Decoupling Control of Bearingless Induction Motors Based on Least Squares Support Vector Machine Inverse

摘要

To realize the nonlinear dynamic decoupling control of a bearingless induction motor (BIM), which are multivariable, strongly coupled and nonlinear, a decoupling control strategy based on least squares support vector machine (LSSVM) inverse is proposed. Firstly, the inverse of original system is analyzed to confirm the existence of BIM's inverse. Then, the LSSVM inverse is used to approximate the inverse model of BIM and the pseudo-linear system is established. The constructed pseudo-linear system is equivalent to two independent linear displacement subsystems, an inde- pendent linear rotor speed subsystem and an independent linear flux linkage subsystem. Thirdly, a closed-loop controller is designed for the pseudo-linear system to improve the performance of the BIM. Finally, the simulation examples evaluate the performance of the proposed strategy. Compared with the inverse system control, system transient response and decoupling results are improved using the LSSVM inverse, and the disadvantages of inverse system method depending on the exact system model are overcome.