The invention discloses a three-dimensional geologic model modeling method, which comprises the following steps of: acquiring drilling data of a plurality of drilling position points of a research area, performing resampling and normalization processing on the drilling data, and predicting the research area by utilizing a trained stacking machine learning model to obtain a three-dimensional geologic model. And acquiring a geological data database of the research area, performing precision evaluation on the three-dimensional geological model by using the thickness and occurrence of the rock stratum, and performing iterative correction on the three-dimensional geological model. According to the method, the stacking machine learning algorithm is integrated, meanwhile, the geoscience knowledge obtained from the geological data database is fused, a model result meeting the geoscience knowledge requirement can be obtained through iterative correction, and the result shows that the stratigraphic classification effect of the integrated stacking machine learning algorithm is better than that of a single classifier; and the precision of the three-dimensional geologic model obtained after iterative correction of the geoscience knowledge database is higher, and the detail expression ability of the model is enhanced.