Method for quantifying the influence of fireworks and crackers on fine particulate matter levels

Abstract

This invention discloses a method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels in the field of air pollutant assessment technology, including obtaining local meteorological factors and local PM2.5 levels. 2.5 Concentration data; combining the meteorological factors and PM2.5. 2.5 Concentration data were processed into training and validation sets; based on the training set, a model was established regarding meteorological factors and PM2.5. 2.5 XGBoost model for concentration data. This invention utilizes local meteorological factors and local PM2.5 concentration data. 2.5 The XGBoost model was trained using a training set derived from concentration data. After the XGBoost model was trained, the impact of fireworks and firecrackers on PM2.5 was calculated. 2.5 When considering the contribution of concentration to the overall effect, PM2.5 concentration can be simulated and calculated by inputting relevant meteorological factor data. 2.5 The concentration reconstruction data is then combined with the actual monitored PM2.5 values. 2.5 Concentration data and PM 2.5 By subtracting the concentration reconstruction values, the impact of fireworks and firecrackers on PM2.5 can be directly determined. 2.5 This method can directly quantify the contribution of concentration to the total PM2.5 concentration. 2.5 The concentration reconstruction data provides fast and efficient calculation results.

Description

Technical Field

[0001] This invention relates to the field of atmospheric pollutant assessment technology, specifically a method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels. Background Technology

[0002] The main chemical components of fireworks and firecrackers include oxidizers, combustibles, flame colorants, and other special agents. Under the high temperature and pressure conditions after ignition, the substances in fireworks and firecrackers undergo a series of chemical reactions, releasing large amounts of sulfur dioxide, nitrogen oxides, heavy metals, and organic matter, and producing a large amount of particulate matter, thus causing a significant deterioration in air quality.

[0003] Currently, the most widely used methods for assessing the impact of fireworks and firecrackers on PM2.5 are... 2.5 The method for assessing the impact of particulate matter (with a diameter of 2.5 micrometers or less) is based on PM. 2.5 / CO ratio method. The impact of fireworks and firecrackers on PM2.5. 2.5 PM 10 SO2 and NO2 concentrations have a significant impact, while their impact on CO concentration is relatively small. Therefore, CO can be used as a reference standard pollutant concentration, utilizing PM2.5 concentrations. 2.5 / CO ratio analysis of the impact of fireworks and firecrackers on PM2.5 2.5 PM's contribution. 2.5 The / CO method assumes a stable CO concentration. However, incomplete combustion during fireworks displays generates CO, causing an increase in CO concentration and thus affecting the PM2.5 concentration obtained from regression. 2.5 The concentration was too high, ultimately underestimating the impact of fireworks on PM2.5. 2.5 The impact. Summary of the Invention

[0004] The purpose of this invention is to provide a method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels, in order to address the aforementioned issue that fireworks and firecrackers generate CO during their combustion, leading to an increase in CO concentration and consequently affecting the PM2.5 levels obtained through regression analysis. 2.5 The problem is that the concentration is too high.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels includes: acquiring local meteorological factors and local PM2.5 levels. 2.5 Concentration data; combining the meteorological factors and PM2.5. 2.5 Concentration data were processed into training and validation sets; based on the training set, a model was established regarding meteorological factors and PM2.5. 2.5 An XGBoost model is used to analyze PM concentration data; meteorological factors from the validation set are input into the XGBoost model, and the XGBoost model outputs PM based on these meteorological factors.2.5 Concentration reconstruction data, including PM 2.5 The concentration reconstruction data are from the XGBoost model simulation of the situation without fireworks or firecrackers; the PM values ​​in the validation set are... 2.5 Concentration data and PM 2.5 The concentration reconstruction values ​​are subtracted to obtain a quantitative concentration difference value, which provides a direct quantification of the impact of fireworks and firecrackers on PM2.5. 2.5 The effect of concentration.

[0007] As a further aspect of the present invention: the acquisition of historical local meteorological factors and local PM2.5... 2.5 Concentration data includes: the meteorological factors include local temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure, wherein temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure are derived from local meteorological monitoring data and environmental monitoring data. Beneficial effects: This method inputs multiple meteorological data into a trained XGBoost model, calculates various factors through the XGBoost model, and simulates the current PM2.5 concentration. 2.5 The reconstructed concentration values ​​are more accurate.

