基于微分几何的无轴承电机神经SMVS解耦控制

Neural SMVS Decoupling Control of Bearingless Motor Based on Differential Geometry

以无轴承永磁薄片电机为研究对象,阐述了其径向力悬浮机理,推导了径向悬浮力数学模型。采用非线性微分几何的方法研究了无轴承永磁薄片电机在负载运行时径向悬浮力之间的解耦控制问题,实现电机径向悬浮力与悬浮力绕组中电流之间的解耦控制。将原耦合系统解耦和基本线性化成独立的伪线性系统,并对解耦后的伪线性子系统设计了神经滑模变结构控制器。最后对设计的控制系统进行仿真试验,验证了这种解耦控制方法的可行性。仿真结果表明,基于非线性微分几何的神经滑模变结构控制器方法能较好实现径向悬浮力的稳定独立控制。

Aiming at a bearingless permanent magnet slice motor,the radial suspension principle of this motor is introduced,and mathematical model of radial suspension forces are deduced.The decoupling control question has been investigated for radial suspension forces of the bearingless permanent magnet slice motor at load adopting the nonlinear differential geometry.The decoupling control has been realized among radical suspension forces and currents in radical suspension force＇s windings.The original coupling system is decoupled and linearized,and the neural network sliding mode variable structure（SMVS）controller is designed for the decoupled linear subsystems.Finally,the feasibility of the method is validated by the results of simulation.The simulation conclusions show that based on the nonlinear differential geometry neural SMVS controller method can achieve better stability of radial suspension force independent control.