Support vector machine regression forecasting of O3 concentrations based on wavelet transformation
SU Xiao-qian1, AN Jun-lin1, ZHANG Yu-xin2
1. Key Laboratory of Meteorological Disaster, Ministry of Education, Joint International Research Laboratory of Climate and Environment Change, Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044, China;
2. Weather Modification Office of Qinghai Province, Xining 810001, China
In the paper, a hybrid model combining multivariable linear regression (MLR) and wavelet transformation (WT) to improve the support vector machine regression (SVR) forecast accuracy of hourly ozone (O3) concentrations, by employing the observations of meteorological variables and O3 and its precursors from June 1to August 15, 2016 at an industrial area in Nanjing. The original time series of O3 concentration was decomposed by WT into a few sub-series with lower variability, the prediction strategy applied to each of them and then summed up all individual prediction results. The squared correlation coefficient (R2) of the forecast was 0.90. The mean absolute error (MAE), mean absolute percentage error (MAPE) and root mean squared error (RMSE) were 3.86×10-9, 28.26% and 5.57×10-9, respectively. The results showed clearly that the performance of the M-WT-SVR was better than the M-SVR and SVR for hourly O3 predictions. The low-level detail signals were mainly related to NO, NO2 and aromatic hydrocarbons, while the coarse approximation signals at higher level were synthetically affected by meteorological conditions, precursors and O3 pre-concentrations. Concerning the O3 concentrations forecasting, M-WT-SVR gave significantly better predictions than MLR methods.
苏筱倩, 安俊琳, 张玉欣. 基于支持向量机回归和小波变换的O3预报方法[J]. 中国环境科学, 2019, 39(9): 3719-3726.
SU Xiao-qian, AN Jun-lin, ZHANG Yu-xin. Support vector machine regression forecasting of O3 concentrations based on wavelet transformation. CHINA ENVIRONMENTAL SCIENCECE, 2019, 39(9): 3719-3726.
Duncan B N, Lamsal L N, Thompson A M, et al. A space-based, high-resolution view of notable changes in urban NOx pollution around the world (2005-2014)[J]. Journal of Geophysical Research Atmospheres, 2016,121(2):976-996.
[2]
Krotkov N A, Mclinden C A, Li C, et al. Aura OMI observations of regional SO2 and NO2 pollution changes from 2005 to 2015[J]. Atmospheric Chemistry and Physics, 2016,16(7):4605-4629.
[3]
Lu Q, Zheng J Y, Ye S Q, et al. Emission trends and source characteristics of SO2, NOx, PM10and VOCs in the Pearl River Delta region from 2000 to 2009[J]. Atmospheric Environment, 2013,76:11-20.
[4]
Smedt I D, Stavrakou T, Hendrick F, et al. Diurnal, seasonal and long-term variations of global formaldehyde columns inferred from combined OMI and GOME-2observations[J]. Atmospheric Chemistry and Physics, 2015,15(8):12241-12300.
[5]
Hamer P D, Bowman K W, Henze D K, et al. The impact of observing characteristics on the ability to predict ozone under varying polluted photochemical regimes[J]. Atmospheric Chemistry and Physics, 2015,15(4):4909-4971.
[6]
Karlsson P E, Klingberg J, Engardt M, et al. Past, present and future concentrations of ground-level ozone and potential impacts on ecosystems and human health in northern Europe[J]. Science of The Total Environment, 2017,576:22-35.
[7]
Goodman J E, Prueitt R L, Sax S N, et al. Ozone exposure and systemic biomarkers:evaluation of evidence for adverse cardiovascular health impacts[J]. Critical Reviews in Toxicology, 2015,45(5):412-452.
[8]
Feng Z Z, Hu E Z, Wang X K, et al. Ground-level O3 pollution and its impacts on food crops in China:a review[J]. Environmental Pollution, 2015,199:42-48.
[9]
Wang X K, Zhang Q Q, Zheng F X, et al. Effects of elevated O3concentration on winter wheat and rice yields in the Yangtze River Delta, China[J]. Environmental Pollution, 2012,171:118-125.
[10]
Zhai L, Li S, Zou B, et al. An improved geographically weighted regression model for PM2.5 concentration estimation in large areas[J]. Atmospheric Environment, 2018,181:145-154.
[11]
安俊琳,王跃思,朱彬.主成分和回归分析方法在大气臭氧预报的应用——以北京夏季为例[J]. 环境科学学报, 2010,30(6):1286-1294. An J L, Wang Y S, Zhu B. Principal component and multiple regression analysis predicting ozone concentrations:case study in summer in Beijing[J]. Acta Scientiae Circumstantiae, 2010,30(6):1286-1294.
[12]
Gao M, Yin L T, Ning J C. Artificial neural network model for ozone concentration estimation and Monte Carlo analysis[J]. Atmospheric Environment, 2018,184:129-139.
