基于距离相关系数和支持向量机回归的PM2.5浓度滚动统计预报方案

针对目前空气质量统计预报方法存在的主要缺陷,本文提出了距离相关系数和支持向量机回归相结合的统计预报方案DC-SVR.利用淮安市2013年1—12月PM_(2.5)观测资料和常规气象观测资料,首先在选入预报当日气象要素的基础上,增加选取前期污染物和气象要素作为预报因子,再采用距离相关系数分季节从预报因子中筛选出重要预报因子,最后采用支持向量机回归对PM_(2.5)浓度值进行逐日滚动统计预报.研究发现,淮安地区气温和气压对PM_(2.5)的距离相关性要高于其他气象要素,夏秋季PM_(2.5)与气象要素的距离相关性较春冬季好.基于距离相关系数和支持向量机回归建立DC-SVR模型,PM_(2.5)的试预报值和实测值的全年相关系数高达0.76,平均偏差仅为1.13μg·m~(-3),平均绝对误差为23.47μg·m~(-3).通过与支持向量机回归、人工神经网络的统计预报效果对比,DC-SVR模型有效降低预报因子维数且能自适应选取最佳参数,预报精度显著优于其他3种统计预报方案,可为业务化预报提供参考.     

Time and cost characterisation of emissions from non-road diesel equipment for infrastructure projects

The construction process contributes to pollutant emissions, particularly through the operation of diesel- and gasoline-powered equipment. In the past decade, a series of investigations were undertaken to quantify these emissions for a variety of non-road construction equipment performing different activities and undergoing different duty cycles, and a model to estimate quantities of six types of pollutant was developed. This paper uses that model to estimate emissions for four street and utility construction projects which no one has done previously. We combined information from company records with standard construction industry manuals to estimate total emissions for the projects and to examine the pollution patterns and magnitudes. The street construction projects all had similar emission profiles with a large peak at the beginning and a steady output of emissions throughout the duration of the project. For example, in two of the projects studied, half of all CO₂ emissions were produced before the projects were 40% completed. Results showed that demolition and earthwork are the activities with the largest contribution. The equipment types with the largest contribution are backhoes, front-end loaders, bulldozers and trenchers. Trenchers, for example, produced 30% of all emissions on the projects on which they were used.     

基于WebGIS和粒子系统的空气污染预测动态可视化展示技术

在完成国家"863"项目"重点城市群大气复合污染综合防治技术与集成示范"(3c-star)的过程中,我们研发了具有自主知识产权的"区域大气环境综合管理和辅助决策系统",其中创造性地利用NetCDF、Silverlight、WebGIS、粒子系统、多媒体等技术实现了空气污染预测动态可视化数据展示。本文阐述了工作的背景、技术要点及成果。     

Neural network forecasting of air pollutants hourly concentrations using optimised temporal averages of meteorological variables and pollutant concentrations

The new method for the forecasting hourly concentrations of air pollutants is presented in the paper. The method was developed for a site in urban residential area in city of Zagreb, Croatia, for four air pollutants (NO₂, O₃, CO and PM₁₀). Meteorological variables and concentrations of the respective pollutant were taken as predictors. A novel approach, based on families of univariate regression models, was employed in selecting the averaging intervals for input variables. For each variable and each averaging period between 1 and 97 h, a separate model was built. By inspecting values of the coefficient of correlation between measured and modelled concentrations, optimal averaging periods for each variable were selected. A new dataset for building the forecasting model was then calculated as temporal moving averages (running means) of former variables. A multi-layer perceptron type of neural networks is used as the forecasting model. Index of agreement, calculated for the entire dataset including the data for model building, ranged from 0.91 to 0.97 for the respective pollutants. As suggested by the analysis of the relative importance of the input variables, different agreements for different pollutants are likely due to different sources and production mechanisms of investigated pollutants. A comparison of the new method with more traditional method, which takes hourly averages of the forecast hour as input variables, showed similar or better performance. The model was developed for the purpose of public-health-oriented air quality forecasting, aiming to use a numerical weather forecast model for the prediction of the part of input data yet unknown at the forecasting time. It is to expect that longer term averages used as inputs in the proposed method will contribute to smaller input errors and the greater accuracy of the model.     

Comparison of the Performance of Statistical Models in Forecasting Monthly Total Dissolved Solids in the Rio Grande1

Abudu, S., J.P. King, Z. Sheng, 2011. Comparison of the Performance of Statistical Models in Forecasting Monthly Total Dissolved Solids in the Rio Grande. Journal of the American Water Resources Association (JAWRA) 48(1): 10‐23. DOI: 10.1111/j.1752‐1688.2011.00587.x Abstract: This paper presents the application of autoregressive integrated moving average (ARIMA), transfer function‐noise (TFN), and artificial neural networks (ANNs) modeling approaches in forecasting monthly total dissolved solids (TDS) of water in the Rio Grande at El Paso, Texas. Predictability analysis was performed between the precipitation, temperature, streamflow rates at the site, releases from upstream reservoirs, and monthly TDS using cross‐correlation statistical tests. The chi‐square test results indicated that the average monthly temperature and precipitation did not show significant predictability on monthly TDS series. The performances of one‐ to three‐month‐ahead model forecasts for the testing period of 1984‐1994 showed that the TFN model that incorporated the streamflow rates at the site and Caballo Reservoir release improved monthly TDS forecasts slightly better than the ARIMA models. Except for one‐month‐ahead forecasts, the ANN models using the streamflow rates at the site as inputs resulted in no significant improvements over the TFN models at two‐month‐ahead and three‐month‐ahead forecasts. For three‐month‐ahead forecasts, the simple ARIMA showed similar performance compared to all other models. The results of this study suggested that simple deseasonalized ARIMA models could be used in one‐ to three‐month‐ahead TDS forecasting at the study site with a simple, explicit model structure and similar model performance as the TFN and ANN models for better water management in the Basin.     

