CN114117893A - Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source - Google Patents
Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source Download PDFInfo
- Publication number
- CN114117893A CN114117893A CN202111310618.4A CN202111310618A CN114117893A CN 114117893 A CN114117893 A CN 114117893A CN 202111310618 A CN202111310618 A CN 202111310618A CN 114117893 A CN114117893 A CN 114117893A
- Authority
- CN
- China
- Prior art keywords
- dust
- source
- fall
- atmospheric
- pollution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000694 effects Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000000428 dust Substances 0.000 claims abstract description 155
- 239000011159 matrix material Substances 0.000 claims abstract description 55
- 239000000126 substance Substances 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 238000004364 calculation method Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000007637 random forest analysis Methods 0.000 claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 11
- 238000011835 investigation Methods 0.000 claims abstract description 10
- 238000000556 factor analysis Methods 0.000 claims abstract description 9
- 238000012821 model calculation Methods 0.000 claims abstract description 8
- 238000012216 screening Methods 0.000 claims abstract description 3
- 238000011160 research Methods 0.000 claims description 20
- 150000002500 ions Chemical class 0.000 claims description 18
- 238000001228 spectrum Methods 0.000 claims description 17
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 14
- 229910052799 carbon Inorganic materials 0.000 claims description 14
- 238000010276 construction Methods 0.000 claims description 10
- 229910052742 iron Inorganic materials 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 6
- 229910052782 aluminium Inorganic materials 0.000 claims description 5
- 229910052791 calcium Inorganic materials 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 229910052710 silicon Inorganic materials 0.000 claims description 5
- 229910052802 copper Inorganic materials 0.000 claims description 4
- 238000005265 energy consumption Methods 0.000 claims description 4
- 229910052748 manganese Inorganic materials 0.000 claims description 4
- 239000002245 particle Substances 0.000 claims description 4
- 229910052700 potassium Inorganic materials 0.000 claims description 4
- 229910052787 antimony Inorganic materials 0.000 claims description 3
- 229910052785 arsenic Inorganic materials 0.000 claims description 3
- 229910052797 bismuth Inorganic materials 0.000 claims description 3
- 229910052793 cadmium Inorganic materials 0.000 claims description 3
- 229910052804 chromium Inorganic materials 0.000 claims description 3
- 238000002485 combustion reaction Methods 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 229910052733 gallium Inorganic materials 0.000 claims description 3
- 229910001385 heavy metal Inorganic materials 0.000 claims description 3
- 229910052745 lead Inorganic materials 0.000 claims description 3
- 229910052759 nickel Inorganic materials 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 239000002689 soil Substances 0.000 claims description 3
- 229910052718 tin Inorganic materials 0.000 claims description 3
- 229910052720 vanadium Inorganic materials 0.000 claims description 3
- 229910052725 zinc Inorganic materials 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 2
- 239000004576 sand Substances 0.000 claims description 2
- 230000001629 suppression Effects 0.000 claims 1
- 230000002265 prevention Effects 0.000 abstract description 7
- 230000008878 coupling Effects 0.000 abstract description 4
- 238000010168 coupling process Methods 0.000 abstract description 4
- 238000005859 coupling reaction Methods 0.000 abstract description 4
- XEEYBQQBJWHFJM-UHFFFAOYSA-N iron Substances [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 12
- 238000012417 linear regression Methods 0.000 description 7
- 229910000831 Steel Inorganic materials 0.000 description 5
- 239000011575 calcium Substances 0.000 description 5
- 229910017053 inorganic salt Inorganic materials 0.000 description 5
- 239000010959 steel Substances 0.000 description 5
- 230000001419 dependent effect Effects 0.000 description 4
- 238000002790 cross-validation Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 239000012080 ambient air Substances 0.000 description 2
- 239000003245 coal Substances 0.000 description 2
- 238000011109 contamination Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 239000003570 air Substances 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 229910052749 magnesium Inorganic materials 0.000 description 1
- 239000013618 particulate matter Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000010517 secondary reaction Methods 0.000 description 1
