These days, many researchers are working diligently on controlling algorithms and motor design to enhance the performance of PMSM

In drive applications where a fast response is required, direct torque control is recommended. Quick response in torque for PMSM is obtained by manipulating the angle quickly. Quick angle manipulation is achieved by omitting the null voltage level so that the angle can be varied. DTC controls torque and flux directly. In this way, DTC's field weakening capabilities are enhanced. It is proposed to implement direct torque control based on space vector (DTC-SVM) to minimize flux and torque ripple in

Space vector modulation-based direct torque control (DTC-SVM) is proposed as the control scheme which reduces the ripple for both torque and flux when it is seen from a generic DTC perspective. A direct torque control system, based on space vector pulse width modulation, is utilized in

The mathematical model of PMSM is depicted by equations below:

The coordinate transformation for dq-abc and αβ-dq is performed by equations (10) and (11) shown below respectively.

The PMSM model, inverter model, signal processing, and switching signal generation are done in a d-q axis using Matlab 2021b. During signal manipulation, flux and torque estimation are performed, which are used for signal generation. The voltage selection for CDTC and TCSF-DTC is executed in such a way that the applied voltage gives a fast response in monitoring flux and torque. According to the work stated in

According to the TCSF-DTC, the flux angle is calculated by adding the output of the torque controller and the flux angle calculated from the estimated flux.

We can represent torque as a function of flux angle, which manipulates the flux mathematically. It is possible to have a positive or negative torque error.

When the reference torque differs from the electromagnetic torque, there is a torque error. Positive errors mean that the reference is greater than needed, and vice versa. The torque equation (4) shows that the quadrature current component and the d-axis flux are significant contributors to torque production. In particular, the quadrature current plays a crucial role. When the torque error is negative, the flux angle decreases, which decreases the flux on the q-axis. According to equation (3), q-axis flux depends on q-axis current.

Whereas the d-axis flux changes from b to a, according to the picture depicted in

Parameter Magnitude Parameter Magnitude Parameter Magnitude Parameter Magnitude Ld 21.3mH Bm 0.001 Lq 24.2mH Tl 10 Nm fs 500hz Ra 0.24 ohm mf 0.542wb ɷ 50 rad/sec Im 10A P 8 J 0.0024 Vdc 200V

The simulation is conducted at a constant speed. The torque was made to change at the magnitude of half of the rated torque, seventy-five percent of the rated torque, and half of the rated torque respectively, and these detachment times are 0 to 1 second, 1 to 2 seconds, and 2 to 3 seconds respectively. Compared to CDTC, TCSF-DTC has one additional PI controller as is shown in figure 1. The PI controller is used to further manipulate the flux angle. Due to this double manipulation on flux, the flux ripple reduction of TCSF-CDTC is better than CDTC. The torque ripple control components are the same for both cases. In addition to the double manipulation of flux magnitude using both PI controller and conventional DTC regulation, the voltage selection is based on the error reduction strategy. So the effect of enhanced voltage selection and double manipulation of flux makes the system have a better performance. In the simulation, the switching state is selected according to the methodology shown in figure 1(b). Simulated outcomes of both the CDTC and TCSF-DTC methods were obtained at a constant speed. Due to TCSF-DTC's double control manipulation on flux values, ripples on flux are minimized when TCSF-DTC is used compared to CDTC. In both cases, the torque performance is similar as shown in

This study presents two direct torque control methods for PMSM at constant speed operation. The two DTC applied for PMSM were compared in terms of torque ripple, flux ripple, and input voltage total harmonic distortion. Simulated outcomes of both the CDTC and TCSF-DTC methods were obtained at a constant speed. According to the simulation result of TCSF-DTC's, due to double control manipulation on flux values, ripples on flux are minimized when TCSF-DTC is used compared to CDTC. In both cases, the torque performance is similar. From this work, it can be seen that CDTC is preferable for torque ripple reduction as the scheme has no computation for angle manipulation which is the case of TCSF-DTC. As the fine control of an electric drive is the hot topic for a research area, an interested person can consider TCSF-DTC with a duty ratio optimization for a torque error minimization as TCSF-DTC has a superior flux ripple performance.