Automatic river flow reorganization method and system based on deep learning

Abstract

The invention discloses an automatic river flow reorganization method and system based on deep learning. The method comprises the steps of selecting an input factor-flow sample with a preset time period length from a historical hydrological big data sample library; dividing the selected input factor-flow sample into a training sample, a verification sample and a test sample; constructing a deep learning reorganization model through the training sample, verifying hyper-parameters of the sample adjustment model, and testing the final evaluation model precision of the sample; and based on the trained deep learning reorganization model, inputting the input factors into the model, and performing real-time calculation to obtain the flow. According to the method, on the basis of fully utilizing hydrological duration time series big data samples with good consistency, reliability and continuity and carrying out hydrological station flow composition physical cause analysis, a deep learning method is utilized to establish a real-time flow calculation model of the station, and automatic flow reorganization is realized; the method does not need any facility equipment, and has the advantages ofhigh automation degree, strong timeliness, high precision and the like.

Description

technical field [0001] The invention relates to the technical field of data compilation, in particular to a method and system for automatic compilation of river flow based on deep learning. Background technique [0002] River flow is one of the most important hydrological elements, usually, the flow can be indirectly converted through the relationship between water level and water level flow; therefore, if figure 1 The stable level-flow relationship shown is an important factor in obtaining flow data quickly and accurately. Affected by complex water flow conditions (such as flood fluctuations, downstream jacking, etc.) and human activities (such as the construction of various wading structures in the test river section), the relationship between water level and flow has become very complicated and unstable, such as figure 2 shown. In order to control the flow change process and reorganization and alignment, hydrological surveyors need to carry out a large number of field ...

AI Technical Summary

Problems solved by technology

[0005] The above three types of methods all have certain shortcomings or limitations: the traditional reorganization method is the benchmark method of other methods, but this method has low timeliness, heavy field workload, and low degree of automation; the online reorganization method relies on the online flow measurement method, There are risks such as large investment and the accuracy being affected by many factors, so it cannot be put into production; the formula method integration method needs to formulate various push-flow formulas in advance, which has the disadvantage of large application limitations

Images