Torque ripple reduction of induction machine a comparison between two approaches DTC-SVM
No Thumbnail Available
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Oum El Bouaghi
Abstract
The paper deals with a comparison between two DTC-SVM approaches: DSVM-DTC with a hysteresis controllers and MPDTC with a predictive controller. In DTC-SVM, the switching frequency is limited by the sampling frequency; its theoretical maximum value is half the sampling frequency. However, in reality the switching frequency is lower than this value, and thus, high flux and torque ripples occur compared with modulator-based control methods. In order to overcome this, an optimization problem is formulated and solved in real time. Thereby, apart from the regulation of the torque and the flux magnitude to their references, an additional control objective should be met: the minimization of the torque and flux ripple. To do so, the time point at which the switches of the inverter should change state is calculated. Following the formulation of both approaches, their implementation in the Matlab-Simulink environment has been treated. The transient behavior as well as the steady state features of the induction motor drive under both DTC-SVM approaches are compared and commented. The comparison is based on several criteria including: static and dynamic performance, structure and implementation complexity, and especially decoupling, torque and flux ripple. It has been found that the second approach MPDTC yields high dynamic performances in the whole speed range. These performances are characterized by a low torque ripple with a high capability to eliminate the demagnetization problem which penalizes the flux regulation at low speeds in the first DTC-SVM approach.
However, the second DTC-SVM approach with a predictor controller requires much more CPU time than the first one.
Description
Keywords
Induction motor; Direct torque control; Predictive direct torque control; :SVM Space vector modulation