Multi-scale space-time diagram convolution blue-green algae forecasting method and device based on hydrodynamic perception

By introducing a multi-scale spatiotemporal map convolution network of hydrodynamic perception in the prediction of cyanobacteria outbreaks, combining the spatial distance and hydrodynamic factors of water quality monitoring stations, the problem of low prediction accuracy in the existing technology is solved, and more efficient cyanobacteria is achieved forecast.

CN119026085AActive Publication Date: 2024-11-26POWERCHINA HUADONG ENG CORP LTD
7 Cites 6 Cited by

Patent Information

Application Number
CN202411489223.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-24
Publication Date
2024-11-26
Estimated Expiration
2044-10-24

AI Technical Summary

Technical Problem

The existing cyanobacteria outbreak prediction methods have low prediction accuracy under complex hydrodynamic conditions and fail to fully consider the spatial structure and hydrodynamic characteristics of the water system, resulting in insufficient prediction accuracy.

Method used

Using the multi-scale spatiotemporal graph convolution network (MS-STGCN) method based on hydrodynamic perception, a multi-scale spatiotemporal convolution network is constructed by obtaining the spatial distance and hydrodynamic factors between water quality monitoring stations, combining the graph diameter and average space The distance and degree matrix is ​​characterized by enhanced to predict cyanobacteria outbreaks.

Benefits of technology

It significantly improves the accuracy of cyanobacteria forecasts, can more accurately capture complex space-time dependencies and water quality changes, and provides a more reliable scientific basis for water environment management.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a multi-scale space-time diagram convolution blue-green algae forecasting method and device based on hydrodynamic perception, and relates to the technical field of water environment prediction.The method comprises the steps that the spatial distance between water quality monitoring stations preset in a to-be-detected area is obtained, and an edge weight matrix is determined based on the spatial distance and preset adjustable scale parameters; setting a graph theory index set, performing feature enhancement processing on the original features based on the graph theory index set, and determining target input features; and constructing a multi-scale space-time convolution network based on the edge weight matrix, sending the target input features to the multi-scale space-time convolution network, carrying out model training processing on the multi-scale space-time convolution network, determining a target multi-scale space-time convolution network, and carrying out blue-green algae outbreak prediction by using the multi-scale space-time convolution network. The accuracy of blue-green algae forecasting can be remarkably improved.
Need to check novelty before this filing date? Find Prior Art