知识与财富的链接
基于CMAQ模型结果,利用高维克里金插值算法,建立了排放控制因子与污染物环境浓度的响应面模型(RSM),实现了大气污染可控源排放与复合污染水平的实时函数响应.研究结果显示,RSM对PM2.5的预测结果与CMAQ实际模拟结果的误差在容许范围内(最大误差小于0.20μg.m-3,3.89%).基于所建立的RSM,开发了RSM-VAT区域大气污染控制可视化辅助决策工具.使用RSM-VAT对美国8个典型城市的PM2.5污染状况进行了控制情景分析,通过"可视化展示"和"图表分析"二大模块,输出不同控制情景下的环境污染物浓度的实时响应图、可视化展示和数据分析图表等结果.
Using CMAQ simulation results obtained from a multivariate design of experiments, a response surface model (RSM) describing the relationship between air pollution and emission control factors is built using the high dimensional Kriging Interpolation Algorithm. The RSM, coupled with a visualization/decision support tool (RSM-VAT), facilitates real-time visualization of 3-D air quality model data under a wide variety of emission control scenarios and supports policy making through a user-friendly graphical interface linking emission control measures to the concentrations of multiple air pollutants. Verification of RSM prediction against CMAQ simulation results of PM2.5 shows acceptable model performance of RSM (deviation of < 0.20 mg·m-3 or < 3.89%). Using the developed RSM-VAT, a case study predicting PM2.5 concentration changes corresponding to different emission control scenarios in eight US metropolitan areas is demonstrated.
