Neural network inverse system decoupling fuzzy self-tuning proportional-derivative control strategy of a bearingless induction motor

In view of the problems that the analytical inverse system decoupling method of bearingless induction motor is sensitive to the change of motor parameters and is greatly affected by unmodeled dynamics, and that of traditional proportional-derivative controller lacking the self-adaptive regulation ability, a neural network inverse system decoupling fuzzy self-tuning proportional-derivative control strategy is proposed for a bearingless induction motor system. First, under the conditions of considering the stator current dynamic of torque winding, and by neural network inverse system method, the bearingless induction motor system is decoupled into four pseudo-linear integral subsystems. Second, the traditional proportional-derivative controller is improved, and the fuzzy control algorithm is used to adjust the parameters of improved proportional-derivative controller adaptively. Thus, a neural network inverse system decoupling fuzzy self-tuning proportional-derivative control system is constructed for a bearingless induction motor. The simulation experimental results show that the proposed control strategy not only can effectively improve the stability, robustness and steady-state control accuracy of bearingless induction motor system, but also can significantly improve the dynamic response speed and the ability to resist the influences of motor parameter change and load disturbance.