Atmospheric particulate matter pollution has attracted much wider attention globally.In recent years,the development of atmospheric particle collection techniques has put forwards new demands on the real-time source a...Atmospheric particulate matter pollution has attracted much wider attention globally.In recent years,the development of atmospheric particle collection techniques has put forwards 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 possible 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 contribution 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(residual 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.展开更多
In this study, we performed a highly time-resolved chemical characterization of nonrefractory submicron particles（NR-PM_1） in Beijing by using an Aerodyne high-resolution time-of-flight aerosol mass spectrometer（HR...In this study, we performed a highly time-resolved chemical characterization of nonrefractory submicron particles（NR-PM_1） in Beijing by using an Aerodyne high-resolution time-of-flight aerosol mass spectrometer（HR-ToF-AMS）. The results showed the average NR-PM_1 mass concentration to be 56.4 ± 58.0 μg/m^3, with a peak at 307.4 μg/m^3. Due to the high frequency of biomass burning in autumn, submicron particles significantly increased in organic content, which accounted for 51% of NR-PM_1 on average. Secondary inorganic aerosols（sulfate ＋ nitrate ＋ ammonium） accounted for 46% of NR-PM_1, of which sulfate,nitrate, and ammonium contributed 15%, 20%, and 11%, respectively. To determine the intrinsic relationships between the organic and inorganic species, we used the positive matrix factorization（PMF） model to merge the high-resolution mass spectra of the organic species and NO＋and NO_2~＋ions. The PMF analysis separated the mixed organic and nitrate（NO＋and NO_2~＋） spectra into four organic factors, including hydrocarbon-like organic aerosol（HOA）, oxygenated organic aerosol（OOA）, cooking organic aerosol（COA）, and biomass burning organic aerosol（BBOA）, as well as one nitrate inorganic aerosol（NIA） factor. COA（33%） and OOA（30%） contributed the most to the total organic aerosol（OA） mass, followed by BBOA（20%） and HOA（17%）. We successfully quantified the mass concentrations of the organic and inorganic nitrates by the NO＋and NO2＋ions signal in the organic and NIA factors. The organic nitrate mass varied from 0.01-6.8 μg/m^3, with an average of 1.0 ±1.1 μg/m^3, and organic nitrate components accounted for 10% of the total nitrate mass in this observation.展开更多
基金supported by Key Laboratory For Environmental Factors Control of Agro-product Quality Safety,Ministry of Agriculture and Rural Affairs(No.2018hjyzkfkt-002)Qian Xuesen Laboratory of Space Technology,CAST(No.GZZKFJJ2020002)National Research Program for Key Issues in Air Pollution Control(No.DQGG-05-30)
文摘Atmospheric particulate matter pollution has attracted much wider attention globally.In recent years,the development of atmospheric particle collection techniques has put forwards 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 possible 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 contribution 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(residual 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.
基金supported by“Strategic Priority Research Program”of the Chinese Academy of Sciences(No.XDB05020201)the Beijing Natural Science Foundation(No.8142034)
文摘In this study, we performed a highly time-resolved chemical characterization of nonrefractory submicron particles（NR-PM_1） in Beijing by using an Aerodyne high-resolution time-of-flight aerosol mass spectrometer（HR-ToF-AMS）. The results showed the average NR-PM_1 mass concentration to be 56.4 ± 58.0 μg/m^3, with a peak at 307.4 μg/m^3. Due to the high frequency of biomass burning in autumn, submicron particles significantly increased in organic content, which accounted for 51% of NR-PM_1 on average. Secondary inorganic aerosols（sulfate ＋ nitrate ＋ ammonium） accounted for 46% of NR-PM_1, of which sulfate,nitrate, and ammonium contributed 15%, 20%, and 11%, respectively. To determine the intrinsic relationships between the organic and inorganic species, we used the positive matrix factorization（PMF） model to merge the high-resolution mass spectra of the organic species and NO＋and NO_2~＋ions. The PMF analysis separated the mixed organic and nitrate（NO＋and NO_2~＋） spectra into four organic factors, including hydrocarbon-like organic aerosol（HOA）, oxygenated organic aerosol（OOA）, cooking organic aerosol（COA）, and biomass burning organic aerosol（BBOA）, as well as one nitrate inorganic aerosol（NIA） factor. COA（33%） and OOA（30%） contributed the most to the total organic aerosol（OA） mass, followed by BBOA（20%） and HOA（17%）. We successfully quantified the mass concentrations of the organic and inorganic nitrates by the NO＋and NO2＋ions signal in the organic and NIA factors. The organic nitrate mass varied from 0.01-6.8 μg/m^3, with an average of 1.0 ±1.1 μg/m^3, and organic nitrate components accounted for 10% of the total nitrate mass in this observation.
