BP神经网络的近地面臭氧估算及时空特征分析  

BP neural network for near-surface ozone estimation and spatial and temporal characteristics analysis

作　　者：李紫微 马庆勋[1] 吕杰[1] LI Ziwei;MA Qingxun;LüJie(Xi’an University of Science and Technology,Xi'an 710054,China)

机构地区：[1]西安科技大学,陕西西安710054

出　　处：《测绘通报》2021年第6期28-32,共5页Bulletin of Surveying and Mapping

基　　金：国家自然科学基金(41674013,41874012)。

摘　　要：近年来,中国东部城市光化学烟雾污染频繁发生,臭氧(O3)作为光化学烟雾的标志性污染物日益成为影响城市或区域大气的首要污染物。为探究京津唐地区近地面臭氧污染特征及时空变化趋势,本文基于前馈(BP)神经网络,结合地面监测站点臭氧浓度数据、卫星遥感臭氧柱总量数据与气象站点气象要素数据的非线性关系建立近地面臭氧浓度反演模型,并对近地面臭氧时空分布进行分析。结果表明:评价模型可靠性的决定系数R2为0.888、RMSE为10.742、MAE为9.596,建立的BP神经网络模型精度较高;2016—2019年京津唐臭氧年平均浓度呈现增加趋势;四季中,京津唐夏季臭氧浓度最大,冬季最小。研究结果为近地面臭氧估算提供了技术参考,同时对环境监测具有重要的现实指导意义。In recent years,photochemical smog pollution frequently occurs in the eastern cities of China.O3,as a landmark pollutant of photochemical smog,is increasingly becoming the primary pollutant affecting urban or regional air pollution.To explore the Beijing-Tianjin-Tangshan(BTT)region surface ozone pollution characteristic and changing trend of time and space,this study based on feedforward(back propagation,BP)neural network,an inversion model of surface ozone concentration was established combined with the nonlinear relationship between the ozone concentration data of ground monitoring stations and the total ozone column data of satellite and meteorological data of meteorological stations.The results show that the R2,RMSE,and MAE of the reliability evaluation model are 0.888,10.742,and 9.596,respectively,and the accuracy of the established neural network model is relatively high.From 2016 to 2019,the annual average ozone concentration in the Beijing-Tianjin-Tangshan region showed an increasing trend.In the four seasons,ozone concentration in Beijing,Tianjin,and Tangshan are the highest in summer and the lowest in winter.The results provided a technical reference for near-surface ozone estimation and have important practical significance for environmental monitoring.

关 键 词：京津唐地区 BP神经网络 臭氧反演 时空特征分析 臭氧浓度变化趋势 

分 类 号：P237[天文地球—摄影测量与遥感]