Prediction of size-fractionated airborne particle-bound metals using MLR, BP-ANN and SVM analyses

被引:75
作者
Leng, Xiang'zi [1 ]
Wang, Jinhua [1 ]
Ji, Haibo [1 ]
Wang, Qin'geng [1 ,2 ]
Li, Huiming [1 ]
Qian, Xin [1 ,2 ]
Li, Fengying [2 ]
Yang, Meng [2 ]
机构
[1] Nanjing Univ, Sch Environm, State Key Lab Pollut Control & Resources Reuse, Nanjing 210023, Jiangsu, Peoples R China
[2] Nanjing Univ Informat Sci & Technol, Sch Environm Sci & Engn, Jiangsu Key Lab Atmospher Environm Monitoring & P, Jiangsu Collaborat Innovat Ctr Atmospher Environm, Nanjing 210044, Jiangsu, Peoples R China
基金
中国国家自然科学基金;
关键词
Airborne particulate matter (PM); Heavy metals; Back propagation artificial neural network (BP-ANN); Support vector machine (SVM); Prediction; ARTIFICIAL NEURAL-NETWORKS; SUPPORT VECTOR MACHINE; PARTICULATE AIR-POLLUTION; VOLUNTEER NATURAL RELOCATION; FINE PARTICULATE; TRACE-ELEMENTS; CHEMICAL SPECIATION; BLOOD-PRESSURE; MATTER; EXPOSURE;
D O I
10.1016/j.chemosphere.2017.04.015
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Size-fractionated heavy metal concentrations were observed in airborne particulate matter (PM) samples collected from 2014 to 2015 (spanning all four seasons) from suburban (Xianlin) and industrial (Pukou) areas in Nanjing, a megacity of southeast China. Rapid prediction models of size-fractionated metals were established based on multiple linear regression (MLR), back propagation artificial neural network (BP-ANN) and support vector machine (SVM) by using meteorological factors and PM concentrations as input parameters. About 38% and 77% of PM2.5 concentrations in Xianlin and Pukou, respectively, were beyond the Chinese National Ambient Air Quality Standard limit of 75 mu g/m(3). Nearly all elements had higher concentrations in industrial areas, and in winter among the four seasons. Anthropogenic elements such as Pb, Zn, Cd and Cu showed larger percentages in the fine fraction (empty set <= 2.5 mu m), whereas the crustal elements including Al, Ba, Fe, Ni, Sr and Ti showed larger percentages in the coarse fraction (empty set > 2.5 mu m). SVM showed a higher training correlation coefficient (R), and lower mean absolute error (MAE) as well as lower root mean square error (RMSE), than MLR and BP-ANN for most metals. All the three methods showed better prediction results for Ni, Al, V, Cd and As, whereas relatively poor for Cr and Fe. The daily airborne metal concentrations in 2015 were then predicted by the fully trained SVM models and the results showed the heaviest pollution of airborne heavy metals occurred in December and January, whereas the lightest pollution occurred in June and July. (C) 2017 Elsevier Ltd. All rights reserved.
引用
收藏
页码:513 / 522
页数:10
相关论文
共 45 条
[1]   Insights Into the Mechanisms and Mediators of the Effects of Air Pollution Exposure on Blood Pressure and Vascular Function in Healthy Humans [J].
Brook, Robert D. ;
Urch, Bruce ;
Dvonch, J. Timothy ;
Bard, Robert L. ;
Speck, Mary ;
Keeler, Gerald ;
Morishita, Masako ;
Marsik, Frank J. ;
Kamal, Ali S. ;
Kaciroti, Niko ;
Harkema, Jack ;
Corey, Paul ;
Silverman, Frances ;
Gold, Diane R. ;
Wellenius, Greg ;
Mittleman, Murray A. ;
Rajagopalan, Sanjay ;
Brook, Jeffrey R. .
HYPERTENSION, 2009, 54 (03) :659-667
[2]   Comparison of support vector machine and artificial neural network systems for drug/nondrug classification [J].
Byvatov, E ;
Fechner, U ;
Sadowski, J ;
Schneider, G .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 2003, 43 (06) :1882-1889
[3]   A hybrid ARIMA and artificial neural networks model to forecast particulate matter in urban areas: The case of Temuco, Chile [J].
Diaz-Robles, Luis A. ;
Ortega, Juan C. ;
Fu, Joshua S. ;
Reed, Gregory D. ;
Chow, Judith C. ;
Watson, John G. ;
Moncada-Herrera, Juan A. .
ATMOSPHERIC ENVIRONMENT, 2008, 42 (35) :8331-8340
[4]   Trace elements in size-segregated urban aerosol in relation to the anthropogenic emission sources and the resuspension [J].
Dordevic, Dragana ;
Stortini, Angela Maria ;
Relic, Dubravka ;
Mihajlidi-Zelic, Aleksandra ;
Huremovic, Jasna ;
Barbante, Carlo ;
Gambaro, Andrea .
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH, 2014, 21 (18) :10949-10959
[5]   Chemical speciation of fine particle bound trace metals [J].
Feng, X. D. ;
Dang, Z. ;
Huang, W. L. ;
Yang, C. .
INTERNATIONAL JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY, 2009, 6 (03) :337-346
[6]   Ozone concentration forecast method based on genetic algorithm optimized back propagation neural networks and support vector machine data classification [J].
Feng, Yu ;
Zhang, Wenfang ;
Sun, Dezhi ;
Zhang, Liqiu .
ATMOSPHERIC ENVIRONMENT, 2011, 45 (11) :1979-1985
[7]   QSRR modeling for diverse drugs using different feature selection methods coupled with linear and nonlinear regressions [J].
Goodarzi, Mohammad ;
Jensen, Richard ;
Vander Heyden, Yvan .
JOURNAL OF CHROMATOGRAPHY B-ANALYTICAL TECHNOLOGIES IN THE BIOMEDICAL AND LIFE SCIENCES, 2012, 910 :84-94
[8]   Short-term Impact of Ambient Air Pollution and Air Temperature on Blood Pressure Among Pregnant Women [J].
Hampel, Regina ;
Lepeule, Johanna ;
Schneider, Alexandra ;
Bottagisi, Sebastien ;
Charles, Marie-Aline ;
Ducimetiere, Pierre ;
Peters, Annette ;
Slama, Remy .
EPIDEMIOLOGY, 2011, 22 (05) :671-679
[9]   Estimation of the Contributions of Brake Dust, Tire Wear, and Resuspension to Nonexhaust Traffic Particles Derived from Atmospheric Measurements [J].
Harrison, Roy M. ;
Jones, Alan M. ;
Gietl, Johanna ;
Yin, Jianxin ;
Green, David C. .
ENVIRONMENTAL SCIENCE & TECHNOLOGY, 2012, 46 (12) :6523-6529
[10]   Short-time forecasting of atmospheric NOx concentration by neural networks [J].
Hoffman, Szymon .
ENVIRONMENTAL ENGINEERING SCIENCE, 2006, 23 (04) :603-609