Development of an automated photolysis rates prediction system based on machine learning  

作　　者：Weijun Pan Sunling Gong Huabing Ke Xin Li Duohong Chen Cheng Huang Danlin Song 

机构地区：[1]State Key Laboratory of Severe Weather&Key Laboratory of Atmospheric Chemistry of CMA,Chinese Academy of Meteorological Sciences,Beijing 100081,China [2]College of Environmental Sciences and Engineering,Peking University,Beijing 100871,China [3]State Environmental Key Laboratory of Reginal Air Quality Monitoring,Guangdong Ecological Environmental Monitoring Center,Guangzhou 510308,China [4]State Environmental Protection Key Laboratory of Formation and Prevention of the Urban Air Complex,Shanghai Academy of Environmental Sciences,Shanghai 200233,China [5]Chengdu Academy of Environmental Sciences,Chengdu 610072,China [6]National Observation and Research Station of Coastal Ecological Environments in Macao,Macao Environmental Research Institute,Macao University of Science and Technology,Macao 999078,China

出　　处：《Journal of Environmental Sciences》2025年第5期211-224,共14页环境科学学报(英文版)

基　　金：supported by the National Key Project of the Ministry of Science and Technology of China(No.2022YFC3701200);the National Natural Science Foundation of China(No.42090030).

摘　　要：Based on observed meteorological elements,photolysis rates(J-values)and pollutant concentrations,an automated J-values predicting system by machine learning(J-ML)has been developed to reproduce and predict the J-values of O^(1)D,NO_(2),HONO,H_(2)O_(2),HCHO,and NO_(3),which are the crucial values for the prediction of the atmospheric oxidation capacity(AOC)and secondary pollutant concentrations such as ozone(O_(3)),secondary organic aerosols(SOA).The J-ML can self-select the optimal“Model+Hyperparameters”without human interference.The evaluated results showed that the J-ML had a good performance to reproduce the J-values wheremost of the correlation(R)coefficients exceed 0.93 and the accuracy(P)values are in the range of 0.68-0.83,comparing with the J-values from observations and from the tropospheric ultraviolet and visible(TUV)radiation model in Beijing,Chengdu,Guangzhou and Shanghai,China.The hourly prediction was also well performed with R from 0.78 to 0.81 for next 3-days and from 0.69 to 0.71 for next 7-days,respectively.Compared with O_(3)concentrations by using J-values from the TUV model,an emission-driven observation-based model(e-OBM)by using the J-values from the J-ML showed a 4%-12%increase in R and 4%-30%decrease in ME,indicating that the J-ML could be used as an excellent supplement to traditional numerical models.The feature importance analysis concluded that the key influential parameter was the surface solar downwards radiation for all J-values,and the other dominant factors for all J-values were 2-m mean temperature,O_(3),total cloud cover,boundary layer height,relative humidity and surface pressure.

关 键 词：J-values Automated prediction system Machine learning Short-term prediction O_(3)simulated improvement 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]