A Novel Method for Regional NO₂ Concentration Prediction Using Discrete Wavelet Transform and an LSTM Network

Abstract

Achieving accurate predictions of urban NO2 concentration is essential for effectively control of air pollution. This paper selected the concentration of NO2 in Tianjin as the research object, concentrating predicting model based on Discrete Wavelet Transform and Long- and Short-Term Memory network (DWT-LSTM) for predicting daily average NO2 concentration. Five major atmospheric pollutants, key meteorological data, and historical data were selected as the input indexes, realizing the effective prediction of NO2 concentration in the next day. Firstly, the input data were decomposed by Discrete Wavelet Transform to increase the data dimension. Furthermore, the LSTM network model was used to learn the features of the decomposed data. Ultimately, Support Vector Regression (SVR), Gated Regression Unit (GRU), and single LSTM model were selected as comparison models, and each performance was evaluated by the Mean Absolute Percentage Error (MAPE). The results show that the DWT-LSTM model constructed in this paper can improve the accuracy and generalization ability of data mining by decomposing the input data into multiple components. Compared with the other three methods, the model structure is more suitable for predicting NO2 concentration in Tianjin.

Prediction of Geological Parameters during Tunneling by Time Series Analysis on In Situ Data

“…With the increasing demand for sequence analysis and the continuous development of machine learning, new methods are constantly being proposed for time sequence analysis. Among them, the LSTM, which was developed based on the ANN and recurrent neural network (RNN), is a particular type of deep learning neural network [ 29 ]. Van Houdt et al [ 30 ] presented a comprehensive review that covers the formulation and training of the LSTM, the relevant applications reported in the literature, and code resources implementing this method for a toy example.…”

A Deep Learning Approach With Wavelets to Forecasting Tropospheric Ozone in a Tropical Metropolitan Region

“…O3 hourly concentration series is nonstationary as it is affected by other air pollutants and meteorological parameters levels variation. A wavelet transform is very suitable for dealing with nonstationary time series including air pollutant data [10]. Air pollution time series generally have short-time transient components.…”

“…They reported results with higher prediction accuracy and stability than other models used for comparison. In [10], the work features a predicting model based on discrete WL transform (DWT) and LSTM network for predict the next day nitrogen dioxide (NO2) concentration in the city of Tianjin-China. Their results show improvement in model accuracy and generalization.…”

ProxySense: A novel approach for gas concentration estimation using Long Short-Term Memory Recurrent Neural Network (LSTM-RNN)

“…Several methods have been proposed for gas concentration estimation in environmental monitoring networks. Liu et al proposed a concentration prediction model based on Discrete Wavelet Transform and Long Short-Term Memory network (DWT-LSTM) for predicting daily average NO 2 concentration [8]. The input to the model consist of different atmospheric pollutants, meteorological data and historic NO 2 data, to realise the prediction of NO 2 concentration for the next day.…”