- 229910052719 titanium Inorganic materials 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- Data Mining & Analysis (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Evolutionary Biology (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Sampling And Sample Adjustment (AREA)
Abstract
A method for analyzing an atmospheric dust-fall pollution source and evaluating the dust-fall marginal effect of the pollution source relates to the field of dust-fall pollution prevention and control, and comprises the following steps: step 1, data is input into the model. Constructing multi-component data by the dust fall quality and the component quality, and inputting the multi-component data into a positive definite matrix factor analysis model; and 2, setting model parameters. Investigating atmospheric dust source types, identifying main emission sources, and setting model calculation parameters; and 3, calculating a source analysis result. Analyzing a dust-fall pollution source by the model, and screening, judging and stretching results through investigation and expert experience; and 4, evaluating the marginal effect of the pollution source on the dust fall by utilizing a random forest algorithm according to the calculation result of the coupling model in the step 4. The invention is coupled with a receptor model and a machine learning model, estimates the contribution of a pollution source to a receptor and evaluates the marginal effect of the pollution source on the dust fall according to the chemical components of the dust fall. The source that can discern the dust fall fast and assess how the pollution source influences the dust fall, solve the pollution source of dust fall and be difficult to discern and the problem that the influence effect of pollution source to the dust fall is undefined, provide the decision-making support for dust fall pollution control, have very strong practical value and popularization and application prospect.
Description
Technical Field
The invention relates to the field of prevention and control of atmospheric dust fall pollution, in particular to a method for analyzing an atmospheric dust fall pollution source and evaluating a dust fall marginal effect of the pollution source by a coupling receptor model and a machine learning model, which is suitable for analyzing the atmospheric dust fall pollution source and evaluating the dust fall marginal effect of the pollution source.
Background
Atmospheric dustfall is particulate matter that falls naturally to the ground by gravity. The atmospheric dust fall is closely related to the urban air quality, can reflect the contribution of a primary emission source of particulate matters, and is an important index for reflecting the urban cleanliness and the fine management level. Meanwhile, the dust fall can also generate smaller particles, and the particles become carriers of various secondary reactions in the ambient air. The characteristics, sources and contributions of chemical components of the atmosphere dust fall are determined to be the key and the premise for managing and controlling the atmosphere dust fall. Nearly thirty years of dust fall monitoring is developed in China, but dust fall analysis still focuses on quantitative analysis, and researches on the aspects of space-time rules of elements such as dust fall chemical component characteristics and particle size distribution, quantitative source analysis of atmospheric dust fall, interrelation of pollution sources and dust fall and the like are lacked. The dust fall source analysis work can clearly give the source constitution of the dust fall, and the direction is indicated for the prevention and control work.
Pollution source analysis technical method mainly including active diffusion model and receptorThe model, receptor model, determines the contribution of the contamination source to the receptor by analysis of the chemical composition of the receptor and contamination source samples. Since the source strength does not need to be known and the meteorological data is not relied on, the receptor model is developed rapidly since the 70 s. The positive definite matrix factor model is an important method of a receptor model, the information of the quantity of the sources and the component spectrum is not needed to be known in advance, and the quantity of the sources, the component spectrum of the sources and the source contribution are calculated based on a large amount of data measured on the same receptor and analyzed. In 1997, Paater proposed an algorithm for positive definite matrix factor model. The positive definite matrix factor model is mainly applied to PM10And PM2.5The method has not been applied to atmospheric dustfall in terms of source analysis.
