Prediction of vertical PM<sub>2.5</sub> concentrations alongside an elevated expressway by using the neural network hybrid model and generalized additive model

doi:10.1007/s11707-016-0593-0

Front. Earth Sci.

2017, Vol. 11

Issue (2) : 347-360 https://doi.org/10.1007/s11707-016-0593-0

RESEARCH ARTICLE

Prediction of vertical PM_2.5 concentrations alongside an elevated expressway by using the neural network hybrid model and generalized additive model

Ya GAO¹, Zhanyong WANG¹, Qing-Chang LU¹(

), Chao LIU², Zhong-Ren PENG^1,²(

), Yue YU

¹. Center for ITS and UAV Applications Research, State Key Laboratory of Ocean Engineering, School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
². Department of Urban and Regional Planning, University of Florida, FL 32611-5706, USA
³. Department of Electrical and Systems Engineering, University of Pennsylvania, PA 19104-6314, USA

Download: PDF(1873 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

A study on vertical variation of PM_2.5 concentrations was carried out in this paper. Field measurements were conducted at eight different floor heights outside a building alongside a typical elevated expressway in downtown Shanghai, China. Results show that PM_2.5 concentration decreases significantly with the increase of height from the 3rd to 7th floor or the 8th to 15th floor, and increases suddenly from the 7th to 8th floor which is the same height as the elevated expressway. A non-parametric test indicates that the data of PM_2.5 concentration is statistically different under the 7th floor and above the 8th floor at the 5% significance level. To investigate the relationships between PM_2.5 concentration and influencing factors, the Pearson correlation analysis was performed and the results indicate that both traffic and meteorological factors have crucial impacts on the variation of PM_2.5 concentration, but there is a rather large variation in correlation coefficients under the 7th floor and above the 8th floor. Furthermore, the back propagation neural network based on principal component analysis (PCA-BPNN), as well as generalized additive model (GAM), was applied to predict the vertical PM_2.5 concentration and examined with the field measurement dataset. Experimental results indicated that both models can obtain accurate predictions, while PCA-BPNN model provides more reliable and accurate predictions as it can reduce the complexity and eliminate data co-linearity. These findings reveal the vertical distribution of PM_2.5 concentration and the potential of the proposed model to be applicable to predict the vertical trends of air pollution in similar situations.

Keywords vertical variations principal component analysis back propagation neural network generalized additive model urban elevated expressway

Corresponding Author(s): Qing-Chang LU,Zhong-Ren PENG

Online First Date: 17 October 2016 Issue Date: 19 May 2017

Cite this article:

Ya GAO,Zhanyong WANG,Qing-Chang LU, et al. Prediction of vertical PM_2.5 concentrations alongside an elevated expressway by using the neural network hybrid model and generalized additive model[J]. Front. Earth Sci., 2017, 11(2): 347-360.

URL:

https://academic.hep.com.cn/fesci/EN/10.1007/s11707-016-0593-0
https://academic.hep.com.cn/fesci/EN/Y2017/V11/I2/347

Fig.1 Schematic of field measurement. (a) Field measurement site. (b) Shangzhong Building. (c) The distribution of monitoring sites in vertical section and within the street canyon.

Tab.1 Descriptions of Traffic and Meteorological Parameters

Fig.2 Wind rose of wind speed and direction measured in the experiment period: (a) February 14th. (b) February 23th.

Fig.3 Architecture of a PCA-BPNN model.

Fig.4 Vertical variations of PM_2.5 concentration with height.

Tab.2 Results of Wilcox-in Rank-Sum test and correlation coefficient analysis.

Fig.5 PCA results of original variables for different cases. (a) Scree plot and respective cumulative variance (%). (b) Communalities of original variables.

Tab.3 Performance of PAC-BPNN model for prediction.

Fig.6 Observed versus training and testing PM2.5 concentrations (all concentration in mg/m³). (a) training Case I. (b) testing Case I. (c) training Case II. (d) testing Case II.

Tab.4 Significance of smooth terms in Case I and Case II.

Fig.7 Estimated effects (on the original scale) of high, temperature, relatively outdoor humidity, sun radiation, traffic volumes from Shangzhong road and Middle-Ring Elevated Expressway and bivariate wind components. (a) Case I. (b) Case II. The dashed lines are the estimated 95% confidence intervals.

