An Improved MTPA Control Based on Amplitude-Adjustable Square Wave Injection

Abstract

In traditional maximum torque per ampere (MTPA) method based on signal injection, fixed wave amplitude and constant integral coefficient are employed, which actually degrade the optimal current angle tracking performance in variable conditions, especially under heavy loads. In order to solve this problem, an improved MTPA method based on amplitude-adjustable square wave injection is proposed. First, the relationships among optimal current angle, square wave amplitude and electromagnetic torque are analyzed. The analysis results indicate that if fixed wave amplitude is adopted, the fluctuations in the obtained current angle increases when load torque rises. Based on this relationship, an amplitude-adjustable square wave virtual signal injection MTPA method is proposed. By adjusting square wave amplitude according to the load torque, steady-state fluctuations of the obtained current angle can be reduced compared to conventional method under heavy loads. Meanwhile, to improve the MTPA dynamic response under load change conditions, an adaptive fuzzy control method is proposed for adjusting integral coefficient. Finally, the validity of the improved MTPA method is verified through experimental results.