[13]
Mao X, Shen T, Feng X. Prediction of hourly ground-level PM2.5 concentrations 3days in advance using neural networks with satellite data in eastern China[J]. Atmospheric Pollution Research, 2017,8(6):1005-1015.
[14]
丁愫,陈报章,王瑾,等.基于决策树的统计预报模型在臭氧浓度时空分布预测中的应用研究[J]. 环境科学学报, 2018,38(8):3229-3242. Ding S, Chen B Z, Wang J, et al. An applied research of decision-tree based statistical model in forecasting the spatial-temporal distribution of O3[J]. Acta Scientiae Circumstantiae, 2018,38(8):3229-3242.
[15]
García Nieto P J, Combarro E F, Montañés E, et al. A SVM-based regression model to study the air quality at local scale in Oviedo urban area (Northern Spain):a case study[J]. Applied Mathematics and Computation, 2013,219(17):8923-8937.
[16]
Lu W Z, Wang W J. Potential assessment of the "support vector machine" method in forecasting ambient air pollutant trends[J]. Chemosphere, 2005,59(5):693-701.
[17]
Quej V H, Almorox J, Arnaldo J A, et al. ANFIS, SVM and ANN soft-computing techniques to estimate daily global solar radiation in a warm sub-humid environment[J]. Journal of Atmospheric and Solar-Terrestrial Physics, 2017,155:62-70.
[18]
赵文婧,赵中军,汪结华,等.基于支持向量机的云量精细化预报研究[J]. 干旱气象, 2016,34(3):568-574. Zhao W J, Zhao Z J, Wang J H, et al. A study on refined forecast of cloud cover based on support vector machine[J]. Arid Meteorology, 2016,34(3):568-574.
[19]
吴波,胡邦辉,王学忠,等.基于近似支持向量机的能见度释用预报研究[J]. 热带气象学报, 2017,33(1):104-110. Wu B, Hu B H, Wang X Z, et al. Visibility forecast based on proximal support vector machine[J]. Journal of Tropical Meteorology, 2017, 33(1):104-110.
[20]
Yeganeh B, Motlagh M S P, Rashidi Y, et al. Prediction of CO concentrations based on a hybrid partial least square and support vector machine model[J]. Atmospheric Environment, 2012,55(3):357-365.
[21]
García Nieto P J, García-Gonzalo E, Sánchez A B, et al. Air quality modeling using the PSO-SVM-based approach, MLP neural network, and M5 model tree in the metropolitan area of Oviedo (Northern Spain)[J]. Environmental Modeling & Assessment, 2017,23(3):229-247.
[22]
Canu S, Rakotomamonjy A. Ozone peak and pollution forecasting using support vectors[R]. Yokohama:IFAC Workshop on Environmental Modeling, 2001.
[23]
García Nieto P J, Sánchez Lasheras F, García-Gonzalo E, et al. PM10 concentration forecasting in the metropolitan area of Oviedo (Northern Spain) using models based on SVM, MLP, VARMA and ARIMA:a case study[J]. Science of the Total Environment, 2018,621:753-761.
[24]
Xu Y Z, Yang W D, Wang J Z. Air quality early-warning system for cities in China[J]. Atmospheric Environment, 2017,148:239-257.
[25]
陈亚玲,赵智杰.基于小波变换与传统时间序列模型的臭氧浓度多步预测[J]. 环境科学学报, 2013,33(2):339-345. Chen Y L, Zhao Z J. A multi-step-ahead prediction of ozone concentration using wavelet transform and traditional time series model[J]. Acta Scientiae Circumstantiae, 2013,33(2):339-345.
[26]
Liu H, Mi X W, Li Y F. Wind speed forecasting method based on deep learning strategy using empirical wavelet transform, long short term memory neural network and Elman neural network[J]. Energy Conversion and Management, 2018,156:498-514.
[27]
Li Y, Tao Y, Li C, et al. Daily PM10 concentration forecasting based on multiscale fusion support vector regression[J]. Journal of Intelligent & Fuzzy Systems, 2018,34(6):3833-3844.
[28]
刘建,吴兑,范绍佳,等.前体物与气象因子对珠江三角洲臭氧污染的影响[J]. 中国环境科学, 2017,37(3):813-820. Liu J, Wu D, Fan S J, et al. Impacts of precursors and meteorological factors on ozone pollution in Pearl River Delta[J]. China Environmental Science, 2017,37(3):813-820.
[29]
Farajzadeh J, Alizadeh F. A hybrid linear-nonlinear approach to predict the monthly rainfall over the Urmia Lake watershed using wavelet-SARIMAX-LSSVM conjugated model[J]. Journal of Hydroinformatics, 2017,20(1):246-262.
[30]
Dunea D, Pohoata A, Iordache S. Using wavelet-feedforward neural networks to improve air pollution forecasting in urban environments[J]. Environmental Monitoring and Assessment, 2015,187(7):477.
