Atmospheric particulate matter pollution has attracted much wider attention globally.In recent years,the development of atmospheric particle collection techniques has put for-wards new demands on the real-time source apportionments techniques.Such demands are summarized,in this paper,as how to set up new restraints in apportionment and how to develop a non-linear regression model to process complicated circumstances,such as the existence of secondary source and similar source.In this study,we firstly analyze the pos-sible and potential restraints in single particle source apportionment,then propose a novel three-step self-feedback long short-term memory(SF-LSTM)network for approximating the source contribution.The proposed deep learning neural network includes three modules,as generation,scoring and refining,and regeneration modules.Benefited from the scoring modules,SF-LSTM implants four loss functions representing four restraints to be followed in the apportionment,meanwhile,the regeneration module calculates the source contribu-tion in a non-linear way.The results show that the model outperforms the conventional regression methods in the overall performance of the four evaluation indicators(resid-ual sum of squares,stability,sparsity,negativity)for the restraints.Additionally,in short time-resolution analyzing,SF-LSTM provides better results under the restraint of stability.

Wei Wang;Weiman Xu;Shuai Deng;Yimeng Chai;Ruoyu Ma;Guoliang Shi;Bo Xu;Mei Li;Yue Li

Trusted Al System Laboratory,College of Computer Science,Nankai University,Tianjin 300350,China;KLMDASR,Tianjin Key Laboratory of Network and Data Security Technology,Tianjin 300350,China;State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control,College of Environmental Science and Engineering,Nankai University,Tianjin 300071,China;Institute of Mass Spectrometry and Atmospheric Environment,Guangdong Provincial Engineering Research Center for on-line source apportionment system of air pollution Jinan University,Guangzhou 510632,China;Guangdong-Hongkong-Macau Joint Laboratory of Collaborative Innovation for Environmental Quality,Guangzhou 510632,China

环境科学学报（英文版）

[1]Wei Wang;Weiman Xu;Shuai Deng;Yimeng Chai;Ruoyu Ma;Guoliang Shi;Bo Xu;Mei Li;Yue Li-.Self-feedback LSTM regression model for real-time particle source apportionment)[J].环境科学学报（英文版）,2022(04):10-20

apportionment,apportionments

Self,feedback,regression,model,real,particle,source,Atmospheric,particulate,matter,pollution,has,attracted,much,wider,attention,globally,In,recent,years,development,atmospheric,collection,techniques,put,wards,new,demands,Such,summarized,this,paper,set,up,restraints,linear,process,complicated,circumstances,such,existence,secondary,similar,study,firstly,analyze,sible,potential,single,then,novel,three,step,self,long,short,term,memory,SF,network,approximating,contribution,proposed,deep,learning,neural,includes,modules,scoring,refining,regeneration,Benefited,from,implants,four,loss,functions,representing,followed,meanwhile,calculates,way,results,show,that,outperforms,conventional,methods,overall,performance,evaluation,indicators,resid,ual,squares,stability,sparsity,negativity,Additionally,resolution,analyzing,provides,better,under

0.541369