A method for atmospheric co2 concentration level on pm 2.5 Method for pollution source emission reduction sensitivity analysis

Abstract

Atmospheric CO2 concentration level on PM 2.5 A pollution source emission reduction sensitivity analysis method relates to the field of coordinated emission reduction of atmospheric pollutants and greenhouse gases, and comprises the following steps: step 1: constructing a PM 2.5 Component and CO2 mixed input concentration matrix and uncertainty matrix; step 2, using a specific factor decomposition method to calculate the PM 2.5 And CO2 pollution source contribution; step 3, constructing a carbon fluctuation index and PM 2.5 Pollution source fluctuation index, to establish a machine learning model training data set; step 4, using the XGBOOST algorithm to construct the nonlinear relationship between CO2 and PM 2.5 Pollution source, to predict the change of CO2 concentration by reducing the proportion of different PM 2.5 Pollution source; step 5, using the XGBOOST algorithm to construct the nonlinear relationship between PM 2.5 Pollution source and CO2 pollution source, to predict the change of PM 2.5 Concentration by reducing the proportion of different CO2 pollution sources; the present application creatively combines the factor decomposition method and the machine learning method, links PM 2.5 Pollution source emission reduction and CO2 concentration, solves the problem that pollution prevention and control is separated from greenhouse gas prevention and control, and has good application prospect in atmospheric pollution prevention and control and climate improvement.

Description

Technical Field

[0001] This invention belongs to the field of synergistic emission reduction of air pollutants and greenhouse gases, specifically involving a coupled PMF and XGBOOST model for analyzing PM2.5 in the atmosphere. 2.5 The contribution of CO2 pollution sources and the assessment of the impact of CO2 concentration on PM2.5 2.5 Methods to reduce the sensitivity of pollution sources. Background Technology

[0002] Air pollution and the greenhouse effect have had a wide-ranging impact on humans and the living environment. PM2.5 2.5 PM2.5 is a major air pollutant, while CO2 is a primary greenhouse gas. 2.5 Since PM2.5 and CO2 share the same origin, it is crucial to design policies that can effectively reduce air pollution while simultaneously reducing carbon emissions. Therefore, it is necessary to study the effect of CO2 concentration levels on PM2.5. 2.5 Sensitivity analysis of pollution source reduction is conducted to identify optimal pollution sources that simultaneously reduce pollution and carbon emissions. The receptor model is currently used for PM2.5 reduction. 2.5 Common source apportionment methods and models include CMB and PMF. With proper calculation, these methods can accurately determine the impact of pollution sources on PM2.5 levels. 2.5 The contribution. However, due to the increasing prominence of the CO2 problem, only PM2.5 was calculated. 2.5 Optimal control of pollution sources is not enough; it is crucial to identify pollution sources that also have carbon reduction effects.

[0003] Faced with the above-mentioned increased control requirements, the main problems with current technical methods are: (1) Due to the complexity of the atmospheric environment, existing receptor models are difficult to realize CO2 and PM2.5. 2.5 (1) Source tracing; (2) Some source tracing methods require complex computing power; (3) CO2 and PM 2.5 The sources exhibit complex nonlinear relationships, making sensitivity analysis difficult. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an effect of atmospheric CO2 concentration levels on PM2.5. 2.5 The method for sensitivity analysis of pollution source emission reduction comprehensively utilizes multi-period particulate matter composition and CO2 monitoring data, and is effective not only for PM2.5 but also for other pollutants. 2.5 The sources of CO2 were monitored and analyzed, and machine learning methods were used to identify optimal PM2.5 control agents that could simultaneously reduce CO2 concentration. 2.5 source.

