In order to maximize the propulsion efficiency of flapping hydrofoil, a new method is proposed in this paper. The effects of heave amplitude, pitch amplitude, and the phase difference between heave and surge on propulsion performance were analyzed by numerical calculation, and it was found that three motion parameters had interactive effects on flapping hydrofoil propulsion performance. BP neural network was used to fit the three motion parameters and propulsion performance. Using this function model, the optimal motion parameters can be obtained under certain thrust. In this study, the optimization matching under certain thrust was carried out by using this method, and the propulsion efficiency was improved by 7.73%.