The atmospheric dust reduction in China mainly monitors the atmospheric dust reduction amount, and the source analysis of atmospheric dust reduction pollution and the relation between a pollution source and dust reduction are lack of deep and systematic research. Atmospheric dustfall is a direct embodiment of the fine management level of cities and is influenced by multiple factors such as weather, terrain, geographical position, city construction and city development. The dust-fall pollution source and contribution are difficult to rapidly and accurately resolve, and more innovations and breakthroughs are required to be searched on the concept and method. The invention utilizes the positive definite matrix factor model to analyze the atmospheric dust-fall pollution source, and uses the machine learning model to evaluate the marginal effect of the pollution source on atmospheric dust-fall, thereby solving the key scientific and technical problems in the atmospheric dust-fall pollution prevention and control process and providing basis and powerful technical support for dust-fall pollution prevention and control.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: providing a method for analyzing an atmospheric dust-fall pollution source and evaluating the dust-fall marginal effect of the pollution source; the invention constructs multi-component data by the atmosphere dust fall quality and the concentrations of chemical components such as elements, carbon, ions and the like, and inputs the multi-component data into a positive definite matrix factor analysis model. By carrying out investigation on atmospheric dust sources in a research area, main emission sources influencing atmospheric dust in the research area are identified, and model calculation parameters such as the number of factors participating in calculation, the number of chemical components, uncertainty of the chemical components, the number of output results and the like are set. Analyzing the pollution source of the atmospheric dust fall according to the set parameters, calculating a source analysis result, and screening, judging and stretching the result according to the survey result and expert experience to obtain a final model calculation result. And evaluating the marginal effect of the pollution source on the dust fall by utilizing a coupled receptor model result by utilizing a random forest algorithm and partial dependence calculation.
The invention provides a method for analyzing an atmospheric dust-fall pollution source and evaluating the dust-fall marginal effect of the pollution source, which adopts the following technical scheme:
and step 1, inputting the atmosphere dustfall multi-component data into a receptor model.
Establishing multi-component data by utilizing the atmospheric dust falling quality monitored by a dust falling sampling instrument and the concentrations of chemical components such as elements, carbon, ions and the like analyzed by an analyzer, and inputting the multi-component data into a positive definite matrix factor analysis model;
calculating the dust reduction amount of a research area by using a gravimetric method for measuring dust reduction of ambient air, analyzing the analysis of dust reduction chemical components (elements, ions and carbon), and measuring water-soluble ions by an ion chromatographic analyzer, wherein the water-soluble ions comprise NH4 +、Na+、K+、Ca2+、Mg2+、SO4 2-、NO3 -And Cl-(ii) a Carbon composition was measured by a semi-continuous OC/EC instrument, including OC and EC; elements are monitored by heavy metal analyzers, including Ca, Si, Al, Fe, K, V, Cr, Mn, Co, Ni, Cu, Zn, Ga, As, Se, Ag, Cd, Sn, Sb, Ba, Au, Hg, Tl, Pb, and Bi components.
And 2, setting parameters calculated by the model.
The method is characterized in that the method comprises the following steps of conducting investigation on atmospheric dust fall sources by means of basic data collection, such as source emission lists, and investigating emission conditions of dust fall sources such as soil dust, road dust, construction dust, storage yard dust, fixed combustion sources, industrial sources and motor vehicle exhaust emission sources in a research area on the spot, and on the basis of investigation on the atmospheric dust fall sources, identifying main emission sources influencing atmospheric dust fall in the research area by combining information such as social economy, energy consumption, urban construction, industrial layout, meteorological data and pollution source emission in the research area, and providing basis for analysis of the atmospheric dust fall sources. Setting the number of lines, columns, factors, the number of chemical components, the uncertainty of the chemical components, the number of output results and the like of the model participating in calculating data;
and 3, calculating a source analysis result.
Analyzing the pollution sources of the atmospheric dust fall according to the set parameters to obtain the number of the pollution sources, a factor spectrum matrix and a factor contribution matrix through analysis, optimizing the factor spectrum matrix and the factor contribution matrix through rotation calculation, identifying the factors into different source types according to chemical identification components in the factor spectrum matrix, and performing multiple linear regression by using the factor contribution matrix and the atmospheric dust fall amount to calculate the contribution of the pollution sources to the atmospheric dust fall pollution.
And 4, evaluating the marginal effect of the pollution source on the dust fall by using the coupled receptor model result through a random forest algorithm and partial dependence calculation.
And (3) forming a data set by taking the analysis result of the receptor model source as an independent variable and the atmospheric dust reduction amount as a dependent variable. And dividing a data set into a training set and a testing set through ten-fold cross validation, and building a machine learning model by using a random forest algorithm to predict the concentration of atmospheric dust fall. Meanwhile, the marginal effect of the pollution source on the dust fall is evaluated by utilizing partial dependence calculation on the basis of the model.
