Sensorless DTC Drive of Induction Motor Using 3-Level Inverter

Abstract

This paper presents a novel switching table based Direct Torque Control (DTC) of induction machine modified for the 3-level inverter. Due to its large number of voltage vectors, torque and flux ripples can be reduced by distinguishing small and large torque and flux errors and choosing different switching vectors for each case. This work also introduces two sensor-less DTC systems: the adaptive flux observer and the Model Reference Adaptive System (MRAS). Based on the dynamic model of the induction machine, the adaptive flux observer estimates the rotor flux and speed according to adaptation laws that must ensure the stability of the system. In another hand, the MRAS technique consists on minimizing the error obtained from the comparison of two of rotor flux models, one based on voltage and current and the other based on current and rotor speed, using an adaptive mechanism. The proposed schemes were verified by simulation under Matlab/Simulink environment. Simulation results show good dynamic performances and low ripples of torque response with the employment of 3-level inverter. As for speed estimation, both techniques allow a good estimation of rotor speed, but MRAS technique gives better dynamic performances.