Hourly PM2.5 concentration forecast using stacked autoencoder model with emphasis on seasonality

Abstract Accurate PM2.5 forecasting provides a possibility for establishing an early warning system to notify the public and take precautionary measures to prevent negative effects on ambient air quality and public health. Considering strong seasonal variation in meteorological conditions, in this paper, a seasonal stacked autoencoder model combining seasonal analysis and deep feature learning is proposed for forecasting the hourly PM2.5 concentration, named DL-SSAE model. The original data are firstly decomposed into four seasonal subseries according to the Chinese calendar, and then the Kendall correlation coefficient method is employed to search inherent relationships between PM2.5 concentrations and meteorological parameters within 1-h ahead for each seasonal time series. The inherent relationships of each seasonal subseries are finally extracted, learned, and modeled by different deep neural networks (stacked autoencoders for regression), and the hourly PM2.5 forecasts are yielded. The addressed model is tested by the dataset collected from three environmental monitoring stations in Beijing, China. The results demonstrate that the proposed model outperforms all other considered models with/without seasonality consideration in this paper.

[1]  Hong Huang,et al.  Relevance analysis and short-term prediction of PM2.5 concentrations in Beijing based on multi-source data , 2017 .

[2]  James M. Wilczak,et al.  PM 2.5 analog forecast and Kalman filter post-processing for the Community Multiscale Air Quality (CMAQ) model , 2015 .

[3]  Martha Cobo,et al.  Discovering relationships and forecasting PM10 and PM2.5 concentrations in Bogotá, Colombia, using Artificial Neural Networks, Principal Component Analysis, and k-means clustering , 2018, Atmospheric Pollution Research.

[4]  G. Pfister,et al.  Spatiotemporal prediction of fine particulate matter during the 2008 northern California wildfires using machine learning. , 2015, Environmental science & technology.

[5]  Yun Bai,et al.  Deep Feature Learning Architectures for Daily Reservoir Inflow Forecasting , 2016, Water Resources Management.

[6]  Yu Xue,et al.  Prediction of particulate matter at street level using artificial neural networks coupling with chaotic particle swarm optimization algorithm , 2014 .

[7]  Jingjing Xie,et al.  Daily reservoir inflow forecasting using multiscale deep feature learning with hybrid models , 2016 .

[8]  Diana Domanska,et al.  Application of fuzzy time series models for forecasting pollution concentrations , 2012, Expert Syst. Appl..

[9]  Jingjing Xie,et al.  Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions , 2016 .

[10]  W. Geoffrey Cobourn,et al.  An enhanced PM2.5 air quality forecast model based on nonlinear regression and back-trajectory concentrations , 2010 .

[11]  P. Goyal,et al.  Artificial intelligence based approach to forecast PM2.5 during haze episodes: A case study of Delhi, India , 2015 .

[12]  Deepak Singh,et al.  Statistical modeling of O3, NOx, CO, PM2.5, VOCs and noise levels in commercial complex and associated health risk assessment in an academic institution. , 2016, The Science of the total environment.

[13]  Bo Zeng,et al.  A multi-pattern deep fusion model for short-term bus passenger flow forecasting , 2017, Appl. Soft Comput..

[14]  Marcella Busilacchio,et al.  Recursive neural network model for analysis and forecast of PM10 and PM2.5 , 2017 .

[15]  Minglei Fu,et al.  Prediction of particular matter concentrations by developed feed-forward neural network with rolling mechanism and gray model , 2015, Neural Computing and Applications.

[16]  Fang Deng,et al.  The MR-CA Models for Analysis of Pollution Sources and Prediction of PM2.5 , 2019, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[17]  Xin Fang,et al.  Spatial modeling of PM2.5 concentrations with a multifactoral radial basis function neural network , 2015, Environmental Science and Pollution Research.

[18]  Luca Delle Monache,et al.  Improving NOAA NAQFC PM2.5 predictions with a bias correction approach , 2017 .

