To eliminate the deviation between the actual pressure and the ideal pressure of the wet clutch of a highpower tractor during the power shifting process, a model free adaptive predictive control strategy (MFAPC) based on the compactform dynamic linearization (CFDL) was proposed to achieve the tracking control of the pressure. Considering the parameter uncertainties of the hydraulic actuators and external disturbances, a nonlinear dynamic model of the wet clutch control system was established and verified in AMESim. Taking the cylinder pressure as the control target, the nonlinear wet clutch actuator system was equivalently converted into a dynamic linearized data model by CFDL. Furthermore, an online MFAPCbased controller was designed to control the clutch pressure in real time. The closedloop clutch pressure control system was built in Matlab/Simulink and the simulation experiments were performed to verify the correctness of the dynamic linearization model and the reliability of the control strategy. Compared with the commonly adopted control strategies such as PID and MFAC, the proposed control algorithm performed better in terms of robustness. MFAPC was proved to output the varying expected pressure accurately and quickly by adjusting the control parameters. Specifically, the response time under the square wave signal was 0.119s, while the steadystate error under sinusoidal signal kept at ±0.0281MPa, which appeared 48.91% lower than the conventional PID. In addition, MFAPC has better antiinterference capability than MFAC, which was reflected by the smallest jerk (16.57m/s3) among PID and MFAC. The results proved that MFAPC generally had better dynamic performance and superiority, and is conducive to improving the shifting quality and guaranteeing the power performance of highpower tractors under continuous operating conditions. 