An improved sensorless vector controlled induction motor drive employing artificial neural networks for stator resistance estimation

This paper describes how an artificial neural network (ANN) can be employed to improve a model reference adaptive system closed loop flux observer (MRAS-CLFO) used for speed estimation in a vector controlled induction motor drive. The system uses the ANN to estimate changes in the stator resistance which enable the MRAS-CLFO models to work more accurately. The overall effect is an improvement in speed estimation and experimental results are presented to verify this.