Design of an Rbf-Pid Controller for Magnetic Levitation Vehicle

To address the open-loop instability and pronounced nonlinearity in the suspension control of a magnetic levitation (maglev) vehicle, a suspension control system based on a radial basis function (RBF) neural network was designed. First, a theoretical model of the maglev vehicle was developed using parameters from a single-suspension experimental platform. Based on this model, a dual-loop control system, consisting of a current loop and a gap loop, was implemented to achieve stable suspension of the single-suspension frame. To overcome the limitations of traditional PID control in suspension systems, an RBF neural network was introduced to optimize the PID parameters of the gap controller in real-time, enabling adaptive control. Simulink simulations show that, compared to the conventional PID controller, the RBF-PID controller yields a smoother suspension process and significantly reduces steadystate error, thus confirming the effectiveness of the proposed approach.