Decoupling control of a 5-degree-of-freedom bearingless induction motor based on least squares support vector machine inverse

Abstract

In order to realize the nonlinear dynamic decoupling control of a 5-degree-of-freedom bearingless induction motor which consists of a 3-degree-of-freedom magnetic bearing and a 2-degree-of-freedom bearingless induction motor, a decoupling control strategy based on least squares support vector machine inverse is proposed. First, on the basis of introducing the structure of 5-degree-of-freedom bearingless induction motor, the mathematical model is derived. At the same time, the reversibility of the mathematical model is analyzed. Second, the inverse model of 5-degree-of-freedom bearingless induction motor is identified using the regress capability for high-dimensional nonlinear least squares support vector machine within limited samples. In addition, the particle swarm optimization algorithm is used to optimize parameters of the least squares support vector machine, which improves the fitting and predictive precision of the model. Third, combining the least squares support vector machine inverse model with the original system constitutes the pseudo linear system and then proportional-integral-derivative closed-loop controllers are designed to realize the compound control for the 5-degree-of-freedom bearingless induction motor system. The dynamic decoupling control among the radial and axial displacements, speed, and flux linkage is achieved. The simulation and preliminary experiment results all verify the effectiveness of the proposed control strategy.