Adaptive levitation control for characteristic model of low speed maglev vehicle

Abstract

Due to the strong coupling and large disturbance between multiple electromagnetic coils, the previous dynamic model of low-speed maglev train cannot accurately model the complex system when it is levitated. Therefore, based on the general linear transfer function, the concept of feature modeling is introduced to describe the dynamic characteristics, environmental characteristics and control performance requirements of the system by defining feature variables. On the premise of satisfying a certain sampling frequency, the original dynamic model is approximately linearized, which is equivalent to a second-order difference equation, i.e. a system with slow time-varying parameters. On this basis, the design and implementation of the controller of the magnetic levitation system are discussed. First, by studying the application of PID controller in suspension system, the difficulty of parameter debugging of PID controller and the insufficiency of restraining coupling interference are pointed out. The generalized predictive control (GPC) is proposed and improved to improve the robustness and anti-jamming of the system. Finally, a large number of simulations and experiments show that the controller can effectively suppress the coupling interference. Compared with single-point levitation controller, it has higher application value in engineering practice.