The positive definite matrix factor analysis model is a model which utilizes ion, element and carbon component data of atmospheric dust fall to extract a plurality of factors from a receptor matrix through positive definite matrix decomposition to obtain factor spectrum matrix information and factor contribution matrix information, identifies the factors into different source classes according to chemical identification components in the factor spectrum matrix, and utilizes the factor contribution matrix to calculate the contribution of each source class.
The random forest algorithm is an algorithm of a machine learning model, is a classifier comprising a plurality of decision trees, and predicts the atmosphere dust reduction amount, wherein the output category is determined by the mode of the category output by individual trees. And the partial dependence calculation shows the marginal effect of the pollution source on the prediction result of the random forest algorithm.
The invention has the technical effects that: the method for analyzing the atmospheric dust-fall pollution source and evaluating the dust-fall marginal effect of the pollution source can quickly and accurately identify the atmospheric dust-fall source and contribution, evaluates the dust-fall marginal effect of the pollution source, has strong practical value, has good popularization and application prospects, and indicates directions for prevention and control work. The invention provides a method for analyzing an atmospheric dust falling source and evaluating a marginal effect, provides a systematic technical guarantee for formulating an atmospheric dust falling pollution prevention and control strategy and improving the environmental quality, and can help a management department to quickly identify a pollution source and evaluate an influence effect when facing a dust falling problem, thereby effectively controlling.
Drawings
FIG. 1 is a flow chart of a method for analyzing an atmospheric dustfall pollution source and evaluating the dustfall marginal effect of the pollution source by a coupling receptor model and a machine learning model.
FIG. 2 is a diagram showing the analysis result of the atmospheric dust reduction source according to the present invention.
Fig. 3 is a graph of the marginal effect of different pollution sources on atmospheric dust fall according to the invention. Wherein: a contributes to the source (%) of the urban raise dust; b contributes (%) to the construction dust source; c contributes to the source of the steel industry (%); d contributes (%) to the coal source; e, contributing (%) to the source of the tail gas emission of the motor vehicle; f is the secondary inorganic salt source contribution (%).
Detailed Description
The atmospheric dustfall pollution source and the method for evaluating the dustfall marginal effect of the pollution source are explained in detail in the following by combining examples and attached drawings.
The invention provides a method for analyzing an atmospheric dust-fall pollution source and evaluating a dust-fall marginal effect of the pollution source, and the whole flow is shown in an attached figure 1. And constructing multi-component data by using the atmospheric dust fall quality and the concentrations of chemical components such as elements, carbon, ions and the like, and inputting the multi-component data into a positive definite matrix factor analysis model. By carrying out investigation on atmospheric dust source types of a research area, main emission sources influencing atmospheric dust in the research area are identified, and parameters of model calculation are set, wherein the parameters comprise the line number, the column number, the factor number, the chemical component number, the uncertainty of the chemical components, the output result number and the like of input data. Analyzing the pollution source of the atmospheric dust fall according to the set parameters, optimizing a factor spectrum matrix and a factor contribution matrix through rotary calculation, identifying the factors into different source classes according to chemical identification components in the factor spectrum matrix, and calculating the contribution of the pollution source to the atmospheric dust fall pollution by using the factor contribution matrix and the atmospheric dust fall amount to perform multiple linear regression. And (3) evaluating the marginal effect of the pollution source on the dust fall by utilizing a coupled receptor model result by utilizing a random forest algorithm and partial dependence calculation, establishing an atmospheric dust fall source analysis and marginal effect evaluation method, and providing a new way and method for quickly and accurately identifying the characteristics of atmospheric dust fall pollution chemical components, pollution sources and contributions, and the relationship between the pollution sources and the dust fall.