[0008] As a further aspect of the present invention: the wind volume vector is decomposed into u-components and v-components; the u-component = wind speed * sin((wind direction - 180) × π / 180); the v-component = wind speed * cos((wind direction - 180) × π / 180); wherein the wind speed and wind direction data are obtained from local meteorological monitoring data. Beneficial effects: By decomposing the wind volume into u-components and v-components, and after the meteorological monitoring equipment detects the wind speed and wind direction data, the u-components and v-components are calculated, and then the wind volume is calculated by combining them. The wind volume calculation is fast, convenient, accurate, and easy to operate.

[0009] As a further aspect of the present invention: the meteorological factors and PM 2.5 Concentration data were processed into training and validation sets, including: the local meteorological factors and local PM2.5. 2.5 A portion of the concentration data was extracted as the training set; the local meteorological factors and local PM2.5... 2.5 The remaining part of the concentration data is the validation set; the training set is the corresponding data monitored during periods when fireworks and firecrackers are not set off; the validation set includes the corresponding data monitored during periods when fireworks and firecrackers are set off.

[0010] As a further aspect of the present invention, the amount of data in the training set is greater than the amount of data in the validation set.

[0011] As a further aspect of the present invention: the establishment of information about meteorological factors and PM based on the training set... 2.5The XGBoost model for concentration data includes: the XGBoost model having several hyperparameters; the hyperparameters include the number of trees, learning rate, maximum tree depth, minimum loss function reduction required for training each tree and splitting the tree; the number of trees represents the number of iterations; the learning rate controls the contribution of each tree to the final prediction; the maximum tree depth controls the tree complexity; the proportion of samples randomly selected during training each tree; and the minimum loss function reduction required for splitting the tree controls the tree generation process.

[0012] As a further aspect of the present invention: the XGBoost model is established by selecting optimized hyperparameters using Yeats optimization to build an optimized XGBoost model; the optimized XGBoost model fits multiple predicted PMs based on the data in the training set. 2.5 Concentration data, based on multiple predicted PM2.5 concentrations. 2.5 Concentration data and corresponding local PM in the training set 2.5 The coefficient of determination R was calculated from the concentration data. 2 (Linear regression fit), root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv), with the coefficient of determination (R) 2 (Linear regression fit), root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv) are used to reflect the stability and accuracy of the optimized XGBoost model.

[0013] As a further aspect of the present invention, it also includes: obtaining PM 2.5 The / CO ratio method is used to calculate PM2.5 during non-firework periods. 2.5 The PM2.5 concentration data and CO concentration data are used to calculate the ratio between the two. This ratio is then multiplied by the CO concentration data for the combustion period to obtain the PM2.5 concentration for that period. 2.5 Regression data will be used to determine the PM2.5 concentration during the fireworks display period. 2.5 measured concentration data minus PM 2.5 Regression data was used to derive the ratio concentration difference, in order to determine the impact of fireworks and firecrackers on PM2.5. 2.5 The impact; compare the quantitative concentration difference with the ratio concentration difference.

[0014] Compared with the prior art, the beneficial effects of the present invention are:

[0015] 1. In this invention, local meteorological factors and local PM2.5 are used. 2.5 The XGBoost model was trained using a training set derived from concentration data. After the XGBoost model was trained, the impact of fireworks and firecrackers on PM2.5 was calculated. 2.5 When considering the contribution of concentration to the overall effect, PM2.5 concentration can be simulated and calculated by inputting relevant meteorological factor data. 2.5 The concentration reconstruction data is then combined with the actual monitored PM2.5 values.2.5 Concentration data and PM 2.5 By subtracting the concentration reconstruction values, the impact of fireworks and firecrackers on PM2.5 can be directly determined. 2.5 This method can directly quantify the contribution of concentration to the total PM2.5 concentration. 2.5 Concentration reconstruction data, in calculating PM 2.5 The numerical process for calculating the contribution of concentration is simple, does not require a lot of computing power, does not rely on a three-dimensional transmission model, and yields fast and efficient results.

