Sensorless Vector Control Based on MRAS for Three-Phase Induction Machines

Squirrel cage induction motors are widely used because of their high robustness and low manufacturing and maintenance costs compared to other machines. However, one major drawback of induction motors is the complexity of speed control, especially when compared to dc motors, which have a much simpler control process. Several methods have been developed to control the speed of these machines, with vector control being particularly renowned. Vector control requires knowledge of the rotor position for implementation. This is where using estimators instead of speed sensors becomes interesting, as they offer greater noise immunity, reliability, and ease of implementation. One of these techniques is the Model Reference Adaptive System (MRAS), which offers significant advantages. These techniques involve using two different models to calculate the variable to be estimated (the reference and the adjustable models), making them particularly reliable when the motor parameters are poorly known or subject to large variations. This paper examines the MRAS scheme based on rotor flux at different speeds to evaluate the performance of the speed estimator and the implemented control.