基于自动机器学习集成模型的日NO_(2)模拟  

Daily NO_(2)Simulation Research Based on Automatic Machine Learning Ensemble Models

作　　者：卢凯凯 李净[1] 刘德仁[2] 徐发昭 张玉娜 朱世鑫 LU Kai-kai;LI Jing;LIU De-ren;XU Fa-zhao;ZHANG Yu-na;ZHU Shi-xing(College of Geography and Environmental Science,Northwest Normal University,Lanzhou 730070,China;College of Civil Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]西北师范大学地理与环境科学学院,兰州730070 [2]兰州交通大学土木工程学院,兰州730070

出　　处：《环境科学》2024年第10期5740-5747,共8页Environmental Science

基　　金：国家自然科学基金项目(42161058)。

摘　　要：为深入了解近地面NO_(2)的空间分布,以长江三角洲地区为研究区,利用NO_(2)站点实测数据和结合对流层观测仪(TROPOMI)的NO_(2)柱浓度数据,充分考虑人口、高程和气象因素对NO_(2)的影响,利用自动机器学习选取了模拟精度较高的5种机器学习算法:ET、RF、XGBoost、LightGBM和Catboost,并将这5种算法利用Stacking模型进行集成后对2020年3月至2021年2月长三角地区逐日NO_(2)浓度进行了模拟.结果表明,Stacking集成模型的RMSE和MAE值分别为7.078和5.270,其精度优于ET、RF、XGBoost、LightGBM和Catboost这5种单一的算法;长三角三省一市的NO_(2)高浓度值空间分布基本呈现以三省交汇处为中心口朝西南方向的U字形格局,以上海市、杭州市、南京市和合肥市为中心形成的城市群污染尤为显著,超过国家标准日限制的城市共27个,常州市是NO_(2)污染最严重的城市,NO_(2)浓度超标14 d,其次是上海市,超标13 d.NO_(2)浓度季节分布特点为:冬季>秋季>春季>夏季,其中夏季7月9日NO_(2)污染最轻,冬季12月23日NO_(2)污染最严重.To understand the spatial distribution of NO_(2)near the surface,we utilized measured data from NO_(2)monitoring stations and combined it with column concentration data from the Tropospheric Monitoring Instrument(TROPOMI),taking the Yangtze River Delta region as the study area.We considered the impact of factors such as population,elevation,and meteorological conditions on NO_(2)levels.We used automated machine learning to select five machine-learning algorithms with high simulation accuracy,namely ET,RF,XGBoost,LightGBM,and Catboost,and then integrated these five algorithms using the Stacking model to simulate the daily NO_(2)concentration in the Yangtze River Delta region from March 2020 to February 2021.The results indicated that the RMAE and MAE values of the Stacking ensemble model were 7.078 and 5.270,respectively,which outperformed the single algorithms of ET,RF,XGBoost,LightGBM,and Catboost.The spatial distribution of high NO_(2)concentrations in the Yangtze River Delta region,consisting of three provinces and one municipality,exhibited a U-shaped pattern with the convergence point located at the intersection of the three provinces,extending towards the southwest.Notably,urban pollution was particularly significant in the urban agglomerations centered around Shanghai,Hangzhou,Nanjing,and Hefei.There were 27 cities that exceeded the national standard daily limit.Changzhou was the city with the most serious NO_(2)pollution,with the NO_(2)concentration exceeding the standard for 14 d,followed by Shanghai,with 13 d.In terms of seasonal variation,the order of severity was as follows:winter,autumn,spring,and summer,with the least NO_(2)pollution occurring on July 9th during the summer,and the most severe NO_(2)pollution was observed on December 23rd during the winter.

关 键 词：自动机器学习 NO_(2) 集成模型 遥感 时空分布 

分 类 号：X51[环境科学与工程—环境工程]