Improved Classic Direct Torque Control Based on Doubly Fed Induction Generator Use Neural Network

Direct Torque Control (DTC) simplifies and improves the dynamic performances when applied on the Doubly Fed Induction Generator. However, the variable switching frequency represents the most significant drawback of the Classical Direct Torque Control strategy, which mainly depends on the sampling frequency, the Look-Up Table (LUT) structure and hysteresis bands. In this paper we proposed an improvement depends on the artificial neural networks (ANN) approach, the main aims are dealing the ripple caused by the hysteresis comparators and decreasing the computational burden caused by the Look-Up Table. As such, an artificial neural network replaces the conventional hysteresis comparators a the Look-Up Table in order to overcome the instability caused by the undesirable ripple in the current and torque side therefore, use the artificial neural network represented a desirable solution to address these types of problems. The simulation results are presented to confirm the effectiveness of the proposed method.