The method comprises the steps of firstly forming a multi-component data set by elements, carbon components and ion data of atmospheric dust fall, determining a research area according to research needs, carrying out dust fall pollution source investigation through methods such as basic information and field investigation, identifying main emission sources of the research area influencing atmospheric dust fall by combining social economy, energy consumption, city construction, industrial layout, meteorological data, pollution source emission and other information of the research area on the basis of the atmospheric dust fall source investigation, providing a basic basis for atmospheric dust fall source analysis, setting model calculation parameters, extracting a plurality of factors from a receptor matrix through a positive matrix factor model to obtain factor spectrum matrix information and factor contribution matrix information, identifying the factors into different source types according to chemical identification components in a factor spectrum matrix, and calculating the contribution of each source type by using the factor contribution matrix. The contribution of an atmospheric dust-fall pollution source is used as an independent variable, the atmospheric dust-fall amount is used as a dependent variable, a data set is divided into a training set and a testing set through cross validation by ten folds, a machine learning model is built by utilizing a random forest algorithm, and the concentration of atmospheric dust fall is predicted. Meanwhile, the marginal effect of the pollution source on the dust fall is evaluated by utilizing partial dependence calculation on the basis of the model.
The method for analyzing the atmospheric dust-fall pollution source and evaluating the dust-fall marginal effect of the pollution source by the coupling receptor model and the machine learning model comprises the following specific steps:
1, inputting atmospheric dust fall multi-component data into a receptor model;
and the monitoring points analyze the dust-settling quality and chemical components by arranging dust-settling cylinders. Wherein utilizes the ambient airDust fall determination gravimetric method analysis research area each point dust fall volume, dust fall chemical composition contains element, water-soluble ion, carbon component, and the element is monitored by heavy metal analyzer, including Ca, Si, Al, Fe, K, V, Cr, Mn, Co, Ni, Cu, Zn, Ga, As, Se, Ag, Cd, Sn, Sb, Ba, Au, Hg, Tl, Pb and Bi component. The water-soluble ions are measured by an ion chromatograph, including NH4 +、Na+、K+、Ca2+、Mg2+、SO4 2-、NO3 -And Cl-. Carbon composition was measured by a semi-continuous OC/EC instrument, including OC and EC; forming a multi-component data set by the dust reduction amount, elements, water-soluble ions and carbon components, and inputting the multi-component data set into a positive definite matrix factor analysis model; .
2 setting parameters of model calculation;
in the example, an atmospheric dust fall research area is Taiyuan city of Shanxi province, the class of atmospheric dust fall sources of the Taiyuan city is investigated, basic data of the Taiyuan city are collected through a source emission list, various statistical yearbooks, atmospheric monitoring station data and the like, the emission conditions of dust fall sources such as soil wind sand dust, road dust, construction dust, storage yard dust, fixed combustion sources, industrial sources, motor vehicle tail gas emission sources and the like of the Taiyuan city are investigated on site, a Taiyuan city atmospheric dust fall pollution source emission database and the like are established through data processing and analysis, and the main emission sources of the Taiyuan city, which affect the atmospheric dust fall, are identified on the basis of the atmospheric dust fall source investigation by combining the information such as the Taiyuan city social economy, energy consumption, city construction, industrial layout, meteorological data and pollution source emission and the like, so that a basis is provided for the atmospheric dust fall source analysis; the row number is 654 rows, the column number is 15 columns, the factor number is 6, and the repeated calculation times are 20 times;
3, calculating a source analysis result;
the model decomposes the matrix X into a factor spectrum matrix F and a source contribution matrix G, the factors are identified into different source types according to chemical identification components in the factor spectrum matrix F, and the source contribution matrix G and the dust reduction amount are utilized to carry out multiple linear regression to calculate the contribution of the pollution source to the atmosphere dust reduction pollution. The positive definite matrix factor model is a model which utilizes multi-component data to extract a plurality of factors from a receptor matrix through positive definite matrix decomposition to obtain a factor spectrum matrix X and a source contribution matrix G, identifies the factors into different source types according to chemical identification components in the factor spectrum matrix X, and utilizes the source contribution matrix G and dust reduction quantity to carry out multiple linear regression to calculate the contribution of a pollution source to atmospheric dust fall pollution. The multivariate linear regression is a regression comprising two or more independent variables, the dust reduction quantity is set as a dependent variable, the source contribution matrix is set as an independent variable, and the multivariate linear regression is carried out.
And 4, evaluating the marginal effect of the pollution source on the dust fall by utilizing a random forest algorithm and partial dependence calculation according to the result of the coupled receptor model.
