In order to further improve the utilization of the core experimental parameters, to enrich oil recovery prediction methods and to further improve the prediction precision, a one-dimensional mathematical model of water flooding was established by utilizing the advantage of various types of core experimental methods, numerical simulation and reservoir engineering approach. The model is based on the one-dimensional reservoir physics experiment parameters, considering effect of the threshold pressure during water flooding.Moreover, using liquid production rate as weight coefficient, the one-dimensional mathematical model is transformed into a two-dimensional cross-sectional model; on this basis, using water flooding sweep coefficient under different well patterns, the two-dimensional cross-sectional model can be transformed into a three-dimensional model to predict oil recovery. The results in Mo005 well block showed that: the predicted recovery using this method is 25.6%, consistent with 25% of the actual standardized recovery. The method can fully use various types of core results, consider interlayer heterogeneity and well pattern, avoiding the singularity and uncertainties by using simple experimental methods or reservoir engineering methods to predict oil recovery. The method can further enrich oil recovery prediction methods and improve prediction precision, at the same time, it provides a new direction and idea for other researchers.