An improved rotor flux based model reference adaptive controller for four-quadrant vector controlled induction motor drives

High quality performance of vector controlled induction motor (IM) drives to a large proportion depends upon the rotor resistance which varies significantly with temperature notably at low speed. Many techniques are proposed till date but, model reference adaptive controller (MRAC) is one of the ace methods among the numerous due to its performance and straightforward stability approach. The rotor flux (RF) -error based MRAC consists of two models, viz. reference and adaptive models used simultaneously to estimate the d-q components of the rotor flux using IM line currents and voltages. However, in the case of conventional RF-MRAC based drive, the accuracy of the measured voltage drastically falls because of the significant amount of voltage drop across the stator resistance in the low speed region. To overcome this limitation, in this paper, the d-q components of the stator voltages are computed using the error signal between the reference and measured d-q current components with Proportional Integral (PI) controllers and representing the relevant equations in the synchronously rotating frame. Hence, instead of using voltage and current sensors, only current sensor is required in the proposed scheme. Thus, the proposed RF-MRAC based drive performs satisfactorily at the low speed as direct dependence on the measured voltage is eliminated. The proposed drive´s performance with the RF-MRAC is validated for various speed ranges and patterns, both by Matlab simulation and experimental verification. A stability study in all the four quadrants of drive´s operation and sensitivity study against motor parameter variations are carried out in Matlab/ Simulink. Experimental results obtained by a dSPACE-1104 based laboratory prototype are presented in the paper.