[0005] This invention provides an atmospheric CO2 concentration level effect on PM2.5 2.5 The technical solution used in the sensitivity analysis of pollution source emission reduction is as follows:

[0006] (1) Building a PM 2.5 Component and CO2 mixture input concentration matrix and uncertainty matrix

[0007] Reduce CO2 observations by 10 5 -10 6 Times the PM 2.5 The component values ​​are combined to form a mixed input matrix and then factorized. The source resolution yields the PM. 2.5 The results for the pollution source contribution of CO2 are more accurate. The specific steps and formulas for calculating the mixed input matrix are as follows:

[0008] 1) CO2 concentration weighted:

[0009]

[0010] in, This represents the weighted CO2 concentration, in p mg / m³. 3 p is the weighting coefficient, and its value is set to 10. 5 -10 6 When left and right, the PM obtained 2.5 The source contribution of CO2 is relatively accurate; Initial CO2 concentration, in mg / m³ 3 .

[0011] 2) PM 2.5 Component and CO2 mixture input matrix construction:

[0012]

[0013] Among them, X mix For inputting PM calculated in PMF 2.5 Components mixed with CO2 input matrix data; For PM 2.5 Component time series matrix data, This is time series data for CO2.

[0014] (2) Calculate PM 2.5 Contribution of CO2 pollution sources

[0015] PM 2.5The chemical components include elements, ions, and carbon. Water-soluble ions include NH4+, Na+, K+, Ca2+, Mg2+, SO42-, NO3-, and Cl-; carbon components include OC and EC; elements include 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. Species can be selected according to specific circumstances. The concentration matrix and uncertainty matrix are incorporated into a positive definite matrix factorization (PMF) model for decomposition. All species types are set to Strong, the number of runs is set to 20-30, and the number of factors is 6-9. The factor spectrum matrix and factor contribution matrix are optimized through rotation calculations. Factors are identified into different source classes based on the chemical identifier components in the factor spectrum matrix. The factor contribution matrix is ​​then compared with atmospheric PM2.5. 2.5 Multiple linear regression was used to calculate the impact of pollution sources on atmospheric PM2.5. 2.5 The contribution of pollution sources to CO2 is obtained by multiplying the factor contribution matrix with the CO2 source spectrum matrix.

[0016] (3) Constructing a carbon fluctuation index and PM 2.5 Pollution Source Fluctuation Index

[0017] Multiply the CO2 concentration by the PM2. 2.5 The carbon fluctuation index is constructed by comparing the concentration of PM2.5 with its average concentration. 2.5 Construct PM2.5 by multiplying the pollution source concentration by 100 and dividing by the CO2 pollution source concentration. 2.5 Pollution source fluctuation index.

[0018] (4) Machine learning methods to calculate the effect of CO2 concentration on PM2.5 2.5 Pollution source emission reduction sensitivity

[0019] The XGBOOST algorithm was used to establish a carbon fluctuation index and PM2.5 correlation. 2.5 The nonlinear relationship of the pollution source fluctuation index was investigated. n_estimators was set to 500-1000, max_depth to 3-20, the training set to 70-80%, and the test set to 20-30%. PM2.5 was then... 2.5 The pollution source fluctuation index is reduced by 10% each time it increases within the range of 10% to 50%, and the corresponding carbon fluctuation index is predicted to calculate the impact of CO2 concentration on PM2.5. 2.5 Sensitivity to pollution source emission reduction.

[0020] (5) Machine learning methods to calculate PM 2.5 Sensitivity of CO2 Concentration to Emission Reduction from Pollution Sources

[0021] PM is built using the XGBOOST algorithm 2.5The nonlinear relationship between the fluctuation index and the CO2 pollution source fluctuation index was investigated. n_estimators was set to 500-1000, max_depth to 3-20, the training set to 70-80%, and the test set to 20-30%. The CO2 pollution source fluctuation index was reduced by increasing by 10% every 10% within the range of 10% to 50%, and the corresponding PM2.5 was predicted. 2.5 Fluctuation index, used to calculate the effect of CO2 concentration on PM2.5. 2.5 Sensitivity to pollution source emission reduction.

