Hybrid Time-Series Framework for Daily-Based PM_2.5 Forecasting

Abstract

The impact of fine particulate matter on health has captured attention worldwide. Many studies have proven that fine particulate matter harms the respiratory system and the cardiovascular system. To prevent people from being harmed, many scientific research studies on PM 2.5 prediction have been conducted in recent years. Accurate PM 2.5 forecasting can not only alert people to stay away from concentrated areas but also provide the government with environmental policies in the future. In this paper, we propose a hybrid time-series prediction framework for daily-based PM 2.5 forecasting. The proposed framework consists of three components: the autoencoder, the dilated convolutional neural network, and the gated recurrent unit. The experimental dataset with 76 monitoring stations from the Taiwan Environmental Protection Administration is applied for comparison of the baseline and the proposed models. The proposed model is not only for the specified city-/county-wide region but also for the particular monitoring station/site to predict PM 2.5 concentration. By considering air quality data, meteorological data, and geographical data simultaneously, the proposed model can increase the accuracy of PM 2.5 prediction. In addition, the proposed PM 2.5 forecasting model can learn the location-centric spatial features and the dailybased temporal features simultaneously. The experimental results show that the prediction accuracy of the proposed model is superior to those of the baseline models.INDEX TERMS autoencoder, dilated convolutional neural network, gated recurrent unit, PM 2.5 forecasting

Machine learning algorithms to forecast air quality: a survey

“…( 2018 ) apply a similar hybridisation in which an MLP is also combined with a CNN and a spatio-temporal LSTM. A hybrid model which combines LSTM, GRU and CNN is developed by Chiang and Horng ( 2021 ) to forecast daily PM 2.5 in Taiwan.…”

Generalizable End-to-End Deep Learning Frameworks for Real-Time Attitude Estimation Using 6dof Inertial Measurement Units}

“…This layer can be used for object recognition, segmentation, and classification by learning patterns from images or videos using multiple layers with different parameters and associated weights. Recent advances in deep learning have shown that CNNs can also be used for time series prediction [79,80] and nonlinear regression [81].…”

Attentive Multi-Task Prediction of Atmospheric Particulate Matter: Effect of the COVID-19 Pandemic

“…In addition, predictive models for PM 10 and PM 2.5 based on neural ensemble techniques were developed [18]. Chiang and Horng proposed a hybrid time-series prediction framework including an autoencoder, dilated CNN, and GRU to predict PM 2.5 [19].…”