[Objective] This paper combined linearized feedback control with adaptive global integral sliding mode control (AGISMC) to optimize the control performance and the control accuracy of the system based on the nonlinear characteristics, uncertainty and susceptibility to external perturbations of the magnetic suspension ball system. [Methods] First, a mathematical model was established using linearized feedback, and adaptive control was employed to achieve real-time estimation of the system parameters to reduce the uncertainty. Second, a global function was introduced to improve the transient response of the system, and the sign function in the sliding mode function was replaced with a saturation function to reduce system chattering. Third, for uncertain disturbances, an extended state observer (ESO) was utilized to estimate unknown disturbance signals in real time. Finally, simulations were conducted in Matlab/Simulink, and an experimental platform was built for validation. [Results] Simulation and experimental