sliding mode prediction;  Approximation law;  jitter;  Active magnetic bearings

Aiming at the jitter problem caused by conventional sliding mode variable structure control methods in electromagnetic bearing control, a power exponential approximation law that can effectively weaken system jitter and ensure system asymptotic stability is proposed, and the approximation law is used as the sliding mode reference trajectory, and the name model of uncertain system is used as the sliding mode prediction model, and the idea of rolling optimization and feedback correction in predictive control is introduced into the design of discrete quasi-sliding mode variable structure control system. Rolling optimization solution is realized. The electromagnetic bearing dynamics model is established and simulated, and the results show that the scheme can effectively reduce the adverse effects of jitter on the system, realize the rapid adjustment and stable suspension of the rotor, and its tracking performance is fast and accurate, and it has strong robustness to the change of system parameters and resistance disturbance.