A Hybrid CNN-LSTM Model for Forecasting Particulate Matter (PM2.5)

Taoying Li,Miao Hua,Xu Wu

doi:10.1109/access.2020.2971348

Taoying Li, Miao Hua + Show 1 more

Open Access

PDF Available

https://doi.org/10.1109/access.2020.2971348

Copy DOI

Export

Save

Cite

Journal: IEEE Access	Publication Date: Jan 1, 2020
Citations: 242	License type: CC BY 4.0

Affiliation: Dalian Maritime University

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

PM2.5 is one of the most important pollutants related to air quality, and the increase of its concentration will aggravate the threat to people’s health. Therefore, the prediction of surface PM2.5 concentration is of great significance to human health protection. In this study, A hybrid CNN-LSTM model is developed by combining the convolutional neural network (CNN) with the long short-term memory (LSTM) neural network for forecasting the next 24h PM2.5 concentration in Beijing, which makes full use of their advantages that CNN can effectively extract the features related to air quality and the LSTM can reflect the long term historical process of input time series data. The air quality data of the last 7days and the PM2.5 concentration of the next day are first set as the input and output of the model due to the periodicity, respectively. Subsequently four models namely univariate LSTM model, multivariate LSTM model, univariate CNN-LSTM model and multivariate CNN-LSTM model, are established for PM2.5 concentration prediction. Finally, mean absolute error (MAE) and root mean square error (RMSE) are employed to evaluate the performance of these models and results show that the proposed multivariate CNN-LSTM model performs the best results due to low error and short training time.

Highlights

In recent years, with the rapid development of China’s economy and industrialization, the problem of environmental pollution is becoming increasingly serious [1]
There are so many models for forecasting PM2.5 concentration, it is essential to test alternative models for identifying the best, so the univariate long short-term memory (LSTM) model, the univariate convolutional neural network (CNN)-LSTM model, the multivariate LSTM model and the multivariate CNN-LSTM model are compared
A hybrid CNN-LSTM deep learning network is proposed based on convolutional neural network and recurrent neural network for predicting the PM2.5 concentration of Beijing

Summary

INTRODUCTION

With the rapid development of China’s economy and industrialization, the problem of environmental pollution is becoming increasingly serious [1]. Duan et al [23] proposed a deep hybrid neural network improved by greedy algorithm for urban traffic flow prediction with taxi GPS trace. These studies mentioned above show that deep learning is a promising approach and some researchers have already applied it to study the air quality. Due to the complexity of PM2.5 formation, the high accuracy and efficiency demand for prediction, and the difficulty of deep learning network model in stability, it is essential to develop more effective model for forecasting PM2.5 concentration.

MATHODOLOGIES

RESULTS AND FINDINGS

CONCLUSION