Open AccessAtmospheric Pollutant Prediction Based on Wavelet Decomposition and Long Short-Term Memory Network
Author(s) -
Lei Li,
HE Zhe-xiang
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/768/7/072059
Subject(s) - pollutant , mean squared error , term (time) , environmental science , meteorology , wavelet , decomposition , statistics , mathematics , computer science , artificial intelligence , geography , chemistry , physics , organic chemistry , quantum mechanics
To solve the problem that current atmospheric pollutant prediction research only pays attention to one pollutant type, a Long Short-term Memory Network (LSTM) atmospheric pollutant prediction model based on Wavelet Decomposition (WD) is proposed, predicting daily average PM 2.5 , PM 10 , SO 2 , NO 2 and O 3 concentration for the next day. Based on data collected from the national control station (NO.1338A) in Changsha from 2015 to 2018, the model was verified. The results show that for the prediction of atmospheric pollutants, compared with the LSTM prediction model established directly with the original atmospheric pollutants, the root mean square error and the absolute average error of the hybrid model are lower, and the accuracy of the level prediction is higher.