REAL-TIME SNOW SIMULATION MODEL FOR THE MONONGAHELA RIVER BASIN1

ABSTRACT: The Pittsburgh District, U.S. Army Corps of Engineers, is responsible for operating two multipurpose reservoirs in the 7384 square mile (19198 square kilometer) Monongahela Basin. A third reservoir, presently under construction, will soon be operating. The real-time forecasting of runoff for operational purposes requires simulation of snow accumulation and snowmelt throughout the Basin during the winter season. This article describes capabilities of SNOSIM, a model being developed for performing such simulation. The application of this model as part of a comprehensive system of water control software, and some initial simulation results are presented.     

AN EVALUATION OF WEEKLY AND MONTHLY TIME SERIES FORECASTS OF MUNICIPAL WATER USE1

ABSTRACT: A cascade model for forecasting municipal water use one week or one month ahead, conditioned on rainfall estimates, is presented and evaluated. The model comprises four components: long term trend, seasonal cycle, autocorrelation and correlation with rainfall. The increased forecast accuracy obtained by the addition of each component is evaluated. The City of Deerfield Beach, Florida, is used as the application example with the calibration period from 1976–1980 and the forecast period the drought year of 1981. Forecast accuracy is measured by the average absolute relative error (AARE, the average absolute value of the difference between actual and forecasted use, divided by the actual use). A benchmark forecast is calculated by assuming that water use for a given week or month in 1981 is the same as the average for the corresponding period from 1976 to 1980. This method produces an AARE of 14.6 percent for one step ahead forecasts of monthly data and 15.8 percent for weekly data. A cascade model using trend, seasonality and autocorrelation produces forecasts with AARE of about 12 percent for both monthly and weekly data while adding a linear relationship of water use and rainfall reduces the AARE to 8 percent in both cases if it is assumed that rainfall is known during the forecast period. Simple rainfall predictions do not increase the forecast accuracy for water use so the major utility of relating water use and rainfall lies in forecasting various possible water use sequences conditioned on sequences of historical rainfall data.     

SHORT-TERM FORECASTING OF SNOWMELT RUNOFF USING ARMAX MODELS1

ABSTRACT: Time series models of the ARMAX class were investigated for use in forecasting daily riverflow resulting from combined snowmelt/rainfall. The Snowmelt Runoff Model (Martinec-Rango Model) is shown to have a form similar to the ARMAX model. The advantage of the ARMAX approach is that analytical model identification and parameter estimation techniques are available. In addition, previous forecast errors can be included to improve forecasts and confidence limits can be estimated for the forecasts. Diagnostic checks are available to determine if the model is performing properly. Finally, Kalman filtering can be used to allow the model parameters to vary continuously to reflect changing basin runoff conditions. The above advantages result in improved flow forecasts with fewer model parameters.     

FLOOD STAGE FORECASTING WITH SUPPORT VECTOR MACHINES1

ABSTRACT: Machine learning techniques are finding more and more applications in the field of forecasting. A novel regression technique, called Support Vector Machine (SVM), based on the statistical learning theory is explored in this study. SVM is based on the principle of Structural Risk Minimization as opposed to the principle of Empirical Risk Minimization espoused by conventional regression techniques. The flood data at Dhaka, Bangladesh, are used in this study to demonstrate the forecasting capabilities of SVM. The result is compared with that of Artificial Neural Network (ANN) based model for one‐lead day to seven‐lead day forecasting. The improvements in maximum predicted water level errors by SVM over ANN for four‐lead day to seven‐lead day are 9.6 cm, 22.6 cm, 4.9 cm and 15.7 cm, respectively. The result shows that the prediction accuracy of SVM is at least as good as and in some cases (particularly at higher lead days) actually better than that of ANN, yet it offers advantages over many of the limitations of ANN, for example in arriving at ANN's optimal network architecture and choosing useful training set. Thus, SVM appears to be a very promising prediction tool.     

PROBABILISTIC GREAT LAKES HYDROLOGY OUTLOOKS1

ABSTRACT: The Great Lakes Environmental Research Laboratory developed a semiautomatic software package for making hydrological outlooks for the Great Lakes. These include basin moisture storages, basin runoff, lake heat storage, lake evaporation, heat fluxes, and net lake supplies, one or more full months into the future. The package combines GLERL's rainfall-runoff and lake evaporation models with near real-time data reduction techniques to represent current system states. Users select historical meteorologic record segments as candidate future scenarios to generate deterministic near real-time hydrological outlooks. GLERL has extended the package to make probabilistic outlooks for a decision-maker who must estimate the risk associated with his decisions. GLERL matches National Weather Service meteorologic outlook probabilities by selecting groups of historical meteorologic sequences, and constructs embedded outlook intervals for each hydrologic variable of interest. Interval probabilities are assigned from comparisons over a recent evaluation period. This physically-based approach for generating outlooks offers the ability, as compared to other statistically-based approaches, to incorporate improvements in the understanding, of process dynamics as they occur in the future and to respond reasonably to conditions initial to a forecast (such as heat and moisture storages), not observed in the past.