[0022] The present invention relates to the effect of atmospheric CO2 concentration level on PM2.5. 2.5 The method for sensitivity analysis of pollution source emission reduction has the following advantages: it creatively combines factor decomposition and machine learning methods to analyze PM2.5 levels. 2.5 The link between pollution source emission reduction and CO2 concentration solves the problem of the separation between pollutant control and greenhouse gas control, and has great application prospects in air pollution prevention and climate improvement. Attached Figure Description

[0023] Figure 1 As a measure of atmospheric CO2 concentration level on PM2.5 2.5 Technical flowchart of pollution source emission reduction sensitivity analysis method;

[0024] Figure 2 PM obtained by the analysis of this invention 2.5 Figure showing the results of source contribution and CO2 source contribution;

[0025] Figure 3 The present invention relates to the effect of atmospheric CO2 concentration levels on PM2.5. 2.5 Pollution source emission reduction sensitivity results diagram;

[0026] Figure 4 For the atmospheric PM of this invention 2.5 A graph showing the sensitivity of CO2 pollution source emission reduction to concentration levels. Detailed Implementation

[0027] The following example illustrates the effect of atmospheric CO2 concentration level on PM2.5. 2.5 The method for sensitivity analysis of pollution source emission reduction is further described in detail so that those skilled in the art can implement it based on the description. Figure 1 This demonstrates the effect of atmospheric CO2 concentration levels on PM2.5. 2.5 The specific procedures for sensitivity analysis of pollution source emission reduction are as follows:

[0028] 1PM 2.5 Construction of component and CO2 mixture input concentration matrix and uncertainty matrix

[0029] Reduce CO2 observations by 10 5-10 6 Times, utilizing online PM monitoring 2.5 Component and CO2 concentration data, per PM 2.5 The components and CO2 are listed as characteristics in the table header, and the table is completed with species concentration data at each monitoring time as one row, thus forming the PM data. 2.5 Component and CO2 mixture input concentration matrix. PM 2.5 The uncertainty matrix is ​​constructed by multiplying the concentration of water-soluble ions in the chemical components by 0.2; the concentration of carbon components by 0.4; and the concentration of elemental components and CO2 by 0.6.

[0030] 2. Calculate PM 2.5 Contribution of CO2 pollution sources

[0031] The concentration matrix and uncertainty matrix were incorporated into a positive definite matrix factorization (PMF) model for decomposition. All species types were set to Strong, the number of runs was set to 20-30, and the number of factors was 6-9. The factor spectrum matrix and factor contribution matrix were optimized through rotation calculations. Factors were identified into different source classes based on the chemical identifier components in the factor spectrum matrix. The factor contribution matrix was then compared with atmospheric PM2.5. 2.5 Multiple linear regression was used to calculate the impact of pollution sources on atmospheric PM2.5. 2.5 The contribution of pollution sources to CO2 is determined by multiplying the factor contribution matrix with the CO2 source spectrum matrix. The results are as follows: Figure 2 As shown.

[0032] 3. Constructing a carbon volatility index and PM2.5 index 2.5 Pollution Source Fluctuation Index

[0033] Multiply the CO2 concentration by the PM2. 2.5 The carbon fluctuation index is constructed by comparing the concentration of PM2.5 with its average concentration. 2.5 Construct PM2.5 by multiplying the pollution source concentration by 100 and dividing by the CO2 pollution source concentration. 2.5 Pollution source fluctuation index.

[0034] 4. Machine learning methods to calculate the effect of CO2 concentration on PM2.5 2.5 Pollution source emission reduction sensitivity

[0035] The XGBOOST algorithm was used to establish a carbon fluctuation index and PM2.5 correlation. 2.5 The nonlinear relationship of the pollution source fluctuation index was investigated. n_estimators was set to 500-1000, max_depth to 3-20, the training set to 70-80%, and the test set to 20-30%. PM2.5 was then... 2.5 The pollution source fluctuation index is reduced by 10% each time it increases within the range of 10% to 50%, and the corresponding carbon fluctuation index is predicted to calculate the impact of CO2 concentration on PM2.5. 2.5 Sensitivity to pollution source reduction. The results are as follows: Figure 3 As shown.

