Direct torque control of induction motor drive system with adaptive sliding-mode neuro-fuzzy compensator

The paper deals with the concept of an adaptive compensator based on the MRAC structure for the Direct Torque Control of an induction motor drive. The adaptive speed compensator uses fuzzy neural network equipped with an additional option for on-line tuning of its chosen parameters. In the paper a sliding-mode PD fuzzy logic controller is used as the speed compensator, whose connective weights are trained on-line according to the error between the state variable of the plant and the reference model. To the stator flux reconstruction the current model is used. It is shown that additional adaptive system in the speed control loop improve the properties of the drive for different drive conditions. The simulation results are verified in experimental tests.