The height of the water-conducting fracture zones (WCFZ) is crucial for ensuring safe coal mining beneath aquifers, particularly considering the secondary development of the WCFZ in upper seams due to repeated mining in close distance coal seams. Accurately predicting this height is essential for mine safety, groundwater protection, and optimal coal resource use. This study compiles extensive measured data from various mining areas in China to analyze the coupling relationship between the WCFZ development height and six influencing factors: mining thickness, mining depth, coal seam spacing, hard rock lithology ratio, and the slope length of working face. Using principal component analysis and fuzzy comprehensive evaluation, we derived the combined weights of these factors. We developed a prediction model based on multiple regression analysis that incorporates the weighted influences of these factors, further refined into a multivariate nonlinear regression model for greater accuracy. Compared against a multivariate linear regression model, empirical formulas, and measured results, our model demonstrated higher accuracy, stability in both absolute and relative errors, and practical applicability. It was successfully applied and validated on the 100,501 working face of Nanyaotou Coal Mine in Shanxi Province, offering a new scientific approach for predicting the WCFZ.