The current assessment of the ecological benefit of reservoirs commonly lacks quantitative calculation of actual ecological distribution. The spatial distribution prediction of vegetation growth generally requires the concurrent application of multiple numerical models, which are complicated and involve numerous parameters. This research comprehensively utilized a surface hydrodynamics numerical model and machine learning method to construct a flow-based vegetation growth prediction model. The ecological impact during the storage period and regulation period after the completion of the Dongzhuang Water Conservancy Hub was calculated and analyzed. The results indicate that the regulation period after the construction of the reservoir has a minor impact on vegetation growth in comparison to pre-construction flow conditions in the normal flow years. The limited extent and magnitude of regional NDVI decline induced by the reservoir storage period will gradually recover after the beginning of reservoir regulation. The formulated vegetation growth prediction model can reflect the relevant influence processes on NDVI to a certain extent and alleviate the complexity of prediction.