The vehicle–guideway coupled self-excited vibration of maglev systems is a common control instability problem in maglev traffic while the train is suspended above flexible girders, and it seriously affects the suspension stability of maglev vehicles. In order to solve this problem, a nonlinear dynamic model of a single-point maglev system with elastic track is established in this paper, and a new and more stable adaptive backstepping control method combined with magnetic flux feedback is designed. In order to verify the control effect of the designed control method, a maglev vehicle–guideway coupled experimental platform with elastic track is built, and experimental verifications under rigid and elastic conditions are carried out. The results show that, compared with the state feedback controller based on the feedback linearization controller, the adaptive backstepping control law proposed in this paper can achieve stable suspension under extremely low track stiffness, and that it shows good stability and anti-interference abilities under elastic conditions. This work has an important meaning regarding its potential to benefit the advancement of commercial maglev lines, which may significantly enhance the stability of the maglev system and reduce the cost of guideway construction.
In electrodynamic suspension (EDS) maglev systems, the time-varying magnetic fields experienced by onboard superconducting magnets can induce excessive AC losses and elevate the risk of quench, posing a critical threat to operational reliability. Conventional three-dimensional finite element analysis, though accurate, requires hours per frequency point and cannot support real-time quench assessment or parametric design optimization. This paper proposes an efficient magnetic shielding analysis method based on a reflection–absorption decoupled lumped-parameter framework. The three-dimensional eddy current field is reduced to a low-order coupled system by decomposing the eddy currents into plate-surface modes, a global circulation loop, and side-wall contributions, all coupled through a unified series magnetic circuit, while the field penetration through the cavity wall is captured via a finite impulse response (FIR) diffusion kernel with a high-frequency correction. The proposed method achieves a simulation time of 3–5 s per frequency point, a speedup of approximately ∼ over finite element analysis, with a root-mean-square shielding effectiveness deviation of 2.03 dB over 10–3000 Hz. The efficiency of the proposed method enables co-optimization of the outer dewar and superconducting magnets, offering a practical analysis tool for quench mitigation and electromagnetic compatibility design in EDS maglev systems.