Application Of Grey Prediction Fuzzy Controller In DTC

Direct torque motor control system is one of the most effective speed control technology of A. C. drive, mathematical model of the motor in direct torque control system is composed by the voltage equation, flux linkage equation and torque equation. When controlling it with traditional control strategy, the dynamic control effective is often poor. There are lag problem in general control theory when responding the change in motion state to the change in controlling force. It will be very difficult to achieve good real-time performance, accuracy, adaptability.This paper summarized an overview of the development of direct torque control system systematically, and introduced the basic theory and principle of direct torque control detailedly. On which bases, to achieve real-time, accurate control of the system, on the one hand, system behavior data is extracted with gray prediction, the law of motion of the system is found, thus the future system motion trend could be predicted according to the law; and according to the future system motion trend, corresponding control decision can be ascertained to achieve accurate predictive control. On the other hand, combined with fuzzy control, firstly a fuzzy control table can be got through off-line calculation , then the mode switch could be chose while looking up the table, so as to achieve fuzzy direct torque control. By the gray prediction algorithm, the combined forecast weight coefficient could be optimized online, and the corresponding fuzzy inference can give the best or second best controlling scenario. So as to solve the problems like large overshoot, slow response, instability and so on in the conventional fuzzy controller in direct torque control process due to the electrical model parameters won't be changed.This paper developed software and tested system in the experimental of platform asynchronous motor direct torque control system using digital signal processor DSP as microprocessors, on a basis of simulation of asynchronous motor direct torque control. Experimental system uses a appropriative motor control DSP chip TMS320F2407. The chip integrated a number of function modules which motor motion control needed, and a reduced the design work of external circuit. Experimental results verified the effectiveness of the combination of gray prediction and fuzzy control method, which overcame the hysteresis effect on the system, allowing the system obtain satisfactory results in dynamic respond, anti-interference ability and robustness, etc. At the same time, the speed, torque and other information under motor load conditions is collected and observed. Experimental results show that the control precision and dynamic qualities of the system are better than ordinary fuzzy controller.