[0016] 2. In this invention, it is not necessary to detect CO concentration data, and therefore the changes in CO concentration caused by setting off fireworks will not affect the PM2.5 levels after subsequent fireworks displays. 2.5 The concentration contributes to the impact, ensuring the accuracy of data on the impact of fireworks and firecrackers on fine particulate matter levels, and achieving good results. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the method flow steps of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example:

[0020] like Figure 1 The diagram illustrates the steps of the method of the present invention. In this embodiment, a method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels includes:

[0021] S1: Obtain local weather factors and local PM2.5. 2.5 Concentration data;

[0022] S2: Combine meteorological factors and PM 2.5 Concentration data were processed into training and validation sets;

[0023] S3: Establish information on meteorological factors and PM based on the training set. 2.5 XGBoost model for concentration data;

[0024] S4: Meteorological factors from the validation set are input into the XGBoost model, and the XGBoost model outputs PM based on the meteorological factors from the validation set. 2.5 Concentration reconstruction data, including PM 2.5The concentration reconstruction values ​​are from XGBoost model simulations of situations where no fireworks or firecrackers were set off.

[0025] S5: Validate the PM in the validation set 2.5 Concentration data and PM 2.5 The concentration reconstruction values ​​are subtracted to obtain a quantitative concentration difference value, which provides a direct quantification of the impact of fireworks and firecrackers on PM2.5. 2.5 The effect of concentration.

[0026] In this invention, local meteorological factors and local PM2.5 are used. 2.5 The XGBoost model was trained using a training set derived from concentration data. After the XGBoost model was trained, the impact of fireworks and firecrackers on PM2.5 was calculated. 2.5 When considering the contribution of concentration to the overall effect, PM2.5 concentration can be simulated and calculated by inputting relevant meteorological factor data. 2.5 The concentration reconstruction data is then combined with the actual monitored PM2.5 values. 2.5 Concentration data and PM 2.5 By subtracting the concentration reconstruction values, the impact of fireworks and firecrackers on PM2.5 can be directly determined. 2.5 This method can directly quantify the contribution of concentration to the total PM2.5 concentration. 2.5 Concentration reconstruction data, in calculating PM 2.5 The numerical calculation of the contribution of concentration to air quality is simple, does not require a lot of computing power, does not rely on a three-dimensional transmission model, and provides fast and efficient results, which is beneficial for subsequent air quality management.

[0027] This invention does not require the detection of CO concentration data, therefore changes in CO concentration caused by fireworks will not affect the PM2.5 levels after subsequent fireworks displays. 2.5 The concentration contributes to the impact, ensuring the accuracy of data on the impact of fireworks and firecrackers on fine particulate matter levels, and achieving good results.

[0028] Furthermore, obtain local meteorological factors and local PM2.5. 2.5 The concentration data are collected continuously in hours or days.

[0029] Furthermore, Extreme Gradient Boosting Decision Tree (XGBoost) is an algorithm or engineering implementation based on Gradient Boosting Decision Tree (GBDT). The basic idea of ​​XGBoost is the same as GBDT, but it incorporates several optimizations. For example, it uses the second derivative in the formula derivation to make the loss function more accurate; it adds the tree model complexity as a regularization term to the optimization objective to avoid overfitting; it pre-sorts the data through parallel computation and then saves it as a block structure, which is repeatedly used in subsequent iterations, greatly reducing the computational load; and it can automatically learn strategies for handling missing values.

[0030] In this embodiment, historical local meteorological factors and local PM2.5 are obtained. 2.5 Concentration data includes meteorological factors such as local temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure. Temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure are derived from local meteorological and environmental monitoring data. The relevant data on temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure detected by local meteorological monitoring equipment are input into the XGBoost model. The XGBoost model calculates multiple data points to simulate the current PM2.5 concentration. 2.5 This method uses multiple meteorological data points as input to a trained XGBoost model. The XGBoost model calculates various factors to simulate the current PM2.5 concentration. 2.5 The reconstructed concentration values ​​are more accurate.

[0031] In this embodiment, the wind volume vector is decomposed into u-components and v-components; u-component = wind speed * sin((wind direction - 180) × π / 180); v-component = wind speed * cos((wind direction - 180) × π / 180); where wind speed and wind direction data come from local meteorological monitoring data. By decomposing the wind volume into u-components and v-components, and then calculating the u-components and v-components after the meteorological monitoring equipment detects the wind speed and wind direction data, the wind volume is calculated. This method is fast, convenient, accurate, and easy to operate.

