Active disturbance rejection generalized predictive control for magnetic levitation system

Abstract:

Considering the problem that the load disturbance and track “step” phenomenon may lead to instability of maglev trains, the active disturbance rejection generalized predictive control (LADRC-GPC) theory was introduced into the levitation system of maglev train, and a new suspension controller was designed. The controller adopted a layered control strategy, and the extended state observer (ESO) was used to dynamically compensate the system in the inner layer. The controlled autoregressive moving average model (CARMA) of the controlled object was obtained to reduce the dependence on the mathematical model of the controlled object. The inner layer was taken as the controlled object by the outer layer, and the generalized predictive control (GPC) was used to dynamically optimize the control system, which improved the tracking performance of the controller. By comparing with PID control algorithm and linear active disturbance rejection control (LADRC) algorithm through simulation and experiment, results show that the LADRC-GPC algorithm had better tracking performance and robustness, and could maintain small error under large load disturbance.