Abstract: The integration of a high proportion of renewable energy into the grid amplifies the uncertainty in supply and demand of the power system, which requires an more targeted automatic generation control (AGC) methods to meet different FM needs. In this paper, the wind power system is taken as the research object. The system control process based on model predictive control (MPC) is first analyzed by constructing the AGC model of interconnected power grid. Then, to address the issue of high computational complexity of traditional online MPCS, a conditional trigger strategy is introduced to reduce the frequency of optimal computation. Trigger rules and thresholds are designed to dynamically adjust the different frequency modulation requirements of various states within the system, so as to achieve optimal trigger and control effects. Finally, the simulation results demonstrate that the proposed method can effectively reduce the online computing complexity of the system while ensuring the control performance at the same time, and it exhibits better system stability in respond to various frequency modulation requirements.