And (3) forming a data set by taking the analysis result of the receptor model source as an independent variable and the atmospheric dust reduction amount as a dependent variable. And dividing a data set into a training set and a testing set through ten-fold cross validation, and building a machine learning model by using a random forest algorithm to predict the concentration of atmospheric dust fall. Meanwhile, the marginal effect of the pollution source on the dust fall is evaluated by utilizing partial dependence calculation on the basis of the model.
Continuously sampling from 11 months in 2019 to 6 months in 2021, monitoring the data with the time resolution of 1 month, and obtaining 654 pieces of receptor data containing Ca, Fe, Al, Si, Mg, K, Ti, Cu, Mn and Cl-、NO3 -、SO4 2-、NH4 +OC and EC are 15 components in total. The input model parameters are as follows: row number 654 rows, column number 15 columns, factor number 6, repeat count 20 times. According to the identified source identification component (urban dust source: Si, Al, Ca, building dust source: Ca, iron and steel industry source: Fe, coal source: OC, EC, SO)4 2-And the tail gas emission source of the motor vehicle is as follows: OC, EC, secondary inorganic salt source: SO (SO)4 2-、NH4 +、NO3 -) The most reasonable result in the repeated calculation is selected. And performing multivariate linear regression on the selected source contribution matrix and the dust reduction quantity to obtain the source concentration and the source contribution, wherein the source analysis result is shown in figure 2.
As can be seen from fig. 2, 6 pollution sources were analyzed by using a positive definite matrix factor analysis model, and the results were sequentially an urban raise dust source (36%), a construction dust source (31%), an iron and steel industry source (14%), a coal-fired source (10%), a secondary inorganic salt source (5%), and a motor vehicle exhaust emission source (4%). The source analysis result fully identifies the pollution source of the dust fall, and accords with the real atmospheric environment.
As can be seen from fig. 3, the influence effect of the main pollution source class on the dustfall change is quantitatively evaluated by using machine learning and partial dependence value calculation. Wherein the dotted line is the average dust amount during monitoring, and the solid line represents the variation relationship between the source type source contribution and the dust amount. The result shows that when the contribution of the urban dust source is lower than 20%, the dust reduction amount has no obvious change, and when the contribution of the urban dust source exceeds 40%, the dust reduction amount is obviously improved. The method shows that the urban fugitive dust source has the most important influence on the change of the Taiyuan dustfall pollution. The source contribution of the building dust source has a large influence on the dust reduction amount, and the dust reduction amount slowly rises along with the rise of the proportion of the building dust source contribution and is higher than the average dust reduction amount. The building dust source plays the most important role in the change of the dust fall pollution in the Taiyuan city. The influence of the source contribution of the iron and steel industry source on the dust reduction amount is stable, and the dust reduction amount slightly rises along with the increase of the source contribution ratio, and is similar to the annual average level. The reduction of the iron and steel industrial source has certain influence on the change of the dust fall pollution of the Taiyuan city. The influence of the source contribution of the coal-fired source on the dust fall amount is small, and the dust fall amount gradually rises and then slowly falls along with the rise of the contribution proportion of the coal-fired source, so that the contribution of the coal-fired source has a certain influence on the change of dust fall pollution in Taiyuan city. The contribution of the motor vehicle exhaust emission source and the secondary inorganic salt source has small influence on the dust reduction amount, and the dust reduction amount slightly decreases and is lower than the annual average level along with the increase of the source contribution ratio, which indicates that the motor vehicle exhaust emission source and the secondary inorganic salt source have little influence on the change of the Taiyuan dust reduction pollution.