[31]
王俊秀,安俊琳,邵平,等.南京北郊大气臭氧周末效应特征分析[J]. 环境科学, 2017,38(6):2256-2263. Wang J X, An J L, Shao P, et al. Characteristic study on the "weekend effect" of atmospheric O3 in northern Suburb of Nanjing[J]. Environmental Science, 2017,38(6):2256-2263.
[32]
An J L, Zhu B, Wang H L, et al. Characteristics and source apportionment of VOCs measured in an industrial area of Nanjing, Yangtze River Delta, China[J]. Atmospheric Environment, 2014,97:206-214.
[33]
Luna A S, Paredes M L L, Oliveira G C G D, et al. Prediction of ozone concentration in tropospheric levels using artificial neural networks and support vector machine at Rio de Janeiro, Brazil[J]. Atmospheric Environment, 2014,98:98-104.
[34]
安俊琳,朱彬,李用宇.南京北郊大气VOCs体积分数变化特征[J]. 环境科学, 2013,34(12):4504-4512. An J L, Zhu B, Li Y Y. Variation characteristics of ambient volatile organic compounds (VOCs) in Nanjing northern suburb, China[J]. Environmental Science, 2013,34(12):4504-4512.
[35]
Daubechies I. Ten Lectures on Wavelets[M]. Philadelphia:SIAM Press, 1992.
[36]
Liu S J, Hu Y W, Li C, et al. Machinery condition prediction based on wavelet and support vector machine[J]. Journal of Intelligent Manufacturing, 2017,28(4):1045-1055.
[37]
Siwek K, Osowski S. Improving the accuracy of prediction of PM10 pollution by the wavelet transformation and an ensemble of neural predictors[J]. Engineering Applications of Artificial Intelligence, 2012,25(6):1246-1258.
[38]
He J J, Gong S L, Yu Y, et al. Air pollution characteristics and their relation to meteorological conditions during 2014~2015 in major Chinese cities[J]. Environmental Pollution, 2017,223:484-496.
[39]
Vapnik V N. The Nature of Statistic Learning Theory[M]. New York:Springer, 2000:17-34.
[40]
Feng X, Li Q, Zhu Y J, et al. Artificial neural networks forecasting of PM2.5 pollution using air mass trajectory based geographic model and wavelet transformation[J]. Atmospheric Environment, 2015,107:118-128.
[41]
齐冰,牛彧文,杜荣光,等.杭州市近地面大气臭氧浓度变化特征分析[J]. 中国环境科学, 2017,37(2):443-451. Qi B, Niu Y W, Du R G, et al. Characteristics of surface ozone concentration in urban site of Hangzhou[J]. China Environmental Science, 2017,37(2):443-451.
[42]
赵辉,郑有飞,吴晓云,等.江苏省大气复合污染特征与相关气象驱动[J]. 中国环境科学, 2018,38(8):2830-2839. Zhao H, Zheng Y F, Wu X Y, et al. Atmospheric compound pollution characteristics and the effects of meteorological factors in Jiangsu Province[J]. China Environmental Science, 2018,38(8):2830-2839.
[43]
Chelani A B. Prediction of daily maximum ground ozone concentration using support vector machine[J]. Environmental Monitoring and Assessment, 2010,162(1-4):169-176.
[44]
林旭,朱彬,安俊琳,等.南京北郊VOCs对臭氧和二次有机气溶胶潜在贡献的研究[J]. 中国环境科学, 2015,35(4):976-986. Lin X, Zhu B, An J L, et al. Potential contribution of secondary organic aerosols and ozone of VOCs in the northern suburb of Nanjing[J]. China Environmental Science, 2015,35(4):976-986.
[45]
杨笑笑,汤莉莉,张运江,等.南京夏季市区VOCs特征及O3生成潜势的相关性分析[J]. 环境科学, 2016,37(2):443-451. Yang X X, Tang L L, Zhang Y J, et al. Correlation analysis between characteristics of VOCs and ozone formation potential in summer in Nanjing urban district[J]. Environmental Science, 2016,37(2):443-451.
[46]
张玉欣,安俊琳,王俊秀,等.南京工业区挥发性有机物来源解析及其对臭氧贡献评估[J]. 环境科学, 2018,39(2):502-510. Zhang Y X, An J L, Wang J X, et al. Source analysis of volatile organic compounds in the Nanjing industrial area and evaluation of their contribution to ozone[J]. Environmental Science, 2018,39(2):502-510.
[47]
王占山,李云婷,陈添,等.北京城区臭氧日变化特征及与前体物的相关性分析[J]. 中国环境科学, 2014,34(12):3001-3008. Wang Z S, Li Y T, Chen T, et al. Analysis on diurnal variation characteristics of ozone and correlations with its precursors in urban atmosphere of Beijing[J]. China Environmental Science, 2014,34(12):3001-3008.
