Abstract: With an aim to predict current control, this paper proposes a two-vector model to solve the problem of large current fluctuation and torque ripple in the traditional predictive current control model of doubly-fed wind power generation systems. This model performs voltage vector selection twice in one sampling period. The optimal vector of the first stage is used as the initial value of the second stage, and, combined with the action time of the candidate voltage vector, the minimum cost function is calculated. As a result, the range of the candidate voltage vector is expanded and the voltage vector selection is more accurate. The simulation results show that the proposed strategy can achieve effective control of the doubly-fed wind power generation system. At the same time, compared with the vector control and the traditional predictive current control model, the strategy effectively reduces current fluctuation and electromagnetic torque ripple.