Adaptive proportional-integral speed controller for direct torque control of induction motor using neural networks

The design and implementation of an adaptive neural network based proportional-integral controller (NNPIC) were carried out on the direct torque control (DTC) system. The proposed controller consists of neural network and projection algorithm, in order to tune its parameters adaptively online under environmental variations. The experimental results proved that the performance of the proposed controller was quite better than the conventional PI controller by the criterion, root-mean-square-error (RMSE).