IEEE Access (Jan 2019)
Torque Ripple Reduction and Flux-Droop Minimization of DTC With Improved Interleaving CSFTC of IM Fed by Three-Level NPC Inverter
Abstract
Direct torque control (DTC) is known for its fast control in AC motor drives due to its simple control structure that directly controls the torque without the need for modulation blocks or frame transformations. However, when used in induction motor (IM) drives it has three main drawbacks: large torque ripples, variable switching frequency, and sector flux-droop at the low-speed region due to the employment of the torque hysteresis controller (THC). Torque ripple minimization can be achieved in DTC drives by replacing the originally proposed two-level inverter with a three-level neutral-point clamped (3L-NPC) inverter. Nevertheless, the switching frequency remains variable and low, which produces an elevated torque ripple and asymmetrical switching signals for the inverter. In addition, sector flux-droop resulting from driving the IM at the low-speed region produces a high current distortion that consequently eliminates the robustness of DTC. To alleviate these problems, an interleaving constant switching frequency torque controller-based DTC (CSFTC-DTC) was implemented. It improves the operation of the IM at the low-speed region by increasing the duty-cycle of the applied active voltage vector and reducing the duration of the applied zero vectors. Although the conventional CSFTC-DTC regulates the stator flux of the IM at the low-speed region and minimizes the total harmonic distortion (THD) of the stator current, it produces a high torque ripple-one of the main disadvantages of classical DTC. In this paper, an interleaving CSFTC-DTC is proposed to subdivide the duty cycle of the applied vectors of the 3L-NPC inverter to limit the influence of the large duty-cycle of the applied vectors on flux-regulation and torque ripples. The simulation and experimental results presented validate the effectiveness of the proposed method over the conventional method.
Keywords