Spatial-temporal data model based city-level PM2.5 concentration prediction method

Abstract

The invention belongs to the application field of mutual cross combination of multiple subjects and technologies, and specifically relates to a spatial-temporal data model based city-level PM2.5 concentration prediction method. The method comprises the steps of firstly acquiring city-level PM2.5 concentration data by using a developed online data acquisition system; determining an optimal model prediction factor by comprehensively considering the model interpretability, the data accessibility and the model complexity, and thus determining the model type and the model structure; then determining an interpolation algorithm for the missing PM2.5 concentration data according to missing conditions of the actually acquired PM2.5 concentration data and the determined model structure; and finally determining an algorithm capable of performing real-time online prediction according to the model prediction precision. According to the invention, the model structure can be determined by making full use of readily available data, and thus an effective city-level PM2.5 concentration prediction model is built.

Description

technical field [0001] The invention belongs to the application field of cross-combination of multiple disciplines and technologies, and in particular relates to a city-level PM2.5 concentration prediction method based on a spatio-temporal data model. Background technique [0002] PM2.5 is a general term for solid particles or droplets with a diameter less than or equal to 2.5 microns in the air. PM2.5 can reduce the visibility of the air, leading to the appearance of haze days; PM2.5 enters the human lungs through the respiratory tract, increasing the incidence of respiratory diseases and cardiovascular diseases. In order to control the harm of PM2.5 to the environment and human health, it is necessary to know the current and future PM2.5 concentration in real time. Therefore, real-time prediction of PM2.5 concentration has important practical significance. [0003] Due to the inability to obtain historical data of PM2.5 concentration values, an Internet-based online PM2....

AI Technical Summary

Problems solved by technology

[0004] The above-mentioned interpolation algorithm for missing PM2.5 concentration has deficiencies: the reverse distance-weighted interpolation is suitable for interpolation calculation of a single missing data, but the actual collected data has continuous missing conditions, which limits the application of reverse distance-weighted interpolation; land Using a regression model using land use type and population density as predictors ignores the temporal and spatial characteristics of PM2.5 concentrations, making interpolation less effective

Therefore, under the conditions that it is very difficult to obtain urban-level PM2.5 concentration data sets, there are few data collection sites, and the accuracy of data sets obtained by remote sensing technology is not high, how to choose appropriate predictors becomes the key to predicting PM2.5 concentration

[0006] The above models for predicting PM2.5 concentration have deficiencies: Barman S C, Singh R and others first used the humidity in the meteorological variable as the predictor of the model, and the predictor of the model is too single; considering that the PM2.5 concentration is affected by the PM2.5 collection site Influenced by the geographical environment, Zou B, Luo Y and others used land use type and population density as the predictors of the model. The prediction accuracy of the model is not high, and the required data are difficult to obtain; Advances in remote sensing technology have been widely used. Song W and Jia H used the aerosol optical depth acquired by satellites to predict PM2.5, and the prediction effect has been significantly improved. However, remote sensing technology requires high image acquisition technology. The accuracy of the acquired data needs to be improved

Images