In order to accurately predict the water richness of the weathered bedrock aquifer, 28 groups of weathered bedrock pumping test borehole data in Zhangjimao minefield were used as training and verification samples, and the lithology combination index, weathering index, thickness, core recovery rate and burial depth of the weathered bedrock were selected as evaluation indexes. Based on whale optimization algorithm-support vector machines (WOA-SVM), a water-rich identification model for weathering bedrock aquifers was proposed. This model can predict the water-rich grade of the weathered bedrock in the area without pumping test data, and realize water-rich zoning of the weathered bedrock in the well field by comprehensive use of the geological information of 249 exploration boreholes. The study shows that the weathered bedrock of Zhangjiamao minefield is weakly water-rich as a whole, and its spatial distribution is uneven. There are strong water-rich areas in the central part of the field and the local area along Wulanbula Gully, but their distribution range is small, there are some moderately water-rich areas in the central-western and southeastern parts, and the northeastern and southwestern areas are weakly and very weakly water-rich almost all the time. The results predicted are more in line with the actual situation, and the research results can provide a reference for the safe production of the mine and a new way of thinking for the prediction of the water-richness of the weathered bedrock.