Claims (7)
1. A method for analyzing an atmospheric dustfall pollution source and evaluating the dustfall marginal effect of the pollution source comprises the following steps:
(1) inputting the atmospheric dust fall multi-component data into a receptor model, constructing the multi-component data by utilizing the atmospheric dust fall quality monitored by a dust fall sampling instrument and the concentrations of chemical components such as elements, carbon and ions analyzed by an analyzer, and inputting the multi-component data into a positive definite matrix factor analysis model;
(2) setting parameters of model calculation, identifying main emission sources influencing atmospheric dust fall in a research area by carrying out survey on atmospheric dust fall sources in the research area, and setting the number of factors participating in calculation, the number of chemical components, the uncertainty of the chemical components and the number of output results;
(3) calculating a source analysis result, analyzing the pollution source of the atmospheric dust fall according to set parameters, and screening, judging and stretching the result through the survey result and expert experience to obtain a final model calculation result;
(4) and evaluating the marginal effect of the pollution source on the dust fall by utilizing a coupled receptor model result by utilizing a random forest algorithm and partial dependence calculation.
2. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: the method is characterized in that the atmospheric dust-fall multi-component data is input into a receptor model, and is model input data formed by monitoring data of atmospheric dust-fall quality and chemical components thereof based on different instruments, wherein the model input data comprises the atmospheric dust-fall quality, element concentration, carbon components and water-soluble ion data;
the mass of the atmospheric dust fall refers to the mass of particles settled from the atmosphere in unit time on a unit area, and the dust fall amount is analyzed by a gravimetric method; the water-soluble ions are measured by an ion chromatograph, including NH4 +、Na+、K+、Ca2+、Mg2+、SO4 2-、NO3 -And Cl-(ii) a Carbon composition was measured by a semi-continuous OC/EC instrument, including OC and EC; elements are monitored by heavy metal analyzers, including Ca, Si, Al, Fe, K, V, Cr, Mn, Co, Ni, Cu, Zn, Ga, As, Se, Ag, Cd, Sn, Sb, Ba, Au, Hg, Tl, Pb, and Bi components.
3. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: the positive definite matrix factor analysis model is a model which utilizes ion, element and carbon component data of atmospheric dust fall to extract a plurality of factors from a receptor matrix through positive definite matrix decomposition to obtain factor spectrum matrix information and factor contribution matrix information, identifies the factors into different source classes according to chemical identification components in the factor spectrum matrix, and utilizes the factor contribution matrix to calculate the contribution of each source class.
4. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: the investigation of the atmospheric dust suppression source comprises the following steps: firstly, collecting basic data; and secondly, investigating and researching the emission conditions of regional soil sand dust, road dust, construction dust, storage yard dust, fixed combustion sources, industrial sources and motor vehicle tail gas emission sources on the spot according to a source emission list.
5. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: the main emission source for identifying the atmospheric dust fall refers to the fact that the main emission source influencing the atmospheric dust fall in the research region is identified by investigating emission conditions of dust fall sources and combining social economy, energy consumption, city construction layout, meteorological data and pollution source emission information in the research region, and a basis is provided for analysis of the atmospheric dust fall source.
6. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: the random forest algorithm is an algorithm of a machine learning model, and is a classifier comprising a plurality of decision trees, the output category is determined by the mode of the category output by individual trees, and the atmosphere dust reduction amount is predicted.
7. The method of resolving atmospheric dustfall pollution sources and their assessment of the dustfall marginal effect on pollution sources as claimed in claim 1, wherein: and the partial dependence calculation shows the marginal effect of the pollution source on the prediction result of the random forest algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111310618.4A CN114117893A (en) | 2021-11-08 | 2021-11-08 | Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111310618.4A CN114117893A (en) | 2021-11-08 | 2021-11-08 | Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114117893A true CN114117893A (en) | 2022-03-01 |
