Direct torque control of electric vehicle drives using hybrid techniques

Abstract

Permanent magnet synchronous motors (PMSM) have the capability of delivering a high torque-to-current ratio, better efficiency and low noise. Because of the above-mentioned factors, PMSMs are commonly employed in variable speed drives, especially in electric vehicle (EV) applications. Without the usage of electromechanical devices, the conventional direct torque control (DTC) can control the speed and torque of PMSM. DTC is highly efficient, fast-tracking and provides smooth torque while limiting its ripple during transient periods. There are many benefits to using a DTC-controlled PMSM drive, including quick and reliable torque reaction, high-performance control speed, and enhanced performance. This research examines the use of the DTC approach to enhance the speed and torque behavior of PMSM. The jellyfish search optimizer (JSO) is used to adjust the DTC's responsiveness and tailor the controller's best gains. In order to train the adaptive neuro-fuzzy inference system (ANFIS) controller, JSO data are utilized. The simulation outcomes demonstrate that the proposed JSO-ANFIS controller achieves a minimal torque ripple of 0.26 Nm and preserves the speed with a harmonic error of 1.21% while contrasted to existing methods.