Decoupling Control of Six-Pole Radial Hybrid Magnetic Bearing Based on Least Square Support Vector Machine Inverse System

In order to solve the problems of the influence of nonlinearity and strong coupling on the control performance of six-pole radial hybrid magnetic bearing (HMB) system, a decoupling control strategy based on least square support vector machine (LSSVM) inverse system is proposed. The structure and operation principle of six-pole radial HMB are introduced, the mathematical model of its suspension force is deduced, and the reversibility of mathematical model is proved. The regression principle of LSSVM is used to approximate the inverse system of the six-pole radial HMB, and the inverse system is connected in series with the original system to form a pseudolinear system, which realizes the linearization and decoupling of six-pole radial HMB system. For the problems of difficulty in tuning the regularization parameter and kernel width of LSSVM, the improved quantum particle swarm optimization (IQPSO) algorithm is proposed to optimize the above parameters. The decoupling simulation system of six-pole radial HMB system is constructed, and the rotor floating, anti-interference, and decoupling performance experiments are conducted to verify the feasibility and reliability of the proposed decoupling control method.