Real-time DSP implementation of DTC neural network-based for induction motor drive

This paper discusses the application of neural networks to control induction machines using direct torque control (DTC). The authors proposes to replace the conventional selector switches statements of the voltage inverter by a selector based on artificial neural networks (ANN), which is able to manage in the same way the switches states. In order to improve the low speed performance a closed loop voltage-current model estimator is employed to evaluate the stator flux. Experimental results for a 3kW induction motor are presented and analyzed using a dSPACE system with DS1104 controller board based on digital processor Texas Instruments TMS320F240 DSP. (5 pages)