Deep learning-based PM2.5 prediction considering the spatiotemporal correlations: A case study of Beijing, China

Pak, Unjin; Ma, Jun; Ryu, Unsok; Ryom, Kwangchol; Juhyok, U.; Pak, Kyongsok; Pak, Chanil

doi:10.1016/j.scitotenv.2019.07.367

Cited by 247 publications

(101 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The combination of these components revealed that LSTM-FC outperforms the vanilla versions of NN and LSTM because it can memorize a long-term dependency. To consider the spatio-temporal dependency, Pak et al [17] proposed a neural network model, called CNN-LSTM, with two components. The first component is a spatio-temporal convolutional neural network (CNN), and the second component is an LSTM model.…”

Section: Prediction With a Neural Network Modelmentioning

confidence: 99%

“…Table 2 shows the list of variables with annotations on their relevant types. Although most variables are frequently utilized features in studies [10,12,15,17,19,20], to our knowledge, there are no prior studies on developing a GPR model with topographic information, traffic information, ultraviolet information, and power plant operation information. Variables, such as wind direction and topographic categories, require further explanations because these two variables are converted into a set of dummy variables by the discretization.…”

Section: Gaussian Process Regressionmentioning

confidence: 99%

See 1 more Smart Citation

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Jang

Shin

Lee

et al. 2020

Sensors

View full text Add to dashboard Cite

Recently, the population of Seoul has been affected by particulate matter in the atmosphere. This problem can be addressed by developing an elaborate forecasting model to estimate the concentration of fine dust in the metropolitan area. We present a forecasting model of the fine dust concentration with an extended range of input variables, compared to existing models. The model takes inputs from holistic perspectives such as topographical features on the surface, chemical sources of the fine dusts, traffic and the human activities in sub-areas, and meteorological data such as wind, temperature, and humidity, of fine dust. Our model was evaluated by the index-of-agreement (IOA) and the root mean-squared error (RMSE) in predicting PM2.5 and PM10 over three subsequent days. Our model variations consist of linear regressions, ARIMA, and Gaussian process regressions (GPR). The GPR showed the best performance in terms of IOA that is over 0.6 in the three-day predictions.

show abstract

Section: Prediction With a Neural Network Modelmentioning

confidence: 99%

Section: Gaussian Process Regressionmentioning

confidence: 99%

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Jang

Shin

Lee

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…There are two major paradigms of methodology related to PM 2.5 prediction, deterministic and statistical methods. The statistical methods have shown better performance because the nonlinear and heterogeneous nature of processes in the formation and transportation of air pollution [8]. With the advance of technology and decreasing cost of sensors for collecting air quality data, data mining and machining learning methods become more and more important, such as time series analysis [9], [10], random forest [11], [12], principal component analysis [13], Kalman filters [14], support vector machines (SVMs) [15], [16], and artificial neural networks (ANNs) [5], [7], [17].…”

Section: Introductionmentioning

confidence: 99%

Dynamic System Approach for Improved PM_2.5 Prediction in Taiwan

et al. 2020

View full text Add to dashboard Cite

“…However, the above research results are based on short-term forecasts. Pak et al [29] gave a next-day prediction, but the data they used were the average concentration of PM2.5 per day, so the predicted result did not show the hourly variation of the next day. Therefore, from the perspective of a prediction problem, it is also a short-term prediction.…”

mentioning

confidence: 99%

“…However, long-term forecasts are very meaningful, especially for the management of air quality. For example, the model of Xu et al [3] is based on the concentration of PM2.5 per hour, giving the concentration of PM2.5 per hour for the next 10 h. Compared with [29], which only gives the average of the next day, Xu's method [3] can provide…”

mentioning

confidence: 99%

Integrated Predictor Based on Decomposition Mechanism for PM2.5 Long-Term Prediction

et al. 2019

View full text Add to dashboard Cite

Featured Application: This work can be used in the intelligent system for the smart city, smart agriculture, etc.Abstract: It is crucial to predict PM2.5 concentration for early warning regarding and the control of air pollution. However, accurate PM2.5 prediction has been challenging, especially in long-term prediction. PM2.5 monitoring data comprise a complex time series that contains multiple components with different characteristics; therefore, it is difficult to obtain an accurate prediction by a single model. In this study, an integrated predictor is proposed, in which the original data are decomposed into three components, that is, trend, period, and residual components, and then different sub-predictors including autoregressive integrated moving average (ARIMA) and two gated recurrent units are used to separately predict the different components. Finally, all the predictions from the sub-predictors are combined in fusion node to obtain the final prediction for the original data. The results of predicting the PM2.5 time series for Beijing, China showed that the proposed predictor can effectively improve prediction accuracy for long-term prediction. Appl. Sci. 2019, 9, 4533 2 of 14 models [14] and vector auto-regression (VAR) [15]. ARIMA is a linear modeling method that provides accurate predictions for approximate linear relationships. However, its prediction performance is not good enough for nonlinear prediction problems.To capture the complex nonlinearity of PM2.5, a back propagation (BP) neural network has been widely applied in time-series data prediction [16][17][18]. Other shallow nonlinear networks based on a machine-learning mechanism such as the improved gray neural network model [19] and the radial basis function (RBF) neural network [20] have also been used to predict time-series data. Xu et al. [3] proposed a supplementary leaky integrator echo state network (SLI-ESN), which added the historical state term of the historical moment to the calculation of a leaky integrator reservoir. Compared with an echo state network (ESN), a leaky integrator ESN (LI-ESN), an extreme learning machine (ELM), a hierarchical ELM (H-ELM), a stacked auto-encoder (SAE), and a traditional SLI-ESN, the proposed SLI-ESN of Xu et al.[3] could achieve good prediction results, but its long-term predictions were not satisfactory.As a shallow network still cannot effectively extract the complex nonlinearity of the data, decomposition must be used; that is, the data must be decomposed into multiple components to reduce its complexity, and then multiple sub-models are used to improve the prediction performance. For example, García et al. [21] decomposed a long-term data series into smaller seasonal component patterns. Jesús et al. [22] divided a pollen-concentration data series into seasonal and random parts, used partial least-squares regression (PLSR) to fit the residuals, and established an airborne pollen time-series model to predict the daily pollen concentration. Ming et al.[23] extracted accurate seasonal signals, a...

show abstract

Deep learning-based PM2.5 prediction considering the spatiotemporal correlations: A case study of Beijing, China

Cited by 247 publications

References 32 publications

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Dynamic System Approach for Improved PM_2.5 Prediction in Taiwan

Integrated Predictor Based on Decomposition Mechanism for PM2.5 Long-Term Prediction

Contact Info

Product

Resources

About

Deep learning-based PM2.5 prediction considering the spatiotemporal correlations: A case study of Beijing, China

Cited by 247 publications

References 32 publications

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Forecasting the Concentration of Particulate Matter in the Seoul Metropolitan Area Using a Gaussian Process Model

Dynamic System Approach for Improved PM2.5 Prediction in Taiwan

Integrated Predictor Based on Decomposition Mechanism for PM2.5 Long-Term Prediction

Contact Info

Product

Resources

About

Dynamic System Approach for Improved PM_2.5 Prediction in Taiwan