Cascade power station hydraulic relation coupling method based on deep learning theory

The invention discloses a cascade power station hydraulic relation coupling method based on a deep learning theory, and relates to the technical field of hydropower operation, and the method comprises the following steps: S1, selecting an input factor based on a conditional mutual information theory in mutual information; S2, constructing a first model, and performing data processing on the input factors through the first model to obtain a deep learning network model of the upstream ex-warehouse flow and the downstream in-warehouse flow; S3, determining a hydraulic coupling relationship between the upstream and downstream cascade power stations according to the deep learning network model of the upstream ex-warehouse flow and the downstream in-warehouse flow, and determining the in-warehouse flow of the downstream power station based on the hydraulic coupling relationship between the upstream and downstream cascade power stations; through the design, coupling calculation of the hydraulic relation of upstream