Tab.5 Performance of GAM model for prediction

Fig.8 Comparison of observations and predictions by the models at eight-floor height.

1	Abdul-Wahab S A, Bakheit C S, Al-Alawi S M (2005). Principal component and multiple regression analysis in modelling of ground-level ozone and factors affecting its concentrations. Environ Model Softw, 20(10): 1263–1271 https://doi.org/10.1016/j.envsoft.2004.09.001
2	Aldrin M, Haff I H (2005). Generalised additive modelling of air pollution, traffic volume and meteorology. Atmos Environ, 39(11): 2145–2155 https://doi.org/10.1016/j.atmosenv.2004.12.020
3	Cai M, Yin Y, Xie M (2009). Prediction of hourly air pollutant concentrations near urban arterials using artificial neural network approach. Transp Res Part D Transp Environ, 14(1): 32–41 https://doi.org/10.1016/j.trd.2008.10.004
4	Carslaw D C, Beevers S D, Tate J E (2007). Modelling and assessing trends in traffic-related emissions using a generalised additive modelling approach. Atmos Environ, 41(26): 5289–5299 https://doi.org/10.1016/j.atmosenv.2007.02.032
5	Chan L Y, Kwok W S (2000). Vertical dispersion of suspended particulates in urban area of Hong Kong. Atmos Environ, 34(26): 4403–4412 https://doi.org/doi:10.1016/S1352-2310(00)00181-3
6	Colls J J, Micallef A (1999). Measured and modelled concentrations and vertical profiles of airborne particulate matter within the boundary layer of a street canyon. Sci Total Environ, 235(1‒3): 221–233 https://doi.org/10.1016/S0048-9697(99)00194-1
7	Chaloulakou A, Saisana M, Spyrellis N (2003). Comparative assessment of neural networks and regression models for forecasting summertime ozone in Athens. Science of the Total Environment, 313(1): 1–13
8	Gardner M W, Dorling S R (2000). Statistical surface ozone models: an improved methodology to account for non-linear behaviour. Atmospheric Environment, 34(1): 21–34
9	Hastie T J, Tibshirani R J (1990). Generalized additive models. London: Chapman and Hall
10	He H D, Lu W Z (2012). Urban aerosol particulates on Hong Kong roadsides: size distribution and concentration levels with time. Stochastic Environ Res Risk Assess, 26(2): 177–187 https://doi.org/10.1007/s00477-011-0465-9
11	He H D, Lu W Z, Xue Y (2014). Prediction of particulate matters at urban intersection by using multilayer perceptron model based on principal components. Stochastic Environ Res Risk Assess, 29(8): 2107–2114
12	He J, Qi Z, Zhao C, Bao X (2009). Simulations of pollutant dispersion at toll plazas using three-dimensional CFD models. Transp Res Part D Transp Environ, 14(8): 557–566 https://doi.org/10.1016/j.trd.2009.08.001
13	Kumar P, Fennell P, Langley D, Britter R (2008). Pseudo-simultaneous measurements for the vertical variation of coarse fine and ultrafine particles in an urban street canyon. Atmos Environ, 42(18): 4304–4319 https://doi.org/10.1016/j.atmosenv.2008.01.010
14	Kumar P, Garmory A, Ketzel M, Berkowicz R, Britter R (2009). Comparative study of measured and modelled number concentrations of nanoparticles in an urban street canyon. Atmos Environ, 43(4): 949–958 https://doi.org/10.1016/j.atmosenv.2008.10.025
15	Li X, Wang J, Tu X D, Liu W, Huang Z (2007). Vertical variations of particle number concentration and size distribution in a street canyon in Shanghai, China. Sci Total Environ, 378(3): 306–316 https://doi.org/10.1016/j.scitotenv.2007.02.040
16	Longley I D, Gallagher M W, Dorsey J R, Flynn M (2004). A case-study of fine particle concentrations and fluxes measured in a busy street canyon in Manchester, UK. Atmos Environ, 38(22): 3595–3603 https://doi.org/10.1016/j.atmosenv.2004.03.040
17	Mazzoleni C, Moosmüller H, Kuhns H D, Keislar R E, Barber P W, Nikolic D, Nussbaum N J, Watson J G (2004). Correlation between automotive CO, HC, NO, and PM emission factors from on-road remote sensing: implications for inspection and maintenance programs. Transp Res Part D Transp Environ, 9(6): 477–496 https://doi.org/10.1016/j.trd.2004.08.006
18	McNabola A, Broderick B M, Gill L W (2009). The impacts of inter-vehicle spacing on in-vehicle air pollution concentrations in idling urban traffic conditions. Transp Res Part D Transp Environ, 14(8): 567–575 https://doi.org/10.1016/j.trd.2009.08.003
19	Milionis A E, Davies T D (1994). Box-Jenkins univariate modelling for climatological time series analysis: an application to the monthly activity of temperature inversions. International Journal of Climatology, 14(5): 569–579.
20	Moseholm L, Silva J, Larson T (1996). Forecasting carbon monoxide concentrations near a sheltered intersection using video traffic surveillance and neural networks. Transp Res Part D Transp Environ, 1(1): 15–28 https://doi.org/10.1016/S1361-9209(96)00002-8
