Sensorless PMSM drive with MRAS-based adaptive speed estimator

The permanent magnet synchronous motor is increasingly playing an important role in advanced motor drives. An encoder or resolver attached to the shaft of the motor typically supplies the feedbacks required for motor speed control. Many advanced motor drives can not tolerate the use of these feedback devices because of reliability concerns in a harsh environment, price, space and weight limitations. In this paper model reference adaptive system-based adaptive strategy has been used for speed estimation. In compare to previously developed methods in the literature such as: EKF, neural networks and sliding mode control this method consumes less computational time and it is easy to implement in microcontrollers and DSP. In this proposed method the Popov´s criterion is used for adaptive speed estimation. The validity of the proposed adaptive strategy has been verified by simulation and experiments