An Application Of Intelligent Methods In Direct Torque Control

The paper entirely summarizes the actuality and development of Direct Torque control (DTC) of the induction motors, and detailedly introduces the basic theory and principles. According to some main problems in the DTC, the application of intelligent method in the DTC is presented.Compared with Vector Control, the main advantage of DTC is that it directly controls flux and torque. The paper analyses the mathematical model and shows that the key of DTC is to select the right stator voltage vectors due to the stator flux and torque. The function of voltage vectors on the stator flux and torque is investigated.In order to realize the combination of the intelligent control methods and DTC theory, this paper systemically introduces the typical methods such as fuzzy logic control and neural networks etc. A series of schemes of intelligent DTC is put forward according to the different peculiarities of fuzzy control and neural network.To overcome the disadvantage of DTC, a new kind of fuzzy controller method is adopted. The results of simulation show that the torque response is faster and the track of steady stator flux is nearly equal to the anticipant one. This method proves highly effective.The paper designs a new kind of speed control model which utilizes the way of off-line training and on-line control for the neural network. Because neural network has an ability of parallel computation this method has a rapid response. This model can be used in the system of the speed estimation without speed sensor. Neural network controller is also presented and simulation results confirm the validity of the method.