|
|
Full decoupling control of bearingless switched reluctance motor based on support vector machine inverse system |
School of Electrical and Information Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China |
|
|
Abstract In view of the magnetic saturation, non-linearity and coupling, it was very difficult to establish mathematic model and realize decoupling for bearingless switched reluctance motor (BSRM). Based on least squares support vector machine (LS-SVM), a novel mathematic model involving magnetic saturation was proposed to establish inverse system to decouple radial suspending forces and torque. The closedloop decoupling control experiments were conducted on dSPACE platform to verify the decoupling ability and the robustness with time-variant parameters. The results show that the proposed method has advantages of both nonlinear approximation ability of LSSVM and decoupling linearization character of inverse system method. The constructed LS-SVM inverse system decoupling controller has strong autoadaptive performance and robustness. The unsuitability for magnetic-saturated condition of existing models and corresponding decoupling methods which are based on nonsaturation hypothesis can be solved.
|
|
|
|
|
|
|
|