| Electric drive automation is the basis of realizing industrial modernization, and variable frequency speed control is the core technique of electric drive automation. With the farther development of variable frequency speed control, more and more commands and methods and put forward in terms of the performances of speed control system. Sorts of nonlinear factors and work variable parameter are always the conversion in the electric drive system. Common PID control can not settled, and in the basis of feedback control, state space control, a new control scheme, intelligent control come out. Intelligent control can validly conquer the effects of nonlinear factors and improve the robustness of system. Intelligent control is the hotpot in contemporary research, many scholars commit themselves to applying intelligent control to electronic control automation.Fuzzy control and neutral nets are used broadly in intelligent control technique. In this paper, speed loop adopting fuzzy control scheme which takes place of common PI controller, is to improve the performance of disturbance rejection. Double fuzzy compound control technique, which is equivalent to fuzzy-PID controller, has been applied in the implementation of the system. But compared with fuzzy-PID controller, the control rules have been greatly reduced, and the chattering phenomenon has been avoided for the applying of fuzzy strobe technique. The simulation results are shown to prove that fuzzy-PID controller can validly conquer the effects of nonlinear factors to the performance of speed control system.Stator resistance is identified by the technique of neutral net in this paper. A nonlinear mapping can be built from stator current to stator resistance, for stable temperature of stator resistance mainly depends on stator current at common operation of the motor. The structure of the system is ensured generally by linear system identification firstly, and then the system is described by neural network. Consequently, the simulation results proved that the convergence rate of neural network can be improved. U-I flux linkage model can be set exactly at low speed for considering the effects of temperature to stator resistance.What's more, in this paper a novel asynchronous machine speed control system by flux field is put forward. Because the intelligent observer of flux linkage is adapting, and rotor's parameters are avoided, the control system presents preferable robustness. |