Abstract: Due to the multi-magnet coupling problem in the suspension frame of maglev train, the coupling effect during operation will cause the unstable suspension of the suspension frame. In order to ensure the stable operation of the multi-magnet module in the suspension frame under external interference, a deep reinforcement learning collaborative control method for the multi-magnet system of the suspension frame of the electromagnetic suspension (EMS) maglev train is proposed. Firstly, considering the coupling, the dynamic modeling of the multi-electromagnet module in the maglev train suspension frame is carried out and its coupling is analyzed. Secondly, a cooperative control method (SAC-CC) of multi-electromagnet system based on SAC algorithm is proposed, and a framework of deep reinforcement learning cooperative control algorithm is constructed. The dynamic model of multi-electromagnet module of suspension frame is transformed into a deep reinforcement learning environment model and a reward function is designed for this model. Then, the SAC-CC controller is obtained by training in a static floating environment, and the control performance of the controller under different reward functions and the stability of the controller under different floating positions are analyzed. Finally, the SAC-CC controller is applied to the suspension control and cooperative control of the multi-electromagnet system under different working conditions. The effectiveness and robustness of the proposed controller are verified by comparing with the traditional proportional-integral-derivative (PID) control method. The results show that under different working conditions, compared with the PID controller, the SAC-CC controller proposed in this paper can not only effectively control the multi-magnet module in the suspension frame to suspend stably near the equilibrium position, but also significantly reduce the coupling between the electromagnet modules. It has better suspension control performance and cooperative control performance. The suspension control performance and cooperative control performance of the SAC-CC controller under different working conditions are increased by 30%-99% and 30%-75%, respectively.