[0032] In this embodiment, meteorological factors and PM 2.5 Concentration data were processed into training and validation sets, including local meteorological factors and local PM2.5. 2.5 A portion of the concentration data was extracted as the training set; local meteorological factors and local PM2.5 were also included. 2.5 The remaining portion of the concentration data serves as the validation set; the training set consists of the corresponding data monitored during periods when fireworks and firecrackers are not set off; the validation set includes the corresponding data monitored during periods when fireworks and firecrackers are set off. Local meteorological factors and local PM2.5 concentrations are then compared. 2.5The training set formed from concentration data is used to train the XGBoost model, which in turn generates meteorological factors and PM2.5. 2.5 Based on the relationship between concentration data, and by inputting meteorological factors, the PM2.5 concentration during simulated periods when fireworks and firecrackers are not set off can be directly quantified. 2.5 Concentration reconstruction data is needed to quantify and verify the impact of concentrated fireworks explosions on PM2.5. 2.5 When considering the impact of concentration, after inputting the meteorological factors verified by Jizhong into the XGBoost model, PM2.5 can be directly quantified. 2.5 Concentration reconstruction data, and then PM in the validation set 2.5 Concentration data and PM 2.5 The concentration reconstruction values ​​are subtracted to obtain a quantified concentration difference value. The magnitude of this quantified concentration difference value is used to intuitively quantify the impact of fireworks and firecrackers on PM2.5. 2.5 The effect of concentration.

[0033] In this embodiment, the training set contains more data than the validation set. A larger training set allows for more stable and accurate training of the XGBoost model.

[0034] Furthermore, the training set accounts for meteorological factors and PM2.5. 2.5 The XGBoost model is trained more stably and accurately using 80% of the total concentration data and the remaining data as a validation set.

[0035] In this embodiment, information about meteorological factors and PM is established based on the training set. 2.5The XGBoost model for concentration data includes: The XGBoost model has several hyperparameters; the hyperparameters include the number of trees, learning rate, maximum tree depth, minimum loss function reduction required for training each tree and splitting a tree; the number of trees represents the number of iterations; the learning rate controls the contribution of each tree to the final prediction; the maximum tree depth controls the complexity of the tree; the proportion of samples randomly selected for training each tree; and the minimum loss function reduction required for splitting a tree controls the tree generation process. Specifically, XGBoost has several hyperparameters, including: the number of trees (n_estimators), which represents the number of iterations; more trees improve the model's fitting ability but can also easily lead to overfitting; the learning rate (learning_rate or eta), which controls the contribution of each tree to the final prediction; a smaller learning rate usually requires more trees to converge but may improve the model's generalization ability; the maximum tree depth (max_depth), which controls the complexity of the trees; overly deep trees are prone to overfitting, while overly shallow trees may not be able to capture the complexity of the data; the proportion of randomly selected samples per tree during training (subsample), which is usually set between 0.5 and 1.0; a value that is too small may lead to underfitting, while a value that is too large may lead to overfitting; and the minimum loss function reduction required for tree splitting (gamma), which controls the tree generation process.

[0036] In this embodiment, Yeats optimization is used to select and optimize hyperparameters to build an optimized XGBoost model. The optimized XGBoost model fits multiple predicted PMs based on the data in the training set. 2.5 Concentration data, based on multiple predicted PM2.5 concentrations. 2.5 Concentration data and corresponding local PM in the training set 2.5 The coefficient of determination R was calculated from the concentration data. 2 (Linear regression fit), root mean square error (RMSE), correlation coefficient (R), coefficient of variation (cv), and coefficient of determination (R²). 2 (Linear regression fit), root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv) are used to reflect the stability and accuracy of the optimized XGBoost model.

[0037] Different combinations of hyperparameters have a significant impact on model performance, based on multiple predicted PM... 2.5 Concentration data and corresponding local PM in the training set 2.5 Concentration data, calculate the coefficient of determination R. 2 The root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv) are used to evaluate the model performance. These parameters are used to intuitively evaluate the XGBoost model and continuously optimize it to achieve the goal of improving the stability and accuracy of the optimized XGBoost model.

[0038] Furthermore, when building the XGBoost model, a Bayesian optimization method is used to select the optimal hyperparameters and establish the optimal XGBoost model. The coefficient of determination R of the XGBoost model is... 2 The root mean square error (RMSE) was 28.69, the correlation coefficient (R) was 0.90, and the coefficient of variation (cv) was 0.62.