21	Muñoz E, Martin M L, Turias I J, Jimenez-Come M J, Trujillo F J (2014). Prediction of PM10 and SO2 exceedances to control air pollution in the Bay of Algeciras, Spain. Stochastic Environ Res Risk Assess, 28(6): 1409–1420 https://doi.org/10.1007/s00477-013-0827-6
22	Nagendra S S, Khare M (2006). Artificial neural network approach for modelling nitrogen dioxide dispersion from vehicular exhaust emissions. Ecol Modell, 190(1‒2): 99–115 https://doi.org/10.1016/j.ecolmodel.2005.01.062
23	Ng H K T, Balakrishnan N, Panchapakesan S (2007). Selecting the best population using a test for equality based on minimal Wilcoxon rank-sum precedence statistic. Methodology and Computing in Applied Probability, 9(2): 263–305
24	Schleicher N J, Norra S, Chai F, Chen Y, Wang S, Cen K, Yu Y, Stüben D (2011). Temporal variability of trace metal mobility of urban particulate matter from Beijing–A contribution to health impact assessments of aerosols. Atmos Environ, 45(39): 7248–7265 https://doi.org/10.1016/j.atmosenv.2011.08.067
25	Schlink U, Dorling S, Pelikan E, Nunnari G, Cawley G, Junninen H, Greig A, Foxall R, Eben K, Chatterton T, Vondracek J, Richter M, Dostal M, Bertucco L, Kolehmainen M, Doyle M (2003). A rigorous inter-comparison of ground-level ozone predictions. Atmos Environ, 37(23): 3237–3253 https://doi.org/10.1016/S1352-2310(03)00330-3
26	Sousa S I V, Martins F G, Alvim-Ferraz M C M, Pereira M C (2007). Multiple linear regression and artificial neural networks based on principal components to predict ozone concentrations. Environ Model Softw, 22(1): 97–103 https://doi.org/10.1016/j.envsoft.2005.12.002
27	Tang T Q, Huang H J, Shang H Y (2015c). Influences of the driver’s bounded rationality on micro driving behavior, fuel consumption and emissions. Transp Res Part D Transp Environ, 41: 423–432 https://doi.org/10.1016/j.trd.2015.10.016
28	Tang T Q, Yu Q, Yang S C, Ding C (2015a). Impacts of the vehicle’s fuel consumption and exhaust emissions on the trip cost allowing late arrival under car-following model. Physica A: Statistical Mechanics and its Applications, 431: 52–62
29	Tang T Q, Li J G, Yang S C, Shang H Y (2015b). Effects of on-ramp on the fuel consumption of the vehicles on the main road under car-following model. Physica A: Statistical Mechanics and its Applications, 419: 293–300
30	Wang J S, Chan T L, Ning Z, Leung C W, Cheung C S, Hung W T (2006). Roadside measurement and prediction of CO and PM2.5 dispersion from on-road vehicles in Hong Kong. Transp Res Part D Transp Environ, 11(4): 242–249 https://doi.org/10.1016/j.trd.2006.04.002
31	Wang J S, Huang Z (2002). Numerical study on impact of urban viaduct on local-scale of atmospheric environment. Shanghai Environmental Sciences, 21(3): 132–135
32	Wang Z, He H D, Lu F, Lu Q C, Peng Z R (2015a). Hybrid model for prediction of carbon monoxide and fine particulate matter concentrations near a road intersection. Transp Res Rec, 2503: 29–38 https://doi.org/10.3141/2503-04
33	Wang Z, Lu F, He H D, Lu Q C, Wang D, Peng Z R (2015b). Fine-scale estimation of carbon monoxide and fine particulate matter concentrations in proximity to a road intersection by using wavelet neural network with genetic algorithm. Atmos Environ, 104: 264–272 https://doi.org/10.1016/j.atmosenv.2014.12.058
40	Wang Z, Lu Q C, He H D, Wang D, Gao Y, Peng Z R (2016). Investigation of the spatiotemporal variation and influencing factors on fine particulate matter and carbon monoxide concentrations near a road intersection. Front. Earth Sci., doi: 10.1007/s11707-016-0564-5
34	Weber S, Kuttler W, Weber K (2006). Flow characteristics and particle mass and number concentration variability within a busy urban street canyon. Atmos Environ, 40(39): 7565–7578 https://doi.org/10.1016/j.atmosenv.2006.07.002
35	Wood S N, Augustin N H (2002). GAMs with integrated model selection using penalized regression splines and applications to environmental modelling. Ecol Modell, 157(2‒3): 157–177 https://doi.org/10.1016/S0304-3800(02)00193-X
36	Zhang C J, Zeng J R, Wen M, Zhang G L, Fang H P, Li Y (2012). Influence of Viaducts on Dispersion of Air Particles in Street Canyons. Research of Environmental Sciences, 25(2): 159–164.
37	Zhang D Z, Peng Z R (2014). Near-road fine particulate matter concentration estimation using artificial neural network approach. Int J Environ Sci Technol, 11(8): 2403–2412 https://doi.org/10.1007/s13762-014-0565-4
38	Zhang K, Batterman S (2010). Near-road air pollutant concentrations of CO and PM2.5: A comparison of MOBILE6. 2/CALINE4 and generalized additive models. Atmos Environ, 44(14): 1740–1748 https://doi.org/10.1016/j.atmosenv.2010.02.008
39	Zhang L D, Zhu W X (2015). Delay-feedback control strategy for reducing emission of traffic flow system. Physica A: Statistical Mechanics and its Applications, 428: 481–492

