GCN-LSTM-based sea temperature space-time prediction method

By adopting the GCN-LSTM method in the SST space-time prediction model, integrating topological maps and implicit maps, combining self-attention mechanisms and LSTM models, the problem of insufficient fusion of spatiotemporal features in the existing technology is solved, and a higher sea Temperature prediction accuracy and reliability.

CN119180302AActive Publication Date: 2024-12-24HEBEI NORMAL UNIVERSITY OF SCIENCE & TECHNOLOGY
6 Cites 0 Cited by

Patent Information

Application Number
CN202411277188.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-12
Publication Date
2024-12-24
Estimated Expiration
2044-09-12

AI Technical Summary

Technical Problem

The existing SST spatiotemporal prediction model fails to effectively integrate spatiotemporal characteristics and dependencies, neglecting the correlation between implicit influencing factors in space and non-neighboring nodes, resulting in insufficient prediction accuracy and reliability.

Method used

Using the SST space-time prediction method based on GCN-LSTM, the space-time prediction method of SST data is learned by constructing dynamic graph modules, time-dependent modules and space-time fusion modules, and the topological graphs, time-dependent implicit graphs and functional correlation implicit graphs. and time characteristics, combined with self-attention mechanism and LSTM model, realize spatiotemporal prediction of sea temperature.

Benefits of technology

It improves the accuracy and reliability of SST prediction, can better capture the dynamic changes characteristics and spatial dependence of SST data, and shows better results in SST space-time prediction tasks.

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

Abstract

The invention discloses a GCN-LSTM-based sea temperature spatio-temporal prediction method, which belongs to the technical field of sea temperature-oriented spatio-temporal prediction, and comprises the following steps: selecting a sea temperature spatio-temporal data set, and defining a sea temperature spatio-temporal data prediction task; a sea temperature space-time prediction model based on GCN-LSTM is constructed; the space-time prediction model STFTIG of the sea temperature based on the GCN-LSTM comprises three modules: a dynamic graph module DGM, a time dependence module TDM and a space-time fusion module STFM; the space-time prediction model STFTIG of the sea temperature is constructed based on a graph convolutional network and a long short-term memory network. The accuracy and reliability of sea temperature prediction are improved by fusing the spatial-temporal characteristics of the data.
Need to check novelty before this filing date? Find Prior Art