Model-Free ANN-based control for Three-Phase Induction Motor Speed Control

The speed control of Induction Motors (IMs) is one of the most complex control problems, and is an active research topic in the control field. In this research work, a neural controller is proposed to address this problem. Our proposal uses an artificial neural network (ANN) with an online learning algorithm. Moreover, we developed a modified version of the Adam optimization algorithm to train the ANN online. Experimental results on a Three-Phase IM test bench show that the ANN-based controller is able to learn in real-time and it outperforms the widely used proportional integral (PI) controller.