[1]	Xue YU, Yue LI, Min XI, Fanlong KONG, Mingyue PANG, Zhengda YU. Ecological vulnerability analysis of Beidagang National Park, China[J]. Front. Earth Sci., 2019, 13(2): 385-397.
[2]	Tao XU, Xiang LI, Christophe CLARAMUNT. Trip-oriented travel time prediction (TOTTP) with historical vehicle trajectories[J]. Front. Earth Sci., 2018, 12(2): 253-263.
[3]	Meilin WU,Youshao WANG,Junde DONG,Fulin SUN,Yutu WANG,Yiguo HONG. Spatial assessment of water quality using chemometrics in the Pearl River Estuary, China[J]. Front. Earth Sci., 2017, 11(1): 114-126.
[4]	David BRAND,Chathurika WIJEWARDANA,Wei GAO,K. Raja REDDY. Interactive effects of carbon dioxide, low temperature, and ultraviolet-B radiation on cotton seedling root and shoot morphology and growth[J]. Front. Earth Sci., 2016, 10(4): 607-620.
[5]	Chengcheng FENG,Xiaolei ZOU,Juan ZHAO. Detection of radio-frequency interference signals from AMSR-E data over the United States with snow cover[J]. Front. Earth Sci., 2016, 10(2): 195-204.
[6]	K. C. TAN,H. S. LIM,M. Z. MAT JAFRI. A statistical model to predict total column ozone in Peninsular Malaysia[J]. Front. Earth Sci., 2016, 10(1): 63-73.
[7]	Sibo ZHANG,Li GUAN. Identifying AMSR-E radio-frequency interference over winter land[J]. Front. Earth Sci., 2015, 9(3): 437-448.
[8]	Xiaoyun FAN, Baoshan CUI, Zhiming ZHANG. Spatial variations of river water quality in Pearl River Delta, China[J]. Front Earth Sci, 2012, 6(3): 291-296.

Viewed

Full text

Abstract

Cited

Shared

Discussed