CNN-based land-water integrated river water quality space-time continuous prediction method

Abstract

The invention discloses a CNN-based land-water integrated river water quality space-time continuous prediction method, which comprises the following steps: carrying out runoff simulation on each sub-basin of a river basin based on an SWAT model to obtain a runoff simulation result of each sub-basin; based on the EFDC model, inputting the runoff result of each sub-basin as the flow boundary of the EFDC model, inputting the pollutant flux as the water quality boundary of the EFDC model, and simulating to obtain a water quality space-time continuous distribution image; constructing a CNN model, inputting the obtained water quality space-time continuous distribution image as a training sample, and training the CNN model; and utilizing the trained CNN model to carry out river water quality space-time continuous prediction, and predicting a space-time continuous distribution water quality result for an input discrete section water quality result. Omnibearing time-space continuous prediction of basin water quality by an artificial intelligence technology is realized, and powerful technical support can be provided for water environment supervision.

Description

technical field [0001] The invention relates to the technical field of river water quality prediction, in particular to a CNN-based land-water integrated river water quality time-space continuous prediction method. Background technique [0002] River water quality prediction is an important part of river basin water environment management. The main measures to solve water quality problems include water quality monitoring, water environment supervision and water quality prediction. In recent years, with the promulgation of policies related to water pollution control in my country and the rapid development of ecological environment big data, the ability to monitor and supervise pollution sources and water quality conditions in river basins has been gradually enhanced, and the occurrence of water quality problems in river basins has been greatly slowed down. [0003] At the same time, the spatiotemporal continuous prediction of river water quality has become an indispensable he...

AI Technical Summary

Problems solved by technology

[0004] Among them, the time-space continuous prediction of river water quality in land-water integration is an important decision support technology for water environment management and control. However, the prediction results of the existing prediction methods for river water quality are discrete and independent sections in space, and cannot represent the flow of the main stream of the river. Water quality status, which limits regulatory capacity for water pollution control

Images