Abstract: Aiming at the issue of inconsistent state of charge (SOC) among lithium battery modules after long-term charging and discharging, traditional centralized balancing circuits suffer from the drawback of low balancing speed. To address this, a balancing control strategy based on continuous control set model predictive control (CCS-MPC) is proposed, utilizing a symmetrical switch array, boost converter, and LC quasi-resonant circuit as the main balancing circuit. Firstly, the balancing system is modeled, and a discrete state-space equation is constructed. Subsequently, a multi-step model predictive algorithm is designed based on the state equation, with the value function defined by the error between the SOC predicted value and the reference value, as well as the difference between the current input and the previous input of the converter’s switching elements. Finally, a quadratic programming is applied to the value function to obtain an optimal control solution online, which is then implemented in the balancing system. By dynamically adjusting the duty cycle, the magnitude of the balancing current is controlled. Compared with single-step prediction, multi-step prediction requires considering the optimality of the controlled variables over multiple cycles, ensuring that the balancer can output the optimal balancing current in each balancing cycle and effectively preventing instability of the balancer. The simulation results show that compared with the conventional PI algorithm, the proposed model prediction algorithm achieves SOC consistency among battery modules, ensures stable output of balancing current, and reduces the balancing time by 17%.