The visual modeling technology of 3D digital geological model, as a component of CAD technology, is also a hot content of current geoscientific development research. This paper is based on the GOCAD platform, and at the same time constructs the corresponding spatial database, and establishes a three-dimensional geological model of the mining area on this basis, and uses the kriging algorithm to derive the grade trend prediction model in the area. The X mining area is taken as the study area to carry out the 3D mineralization prediction research, through the quantitative extraction of the stratigraphy, rock bodies and other mineralization factors in the area, the introduction of the weight of evidence method to construct the mineral search model, and the use of the ROC curve to evaluate the performance of the model. Through the study, four mineral searching target areas were circled in the X mining area, and the a posteriori probability of mineralization can be divided into four levels: high, higher, lower, and low, and the AUC value of the evidence weight model is 0.94, which is a better prediction performance, and the probability of finding minerals is about 78.2%.
