As an engineering machine which loads and unloads gravel frequently, wheel loaders own large potential energy when the unloading process. In order to recover and utilize the potential energy, a closed pumpcontrolled threechamber hydraulic cylinder lifting device was proposed. The servo motor and accumulator supplied power for the device. A potential energy recovery chamber was added to the original boom hydraulic cylinder to form a threechamber hydraulic cylinder. The potential energy was recovered and utilized by an accumulator connected with the added chamber when the wheel loader dropped. The closed hydraulic system composed of the fixed displacement pump and the threechamber hydraulic cylinder was used to drive the lifting device, and the servo motor was used to drive the pump. In order to accurately control the extension and retraction of the hydraulic cylinder, the displacement feedback of the hydraulic cylinder was adopted to closedloop control of the motor speed. In the research, the mathematical model and the transfer function of closed pumpcontrolled threechamber hydraulic cylinder lifting device were established. Firstly, the static characteristics of the device were analyzed, and the corresponding control strategies were put forward. Then, according to the working principle and control strategies of the proposed device, a multidisciplinary electromechanicalhydraulic simulation model was constructed, and its feasibility, dynamic and energy consumption characteristics were analyzed. Finally, a test system of the proposed device was constructed to verify its dynamic and energy consumption characteristics. Based on the simulation and experimental results, the energy consumption of the proposed device was reduced by 21.2% in one working cycle under halfload condition comparing with the closed pumpcontrolled twochamber hydraulic cylinder lifting device. Compared with the original valvecontrolled hydraulic lifting device, the energy consumption of the proposed device under noload, halfload and fullload conditions were decreased by 22.7%, 20.9% and 21.5%, respectively.