Prediction method and device for air quality, computer device and readable storage medium

Abstract

The invention provides a prediction method and a device for air quality, a computer device and a readable storage medium. The prediction method for air quality comprises steps of obtaining historicalpollutant data at a pollutant collection spot and historical meteorological data at a meteorological collection spot; weighting the historical pollutant data according to location information of the pollutant collection spot and the meteorological collection spot to obtain historical pollutant mapping data at the meteorological collection spot; training the historical pollutant mapping data and the historical meteorological data as a training set with an extreme gradient ascent model to obtain a prediction model; using the prediction model to obtain predicted air quality data based on the current pollutant data and the current meteorological data. The prediction method for air quality proposed by the invention does not need to preprocess missing data, so that the whole deep learning process is faster, and the historical pollutant data collected at the pollutant collection spot is mapped to the meteorological collection spot, so as to obtain more accurate sample data, improving prediction accuracy.

Description

technical field [0001] The present invention relates to the field of environmental science, in particular to an air quality prediction method, an air quality prediction device, a computer device, and a computer-readable storage medium. Background technique [0002] The traditional EMOS (Ensemble Model Output Statistics) model and ARMA (Auto-Regressive Moving Average) model based on feature engineering are used for meteorological prediction of SO 2 Concentration prediction tasks can achieve better results, but these methods and models require researchers to spend a lot of effort on hand_crafted feature extraction, model building, and parameter adjustment. The deep learning Conv_LSTM model can capture the spatial and temporal relationship very well. When monitoring stations acquire air quality data, there are usually many missing values ​​in the air quality data due to equipment failure or network delay delays. In data preprocessing, especially missing value processing, most ...

AI Technical Summary

Problems solved by technology

[0003] The present invention aims to at least solve the technical problem in the prior art that if data is missing in the air quality predic

Images