[0036] 5 Machine Learning Methods for Calculating PM 2.5 Sensitivity of CO2 Concentration to Emission Reduction from Pollution Sources

[0037] PM is built using the XGBOOST algorithm 2.5 The nonlinear relationship between the fluctuation index and the CO2 pollution source fluctuation index was investigated. n_estimators was set to 500-1000, max_depth to 3-20, the training set to 70-80%, and the test set to 20-30%. The CO2 pollution source fluctuation index was reduced by increasing by 10% every 10% within the range of 10% to 50%, and the corresponding PM2.5 was predicted. 2.5 Fluctuation index, used to calculate the effect of CO2 concentration on PM2.5. 2.5 Sensitivity to pollution source reduction. The results are as follows: Figure 4 As shown.

Claims

1. The effect of atmospheric CO2 concentration level on PM2.5 2.5 The method for sensitivity analysis of pollution source emission reduction is characterized by: This method is implemented through the following steps: 1) Building a PM 2.5 The mixed input concentration matrix and uncertainty matrix of components and CO2 reduce the CO2 observations by 10. 5 -10 6 Times the PM 2.5 The component values ​​together form the input data concentration matrix, PM 2.5 The component values ​​and CO2 concentrations are multiplied by 0.2-0.6 to form the uncertainty matrix of the input data; 2) Calculate PM 2.5 To account for the pollution source contribution of CO2, the concentration matrix and uncertainty matrix were incorporated into a positive definite matrix factorization (PMF) model. All species types were set to Strong, the number of runs was set to 20-30, and the number of factors was 6-9. When all factors could be explained and the Q / Qexp value was between 0.75 and 1.25, the PM2.5 concentration was calculated. 2.5 And the contribution of CO2 pollution sources; 3) Based on the contributions of the aforementioned pollution sources, construct a carbon fluctuation index and a PM2.5 index. 2.5 The pollution source fluctuation index, in which the carbon fluctuation index characterizes the relative changes in CO2 contribution from pollution sources, is calculated by multiplying the CO2 concentration by the PM2.5 concentration. 2.5 The ratio of concentration to its average concentration, PM 2.5 The pollution source fluctuation index is used to characterize PM2.

5. 2.5 The variation characteristics of the contribution of pollution sources relative to the contribution of CO2 pollution sources are calculated by converting PM2.5 into the contribution of CO2 pollution sources. 2.5 Multiply the pollution source concentration by 100 and divide by the corresponding CO2 pollution source concentration; 4) For the target PM 2.5 A proportional reduction perturbation is applied to the contribution of pollution sources, and the carbon fluctuation index is established using the XGBoost algorithm in relation to PM2.

5. 2.5 The nonlinear relationship between pollution source fluctuation indices, respectively, will affect PM2.5 levels. 2.5 The pollution source fluctuation index is reduced in 10% increments within the range of 10%–50% to predict the corresponding carbon fluctuation index, thereby calculating the impact of CO2 concentration on PM2.

5. 2.5 Sensitivity to pollution source emission reduction; 5) Apply proportional reduction perturbations to the contribution of target CO2 pollution sources, and establish PM using the same method. 2.5 The nonlinear relationship between the volatility index and the carbon source volatility index is used to calculate PM2.

5. 2.5 Sensitivity of concentration to CO2 pollution source emission reduction.

2. The effect of atmospheric CO2 concentration level on PM2.5 as described in claim 1. 2.5 The method for sensitivity analysis of pollution source emission reduction is characterized by: Reduce CO2 observations by 10 5 -10 6 Times the PM 2.5 The component values ​​are combined to form a mixed input matrix and then factored to improve PM. 2.5 To improve the stability and accuracy of CO2 pollution source contribution analysis, the calculation steps for the mixed input matrix are as follows: 1) CO2 concentration weighted: in, This represents the weighted CO2 concentration, in p mg / m³. 3 p is the weighting coefficient, and its value is set to 10. 5 -10 6 When left and right, the PM obtained 2.5 The source contribution of CO2 is relatively accurate; Initial CO2 concentration, in mg / m³ 3 ; 2) PM 2.5 Component and CO2 mixture input matrix construction: Among them, X mix For inputting PM calculated in PMF 2.5 Components mixed with CO2 input matrix data; For PM 2.5 Component time series matrix data, This is time series data for CO2.

Images