWM High-Frequency Electromagnetic Vibration Suppression of Permanent Magnet Synchronous Machine Based on Improved Mersenne Twister Algorithm
Chen Zhenfei, Wan Xiangmin, Chen Shutong, Wang Lei, Wang Feng
School of Electrical and Power Engineering Hohai University Nanjing 211100 China
Abstract：In the inverter-driven permanent magnet synchronous machine (PMSM) control system, high-frequency current harmonics near the switching frequency and its multiples are generated using space vector pulse width modulation (SVPWM), which brings high-frequency electromagnetic vibration. Therefore, a double random SVPWM control method combining random switching frequency and random zero vector is applied to the high-frequency current harmonic spectrum expansion. Meanwhile, the random number is generated by the improved Mersenne twister (MT) algorithm, which enhances the random performance of the random sequences and ensures the spreading effect of the double random SVPWM control method. Firstly, random zero vector control can be achieved by changing the action time of the zero vector and reassigning the randomized zero vectors into the space vectors. Secondly, the switching frequency of the traditional SVPWM control method is fixed, and the random switching frequency control can be achieved by changing the fixed switching frequency of the inverter and dispersing the harmonics at the switching frequency and its multiples into the specified frequency domain. Thirdly, the random number is generated by the improved MT algorithm, which is applied to the double random SVPWM control of the permanent magnet synchronous machine to enhance the degree of freedom and spatial traversal of the random sequence. The new control method proposed is named LKMT-DRC. Experimental verifications are conducted on a 4.4 kW fractional-slot permanent magnet synchronous machine. The PMSM phase current harmonics and vibration acceleration under the conventional SVPWM control and LKMT-DRC control are compared, and the vibration suppression effect of the LKMT-DRC control is analyzed using different spreading ranges. Under the inverter power supply mode, the harmonic frequencies introduced by the SVPWM control mode are mainly fk±2f0 and fk±4f0, and the frequency of the introduced high-frequency radial electromagnetic force wave is mainly fk±f0. Compared with the traditional SVPWM control mode, the high-frequency harmonics concentrated on the switching frequency. Its integer multiples can be effectively dispersed using LKMT-DRC control mode, and the vibration suppression effect caused by the high-frequency electromagnetic force wave can be significantly reduced. Meanwhile, the random numbers generated by the proposed LKMT algorithm can improve the randomness and spatial traversability of the random sequences, which ensures the effectiveness of the double-random SVPWM control method. The contributions of the proposed double random SVPWM control method based on the improved MT algorithm are as follows. (1) A mathematical model of the high-frequency radial electromagnetic force introduced by PMSM under inverter power supply conditions is derived, and the effect of the high-frequency electromagnetic force under inverter power supply mode on motor vibration is analyzed. (2) The LKMT-DRC control method is proposed, which effectively reduces high-frequency harmonic content and suppresses high-frequency electromagnetic vibration. (3) The effects of different spreading ranges on the final damping under the LKMT-DRC control mode are analyzed, and the appropriate ranges are indicated.
2025, Vol. 40 
