PM2.5 concentration space-time change prediction method and system based on space-time diagram neural network

Abstract

The invention relates to a PM2.5 concentration spatio-temporal change prediction method and system based on a spatio-temporal diagram neural network. The method comprises steps of acquiring atmospheric pollutant concentration monitoring historical data of each atmospheric monitoring station, meteorological data of a national meteorological station, predicted meteorological data and elevation data; constructing a sample, and constructing an adjacent matrix M and a weight matrix W by using the longitude and latitude of each atmosphere monitoring station; and constructing a neural network model based on the space-time diagram to predict a prediction result of each station, and obtaining a corresponding PM2.5 concentration prediction value. According to the method, observation data of about 1500 atmospheric monitoring stations in the whole country is taken as a training set, various data sources such as meteorology and elevation are combined, the neural network based on the space-time diagram is used, a unified prediction framework is constructed, the PM2.5 concentration change in a large area can be predicted at the same time, and the prediction precision is improved.

Description

technical field [0001] The present invention relates to the technical field of meteorological forecasting, in particular to a method and system for predicting temporal and spatial changes in PM2.5 concentration based on temporal and spatial graph neural networks. Background technique [0002] Air pollution is one of the major environmental issues affecting health. Accurate short-term prediction of air pollution can provide a basis for government decision-making, make timely preventive measures, and reduce the occurrence of pollution incidents. [0003] Existing PM2.5 prediction methods are mainly divided into mechanism models and data-driven methods. The mechanism model can predict by simulating known physical laws and inputting the required historical observation data to initialize the model. However, due to the fact that the mechanism process of pollutant formation has not been fully ascertained and is also limited by computing resources, it cannot be fully simulated. , ...

AI Technical Summary

Problems solved by technology

Therefore, most of the existing methods only consider the temporal relationship but not the spatial relationship, and the prediction accuracy needs to be improved.

Images