Family
ID=80381004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111310618.4A Pending CN114117893A (en) | 2021-11-08 | 2021-11-08 | Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114117893A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116628598A (en) * | 2023-05-15 | 2023-08-22 | 生态环境部华南环境科学研究所(生态环境部生态环境应急研究所) | Dioxin source analysis method and system based on big data and NMF model |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108802262A (en) * | 2018-06-19 | 2018-11-13 | 南开大学 | Based on target elements conversion-PMF coupling algorithms associated with multicomponent online monitoring data and Factor Analysis Model |
-
2021
- 2021-11-08 CN CN202111310618.4A patent/CN114117893A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108802262A (en) * | 2018-06-19 | 2018-11-13 | 南开大学 | Based on target elements conversion-PMF coupling algorithms associated with multicomponent online monitoring data and Factor Analysis Model |
Non-Patent Citations (4)
| Title |
|---|
| 夏晓圣;陈菁菁;王佳佳;程先富;: "基于随机森林模型的中国PM_(2.5)浓度影响因素分析", 环境科学, vol. 41, no. 05, 31 May 2020 (2020-05-31), pages 2057 - 2065 * |
| 杨佳美;戴启立;刘保双;毕晓辉;宋文斌;吴建会;冯银厂;: "关中地区背景点位环境空气PM_(2.5)来源解析与多模型结果对比", 环境科学研究, no. 02, 28 October 2016 (2016-10-28) * |
| 王燕;彭林;李丽娟;王毓秀;张腾;刘海利;牟玲;: "晋城城市扬尘化学组成特征及来源解析", 环境科学, no. 01, 31 January 2016 (2016-01-31) * |
| 王露;毕晓辉;刘保双;吴建会;张裕芬;冯银厂;张勤勋;: "菏泽市PM_(2.5)源方向解析研究", 环境科学研究, no. 12, 28 February 2017 (2017-02-28) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116628598A (en) * | 2023-05-15 | 2023-08-22 | 生态环境部华南环境科学研究所(生态环境部生态环境应急研究所) | Dioxin source analysis method and system based on big data and NMF model |
| CN116628598B (en) * | 2023-05-15 | 2024-03-12 | 生态环境部华南环境科学研究所(生态环境部生态环境应急研究所) | Dioxin source analysis method and system based on big data and NMF model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111368401A (en) | Tracing method and device for pollution source and storage medium | |
| Hedberg et al. | Source contributions to PM10 and arsenic concentrations in Central Chile using positive matrix factorization | |
| Hopke | Recent developments in receptor modeling | |
| Gugamsetty et al. | Source characterization and apportionment of PM10, PM2. 5 and PM0. 1 by using positive matrix factorization | |
| Kim et al. | Improving source identification of Atlanta aerosol using temperature resolved carbon fractions in positive matrix factorization | |
| CN110489785B (en) | Online source analysis method and system for atmospheric pollutants | |
| Furusjö et al. | PM10 source characterization at urban and highway roadside locations | |
| CN105631203A (en) | Method for recognizing heavy metal pollution source in soil | |
| Tian et al. | PM2. 5 source apportionment during severe haze episodes in a Chinese megacity based on a 5-month period by using hourly species measurements: Explore how to better conduct PMF during haze episodes | |
| CN107300550A (en) | A kind of method based on BP neural network model prediction atmosphere heavy metal concentration | |
| CN111798928B (en) | Atmospheric particulate pollution source analysis method and device | |
| CN112613675A (en) | Analyzing pollution source and meteorological factor to PM of different degrees2.5Machine learning model of pollution impact contributions and effects | |
| Milando et al. | Trends in PM2. 5 emissions, concentrations and apportionments in Detroit and Chicago | |
| CN109187287A (en) | Source of atmospheric particulate matter analytic method based on stable element particle diameter distribution information | |
| CN112967764B (en) | Multi-technology coupled pollutant source analysis method and device | |
| Hopke | Chemometrics applied to environmental systems | |
| CN105844301A (en) | Soil heavy metal pollution source analysis method based on Bayes source identification | |
| Manousakas et al. | Quantitative assessment of the variability in chemical profiles from source apportionment analysis of PM10 and PM2. 5 at different sites within a large metropolitan area | |
| CN116205509A (en) | Research method for comprehensively evaluating heavy metal pollution condition of soil | |
| CN114117893A (en) | Method for analyzing atmospheric dust-fall pollution source and evaluating dust-fall marginal effect of pollution source | |
| CN115453064A (en) | Fine particle air pollution cause analysis method and system | |
| Mossetti et al. | Assessing the impact of particulate matter sources in the Milan urban area | |
| Sonu et al. | Linear regression based air quality data analysis and prediction using python | |
| CN109085291B (en) | Missing component iterative inversion calibration nesting-PMF source analysis algorithm | |
| Lowenthal et al. | Evaluation of regional-scale receptor modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |