杭州城市可吸入颗粒物污染与气象条件关系的分级评价和BP神经网络评估

为了解杭州城市环境空气质量与气象条件之间的关系,利用杭州市区2003-2007年的可吸入颗粒物(PM10)浓度数据和气象资料,通过分级评价的方法和基于BP神经网络的污染物浓度评估模型,得到PM10浓度与气象条件的对应关系.结果表明,随着日降水量的增大,PM10浓度减小;风速与PM10浓度呈明显的负相关,随着风速的增大,PM10浓度明显减小;气象因素与PM10浓度之间呈非线性关系,大气能见度对PM10和相对湿度的变化极为敏感.随着PM10浓度的增大,大气能见度迅速降低,相对湿度越高,大气能见度则越低;近几年杭州市气象条件不利于大气污染物的扩散和清洗,是杭州城市环境空气质量上升缓慢的主要原因之一.     

基于BP人工神经网络的城市PM2.5浓度空间预测

针对PM2.5日均质量浓度,采用BP人工神经网络模型,预测研究区空气中PM2.5浓度的空间变异,通过与普通克里格(Ordinary Kriging)插值方法对比验证BP人工神经网络预测模型的精度.结果表明:BP人工神经网络预测模型下研究区检验样本点位置的PM2.5仿真浓度与观测浓度之间的均方差、平均绝对误差、平均相对偏差和相关系数分别为0.296 μg2/m6、0.412 μg/m3、1.650％和0.851;而与此同时,普通克里格插值方法下的对应结果分别为1.041 μg2/m6、0.689 μg/m3、11.910％、0.638.研究成果在肯定BP人工神经网络预测模型可用于揭示PM2.5浓度空间变异特征的同时,也证实了其相对于普通克里格插值方法在固定空间点位准确预测PM2.5浓度方面的优势.     

基于改进型灰色神经网络组合模型的空气质量预测

基于空气质量数据不足及波动较大的情况,将灰色GM（1,1）模型与人工神经网络模型组合并改进,建立改进型灰色神经网络组合模型。利用天津市2001—2008年PM10、SO2和NO2年均值作为原始数据预测2009—2010年PM10、SO2和NO2的浓度以进行模型精度检验,最后利用该模型预测2011—2015年天津市空气质量状况。结果表明,与灰色GM（1,1）模型、传统灰色神经网络组合模型相比,所建立的改进型灰色神经网络组合模型相对模拟误差小,预测结果更为可靠,可以用于空气质量预测。     

深圳市区空气污染的人工神经网络预测

利用深圳市2006至2013年的大气污染物监测浓度数据和气象资料,分析深圳市空气质量的逐月分布变化特征。采用Pearson相关分析,选择显著相关因子,分别以BP神经网络和RBF神经网络构建空气质量预测模型,对该市2013年SO2、NO2、PM103种空气污染物的月均值进行预测。实验结果表明,通过Pearson相关分析建立的预测模型有更高的预报精度。BP和RBF 2种网络预测效果都比较理想,对不同污染物的预测精度各有高低。但BP网络的构建和参数优化过程较为复杂且网络训练结果不稳定,而RBF网络构建和训练简单,时间短而结果稳定。在综合性能上,RBF网络用于环境空气污染物浓度的预测具有更强的适用性。     

山东省空气质量预报平台设计及其预报效果评估

基于排放源处理(SMOKE)模型、综合空气质量(CMAQ)模型与气象预报(WRF)模型初步搭建山东省空气质量预报平台,利用污染物在线监测数据和气象站观测数据检验预报平台的预报效果。结果表明,预报平台气象模块的预测效果与文献研究结果较一致;由CMAQ模型对2014年济南、淄博、烟台、威海的SO2、NO2、PM2.5质量浓度进行预测,SO2、NO2、PM2.5预报平均值分别在17.65~48.97、18.69~45.43、34.97~79.15μg/m3;SO2、NO2、PM2.5预报值与监测值的相关系数在0.52~0.74,标准化平均偏差、标准化平均误差、平均相对偏差、平均相对误差分别在-34.00%~-5.73%、11%~47%、-25.00%~-10.21%、20%~42%,预报平台具有良好的预报性能。最后,对未来空气质量数值预报平台的发展提出建议。     

NO模拟烟气水净化实验研究

通过模拟实验研究了在NO=120～480mL／Nm^3低浓度和含氧量5．5％、6．0％和20％体积浓度及常压与≤36℃条件下，以纯水净化难溶有害污染成分NO构成的模拟烟气。在相同的NO配气体积浓度10％，NO被净化吸收量随其流量的增加而增加，而吸收率η^LENO却下降了。常压大气中吸收率值为5．6％～20．6％；5．5％ O2时的吸收率值为1．3％～4．6％；添加少量氧化剂使O2达6．0％时的吸收率值为1．3％～16．7％。     

南宁市大气颗粒物TSP、PM10、PM2.5污染水平研究

2002年在南宁市的5个典型城市功能区内,共采集了125个大气样品(按季节分别采集),初步调查了大气中颗粒物TSP、PM10、PM2.5的污染状况。结果表明,南宁市TSP、PM10、PM2.5的污染很严重,超标率分别为67．5％、82．5％、92．5％,对人体健康危害更大的PM2．5占到了PM10的63．5％左右。重污染区PM2.5的浓度超过轻污染区近一倍。     

BP神经网络对蚯蚓滤池处理COD的模拟预测

基于蚯蚓滤池处理去除污染物的非线性特点,利用BP神经网络建立了蚯蚓滤池处理COD的基本模型结构。同时对实验数据进行了验证和预测,通过权值贡献率分析确定了各种输入因素对COD出水浓度的影响。结果表明:COD的出水模型预测值与实际值平均误差较小,模型稳定,预测效果好。输入神经元为4,隐含神经元为8,输出神经元为1,学习速率为0.1,动量为0.1,训练次数为10 000的BP神经网络模型,预测的COD出水值最接近真实值。COD进水浓度对COD出水影响最大,符合理论研究结果。BP神经网络模型建立的成功为后续生活污水智能化控制的研究提供了相应的理论基础。     

气相色谱法测定空气中的1，2-二氯乙烷

建立了一种用气相色谱法测定空气中1，2-二氯乙烷的方法。空气中1，2-二氯乙烷用活性炭吸附，二硫化碳解析，直接进样分析。其回收率为97．6％～100．8％，最低检出浓度为0．005mg／m^3，相对标准偏差为0．7％-2．4％；该方法具有前处理简单、速度快、干扰少、灵敏度高等特点，可用于空气中1，2-二氯乙烷的测定。     

固相萃取-气相色谱/质谱法测定水中多环芳烃

建立了固相萃取-气相色谱／质谱联用测定水中多环芳烃(PAHs)的分析方法．优化了固相萃取条件。结果表明,固相萃取效率高、萃取时间短,采用MS的选择离子检测方式对实际水样中PAHs进行定性定量分析,平均回收率在80．4％～115％之间,相对标准偏差为7．03％～18．5％,方法的检出限在0．010～0．020μg／L之间。通过实际样品中PAHs的分析表明,该法快速,溶剂用量少,能满足痕量分析的要求。     

Carbon monoxide concentration forecasting in Santiago, Chile

In the city of Santiago, Chile, air quality is defined in terms of particulate matter with an aerodynamic diameter < or = 10 microm (PM10) concentrations. An air quality forecasting model based on past concentrations of PM10 and meteorological conditions currently is used by the metropolitan agency for the environment, which allows restrictions to emissions to be imposed in advance. This model, however, fails to forecast between 40 and 50% of the days considered to be harmful for the inhabitants every year. Given that a high correlation between particulate matter and carbon monoxide (CO) concentrations is observed at monitoring stations in the city, a model for CO concentration forecasting would be a useful tool to complement information about expected air quality in the city. Here, the results of a neural network-based model aimed to forecast maximum values of the 8-hr moving average of CO concentrations for the next day are presented. Forecasts from the neural network model are compared with those produced with linear regressions. The neural network model seems to leave more room to adjust free parameters with 1-yr data to predict the following year's values. We have worked with 3 yr of data measured at the monitoring station located in the zone with the worst air quality in the city of Santiago, Chile.     

Model for forecasting expressway fine particulate matter and carbon monoxide concentration: application of regression and neural network models

The Borman Expressway is a heavily traveled 16-mi segment of the Interstate 80/94 freeway through Northwestern Indiana. The Lake and Porter counties through which this expressway passes are designated as particulate matter < 2.5 microm (PM2.5) and ozone 8-hr standard nonattainment areas. The Purdue University air quality group has been collecting PM2.5, carbon monoxide (CO), wind speed, wind direction, pressure, and temperature data since September 1999. In this work, regression and neural network models were developed for forecasting hourly PM2.5 and CO concentrations. Time series of PM2.5 and CO concentrations, traffic data, and meteorological parameters were used for developing the neural network and regression models. The models were compared using a number of statistical quality indicators. Both models had reasonable accuracy in predicting hourly PM2.5 concentration with coefficient of determination -0.80, root mean square error (RMSE) <4 microg/m3, and index of agreement (IA) > 0.90. For CO prediction, both models showed moderate forecasting performance with a coefficient of determination -0.55, RMSE < 0.50 ppm, and IA -0.85. These models are computationally less cumbersome and require less number of predictors as compared with the deterministic models. The availability of real time PM2.5 and CO forecasts will help highway managers to identify air pollution episodic events beforehand and to determine mitigation strategies.     

Prediction of maximum of 24-h average of PM10 concentrations 30 h in advance in Santiago,Chile

We have developed a neural network based model that uses values of PM10 concentrations measured until 6 p.m. on the present day plus measured and forecasted values of meteorological variables as input in order to predict the level reached by the maximum of the 24-h moving average (24MA) of PM10 concentration on the next day. We have adjusted the parameters of the model using 1998 data to predict 1999 conditions and 1999 data to forecast 2000 maximum concentrations. We have found that among the relevant meteorological input variables, the forecasted difference between maximum and minimum temperature is the most important. Due to the fact that local authorities impose restrictions to emissions on days when the maximum of 24MA of PM10 concentration is expected to exceed 240 μg/m³, we have corrected the measured concentrations on these days before evaluating the efficacy of the forecasting model. Percent errors in forecasting the numerical value are of the order of 20%. The performance of the neural network is better than that of a linear model with the same inputs, but the difference being not great is an indication that the right choice of input variables may be more important than the decision to use a linear or a nonlinear model.     

Neural network and multiple regression models for PM10 prediction in Athens: a comparative assessment

Particulate atmospheric pollution in urban areas is considered to have significant impact on human health. Therefore, the ability to make accurate predictions of particulate ambient concentrations is important to improve public awareness and air quality management. This study examines the possibility of using neural network methods as tools for daily average particulate matter with aerodynamic diameter <10 microm (PM10) concentration forecasting, providing an alternative to statistical models widely used up to this day. Based on a data inventory, in a fixed central site in Athens, Greece, ranging over a two-year period, and using mainly meteorological variables as inputs, neural network models and multiple linear regression models were developed and evaluated. Comparison statistics used indicate that the neural network approach has an edge over regression models, expressed both in terms of prediction error (root mean square error values lower by 8.2-9.4%) and of episodic prediction ability (false alarm rate values lower by 7-13%). The results demonstrate that artificial neural networks (ANNs), if properly trained and formed, can provide adequate solutions to particulate pollution prognostic demands.     

基于因子分析法优选河流水体中氮浓度预测方法的研究

定量的河流水体中氮浓度预测方法有很多种,如何优选出预测精度较高的方法一直是学术界多年来致力于研究的重点。本研究采用因子分析法对预测方法的精度评价指标进行分析,并建立了预测方法精度的评价模型,对回归分析法、神经网络法、灰色系统法和增长率统计法4种水体氮浓度预测方法进行综合评估,优选出精度较高的河流水体氮浓度预测模型——BP神经网络预测模型。结果表明,此评估模型对类似研究具有一定的参考价值,能为选择出合适的河流水体氮浓度预测方法提供依据。     

Defining the photochemical contribution to particulate matter in urban areas using time-series analysis

The objective of this project is to demonstrate how the ambient air measurement record can be used to define the relationship between O3 (as a surrogate for photochemistry) and secondary particulate matter (PM) in urban air. The approach used is to develop a time-series transfer-function model describing the daily PM10 (PM with less than 10 microm aerodynamic diameter) concentration as a function of lagged PM and current and lagged O3, NO or NO2, CO, and SO2. Approximately 3 years of daily average PM10, daily maximum 8-hr average O3 and CO, daily 24-hr average SO2 and NO2, and daily 6:00 a.m.-9:00 a.m. average NO from the Aerometric Information Retrieval System (AIRS) air quality subsystem are used for this analysis. Urban areas modeled are Chicago, IL; Los Angeles, CA; Phoenix, AZ; Philadelphia, PA; Sacramento, CA; and Detroit, MI. Time-series analysis identified significant autocorrelation in the O3, PM10, NO, NO2, CO, and SO2 series. Cross correlations between PM10 (dependent variable) and gaseous pollutants (independent variables) show that all of the gases are significantly correlated with PM10 and that O3 is also significantly correlated lagged up to two previous days. Once a transfer-function model of current PM10 is defined for an urban location, the effect of an O3-control strategy on PM concentrations is estimated by calculating daily PM10 concentrations with reduced O3 concentrations. Forecasted summertime PM10 reductions resulting from a 5 percent decrease in ambient O3 range from 1.2 microg/m3 (3.03%) in Chicago to 3.9 microg/m3 (7.65%) in Phoenix.     

Analysis of a summertime PM2.5 and haze episode in the mid-Atlantic region

Observations of the mass and chemical composition of particles less than 2.5 microm in aerodynamic diameter (PM2.5), light extinction, and meteorology in the urban Baltimore-Washington corridor during July 1999 and July 2000 are presented and analyzed to study summertime haze formation in the mid-Atlantic region. The mass fraction of ammoniated sulfate (SO4(2-)) and carbonaceous material in PM2.5 were each approximately 50% for cleaner air (PM2.5< 10 microg/m3) but changed to approximately 60% and approximately 20%, respectively, for more polluted air (PM2.5>30 microg/m3). This signifies the role of SO4(2-) in haze formation. Comparisons of data from this study with the Interagency Monitoring of Protected Visual Environments network suggest that SO4(2-) is more regional than carbonaceous material and originates in part from upwind source regions. The light extinction coefficient is well correlated to PM2.5 mass plus water associated with inorganic salt, leading to a mass extinction efficiency of 7.6 +/- 1.7 m2/g for hydrated aerosol. The most serious haze episode occurring between July 15 and 19, 1999, was characterized by westerly transport and recirculation slowing removal of pollutants. At the peak of this episode, 1-hr PM2.5 concentration reached approximately 45 microg/m3, visual range dropped to approximately 5 km, and aerosol water likely contributed to approximately 40% of the light extinction coefficient.     

Neural Network and Multiple Regression Models for PM10 Prediction in Athens: A Comparative Assessment

Abstract

Particulate atmospheric pollution in urban areas is considered to have significant impact on human health. Therefore, the ability to make accurate predictions of particulate ambient concentrations is important to improve public awareness and air quality management. This study examines the possibility of using neural network methods as tools for daily average particulate matter with aerodynamic diameter <10 µm (PM₁₀) concentration forecasting, providing an alternative to statistical models widely used up to this day. Based on a data inventory, in a fixed central site in Athens, Greece, ranging over a two-year period, and using mainly meteorological variables as inputs, neural network models and multiple linear regression models were developed and evaluated. Comparison statistics used indicate that the neural network approach has an edge over regression models, expressed both in terms of prediction error (root mean square error values lower by 8.2–9.4%) and of episodic prediction ability (false alarm rate values lower by 7–13%). The results demonstrate that artificial neural networks (ANNs), if properly trained and formed, can provide adequate solutions to particulate pollution prognostic demands.