Application of neuron speed observer and dead-time compensation method in direct torque control system

On the basis of artificial neural network theory, this paper has put forward a new kind of speed observer. The observer is good in robustness of motor parameter changes, high in the observation precision. The speed identification convergence is fast, especially keeping higher precision in low speed. At the same time, the paper puts forward a new method of dead-time compensation, applies the new observer and a new method to direct torque control (DTC) system, which have improved the performance of DTC in an all-round way. The simulation results show that this control system has good static property and dynamic performance.