Active Disturbance Rejection Decoupling Control for Three-Degree-of- Freedom Six-Pole Active Magnetic Bearing Based on BP Neural Network
Shaoshuai Wang, Huangqiu Zhu, Mengyao Wu, Weiyu Zhang
Abstract
In order to decouple the three-degree-of-freedom six-pole active magnetic bearing (3-DOF 6-pole AMB) with strong couplings, nonlinear and unstable disturbance, an active disturbance rejection control strategy based on BP neural network (ADRC-BP) is put forward. Firstly, the configuration, magnetic circuit, working principle and mathematical model of 3-DOF 6-pole AMB are introduced and established. Then, taking the 3-DOF 6-pole AMB as the controlled object, an ADRC-BP system is designed and a method of parameter tuning is proposed. Finally, the effectiveness of the control strategy is verified by simulation and experiments. The results show that the ADRC-BP has strong robustness and adaptability to the uncertainty of magnetic bearing model and the variation of external disturbance. The decoupling control effect of ADRC-BP is better than that of conventional ADRC.