Study on application of the asynchronous motor direct torque control based on neural network

A new improved direct torque control (DTC) method is proposed for an asynchronous motor direct torque control system, in this paper, after studying and analyzing model reference adaptive control (MRAC) and adaptive linear neural network motor speed identification. This method adopts an MRAC identification model with high-pass filter part and low-pass filter part. Based on this condition, we design a new speed observer. The efficiency and superiority have been demonstrated by comparing the simulation results of this method and that of others, and it has a better control performance, higher identification precision and practical significance. We built a relevant experimental system adopting the digital signal processor TMS320F240, and the validity of this method has been demonstrated by the experimental results. Finally, the applicable area and research orientation of this method have been indicated in this paper.