Watershed scale runoff multi-step space-time prediction method based on combination of static and dynamic graphs

A drainage basin scale runoff multi-step space-time prediction method based on combination of static and dynamic graphs comprises the steps that time and space feature extraction is carried out on runoff of runoff monitoring stations in a predicted drainage basin in a period of historical time, a static river channel distance graph, a static elevation difference graph and a dynamic graph are generated for all the runoff monitoring stations, the three graphs are fused, and a drainage basin scale runoff multi-step space-time prediction result is obtained. And establishing a runoff prediction model to predict runoff of all runoff monitoring stations. The runoff prediction model comprises an input layer, M levels of processing structures and a linear layer, and the input layer inputs a runoff sequence of each runoff monitoring station in a prediction drainage basin within a period of historical time; the processing structure of each level is subjected to time and space feature extraction to obtain the output of t