Sensorless speed identification of vector—controlled induction drives via neural network based estimation

Abstract

An application of neural network technique to sensorless speed identification of vector-controlled induction drives is implemented in this paper. In the proposed scheme, neural networks help estimate the rotor speed through stator measurements, allowing investment on speed sensors to be saved. High performance of this induction drive however, can be ensured. We evaluated our approach based on the speed response and temperature rise perspectives. Load disturbance effects on the proposed approach were also included. The test results demonstrate the feasibility and superiority of the proposed method for the applications.