[19]  Koji Zettsu,et al.  Dynamically pre-trained deep recurrent neural networks using environmental monitoring data for predicting PM2.5 , 2015, Neural Computing and Applications.

[20]  Vikas Singh,et al.  Higher Pollution Episode Detection Using Image Classification Techniques , 2016, Environmental Modeling & Assessment.

[21]  Pascal Vincent,et al.  Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives , 2012, ArXiv.

[22]  F. Hajiloo,et al.  Impact assessment of meteorological and environmental parameters on PM2.5 concentrations using remote sensing data and GWR analysis (case study of Tehran) , 2018, Environmental Science and Pollution Research.

[23]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[24]  Ping-Huan Kuo,et al.  A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities , 2018, Sensors.

[25]  Sachit Mahajan,et al.  Improving the Accuracy and Efficiency of PM2.5 Forecast Service Using Cluster-Based Hybrid Neural Network Model , 2018, IEEE Access.

[26]  Muhammad Mushtaq Khan,et al.  Triangular relationship among energy consumption, air pollution and water resources in Pakistan , 2016 .

[27]  Le Jian,et al.  An application of ARIMA model to predict submicron particle concentrations from meteorological factors at a busy roadside in Hangzhou, China. , 2012, The Science of the total environment.

[28]  George D. C. Cavalcanti,et al.  Hybrid intelligent system for air quality forecasting using phase adjustment , 2014, Eng. Appl. Artif. Intell..

[29]  Xiang Li,et al.  Deep learning architecture for air quality predictions , 2016, Environmental Science and Pollution Research.

[30]  Diego Cabrera,et al.  Multimodal deep support vector classification with homologous features and its application to gearbox fault diagnosis , 2015, Neurocomputing.

[31]  Lifeng Wu,et al.  Prediction of air quality indicators for the Beijing-Tianjin-Hebei region , 2018, Journal of Cleaner Production.

[32]  Alexei Stepanov,et al.  On the Kendall Correlation Coefficient , 2015, 1507.01427.

[33]  Yu Liu,et al.  Autoencoder-based deep belief regression network for air particulate matter concentration forecasting , 2018, Journal of Intelligent & Fuzzy Systems.

[34]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[35]  Mingxing Chen,et al.  The impacts of Chinese Nian culture on air pollution , 2016 .

[36]  Rakesh Kumar,et al.  Evaluation of control strategies for industrial air pollution sources using American Meteorological Society/Environmental Protection Agency Regulatory Model with simulated meteorology by Weather Research and Forecasting Model , 2016 .

[37]  Gabriela Polezer,et al.  Assessing the impact of PM2.5 on respiratory disease using artificial neural networks. , 2018, Environmental pollution.

[38]  Yoshua Bengio,et al.  Extracting and composing robust features with denoising autoencoders , 2008, ICML '08.

[39]  Uwe Stilla,et al.  Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks , 2016, IEEE Geoscience and Remote Sensing Letters.

[40]  David Morley,et al.  A land use regression variable generation, modelling and prediction tool for air pollution exposure assessment , 2018, Environmental Modelling & Software.

[41]  Y. Hao,et al.  How harmful is air pollution to economic development? New evidence from PM2.5 concentrations of Chinese cities , 2018 .

[42]  Donald Huisingh,et al.  On moving towards an ecologically sound society: with special focus on preventing future smog crises in China and globally☆ , 2014 .

[43]  Daniel Dunea,et al.  Using wavelet–feedforward neural networks to improve air pollution forecasting in urban environments , 2015, Environmental Monitoring and Assessment.

[44]  Beibei Sun,et al.  Analysis and forecasting of the particulate matter (PM) concentration levels over four major cities of China using hybrid models , 2014 .

[45]  Prashant Kumar,et al.  Developing a methodology to predict PM10 concentrations in urban areas using generalized linear models , 2016, Environmental technology.

[46]  Wei Sun,et al.  Daily PM2.5 concentration prediction based on principal component analysis and LSSVM optimized by cuckoo search algorithm. , 2017, Journal of environmental management.