基于神经网络逆系统理论无轴承异步电动机解耦控制

Decoupling Control of Bearingless Induction Motors Based on Neural Network Inverse System Method

无轴承异步电动机是一个多变量、非线性、强耦合的系统,其径向力和转速之间存在交叉耦合,若要实现电动机转子稳定悬浮和运行,必须对电动机转速和径向力进行动态解耦控制。为此,本文提出了一种基于神经网络逆系统的无轴承异步电动机解耦控制方法。理论分析表明,此方法可以将无轴承异步电动机动态解耦成位移子系统、转速子系统和磁链子系统,从而简化外环控制器的设计,进一步提高整个系统的控制性能。最后,对采用所提解耦方法的整个无轴承异步电动机控制系统进行了仿真和初步的实验研究,结果验证了该解耦方法的有效性。

A bearingless induction motor is a multi-variable,nonlinear and strong-coupled system.The cross-coupled relation exists among speed and radial forces.In order to realize the rotor suspending and the motor working steadily,it is necessary to realize dynamic decoupling control among speed and radial forces.In this paper,a novel decoupling control method based on neural networks inverse system is proposed for bearingless induction motors.Theoretical analysis shows that by using the proposed method the system is decoupled into two independent linear displacement subsystems,an independent linear rotor speed subsystem and an independent linear magnetic linkage subsystem.The design of outer controllers becomes easier,and the whole system control performance is further improved.At last,taking bearingless induction motor system which adapts the proposed method as object,the study of simulation and primary experiment are done.The validity of the proposed method is proved by the simulation and primary experiment results.