[0039] The specific definition of the Bayesian optimization objective function is as follows:

[0040]

[0041]

[0042] Substituting hourly meteorological data into the model, the reconstructed PM2.5 concentration function, unaffected by fireworks and firecrackers, is as follows:

[0043] Predict using the optimal model

[0044] y_pred=pd.Series(best model.predict(X nofire),index=Xnofire.index)

[0045] # Create a result data frame

[0046] result_df = pd.DataFrame({

[0047] Real-time value':y_nofire,

[0048] Reconstructed value':y pred,

[0049] The impact of fireworks and firecrackers:y_nofire-y_pred

[0050] })

[0051] PM 24 hours a day 2.5 The daily average value is calculated from the reconstructed concentration values, and then compared with the actual monitored PM2.5 daily average. 2.5 By subtracting the concentration data, the impact of daily fireworks and firecracker use on PM2.5 can be intuitively quantified. 2.5 The effect of concentration.

[0052] This embodiment also includes: obtaining PM 2.5 The / CO ratio method is used to calculate PM2.5 during non-firework periods. 2.5 The PM2.5 concentration data and CO concentration data are used to calculate the ratio between the two. This ratio is then multiplied by the CO concentration data for the combustion period to obtain the PM2.5 concentration for that period. 2.5 Regression data will be used to determine the PM2.5 concentration during the fireworks display period. 2.5measured concentration data minus PM 2.5 Regression data was used to derive the ratio concentration difference, in order to determine the impact of fireworks and firecrackers on PM2.5. 2.5 The impact; compare the quantitative concentration difference with the ratio concentration difference.

[0053] The main chemical components of fireworks and firecrackers include oxidizers, combustibles, flame colorants, and other special agents.

[0054] Under the high temperature and high pressure conditions after ignition, the substances in fireworks and firecrackers undergo a series of chemical reactions, releasing large amounts of sulfur dioxide, nitrogen oxides, heavy metals, and organic matter, and producing a large amount of particulate matter, thus causing a significant deterioration in air quality.

[0055] Currently, the most widely used methods for assessing the impact of fireworks and firecrackers on PM2.5 are... 2.5 The method of influence is based on PM 2.5 Methods for determining the CO / O ratio. The impact of fireworks and firecrackers on PM2.5. 2.5 PM 10 SO2 and NO2 concentrations have a significant impact, while their impact on CO concentration is relatively small. Therefore, CO can be used as a reference standard pollutant concentration, utilizing PM2.5 concentrations. 2.5 / CO ratio analysis of the impact of fireworks and firecrackers on PM2.5 2.5 Contributions:

[0056]

[0057] PM 2.5 f =PM 2.5 -PM 2.5 r

[0058] In the formula: Average PM2.5 during non-fireworks periods 2.5 The ratio of concentration to average CO concentration, CO, PM 2.5 r These represent hourly CO concentration and hourly PM2.5 obtained from regression, respectively. 2.5 Concentration, PM 2.5 For hourly PM 2.5 Concentration, PM 2.5 f The ratio is the concentration difference, used to show the effect of fireworks and firecrackers on PM2.5. 2.5 The hourly contribution.

[0059] After obtaining PM 2.5 The ratio concentration difference obtained by the / CO ratio method is compared with the quantitative concentration difference calculated by this method. The two detection methods can then be used to verify each other, further improving the accuracy of the simulation results.

[0060] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels, characterized in that, include: Obtain local meteorological factors and local PM2.

5. 2.5 Concentration data; The meteorological factors and PM 2.5 Concentration data processing for training and validation sets includes: A portion of the local meteorological factors and local PM2.5 concentration data is used as the training set; the remaining portion of the local meteorological factors and local PM2.5 concentration data is used as the validation set; the training set consists of the corresponding data monitored during periods when fireworks and firecrackers are not set off; the validation set includes the corresponding data monitored during periods when fireworks and firecrackers are set off. Based on the training set, establish information on meteorological factors and PM. 2.5 XGBoost model for concentration data; The meteorological factors in the validation set are input into the XGBoost model, and the XGBoost model outputs PM based on the meteorological factors in the validation set. 2.5 Concentration reconstruction data, including PM 2.5 The concentration reconstruction values ​​are from XGBoost model simulations of situations where no fireworks or firecrackers were set off. PM in the verification set 2.5 Concentration data and PM 2.5 The concentration reconstruction values ​​are subtracted to obtain a quantitative concentration difference value, which provides a direct quantification of the impact of fireworks and firecrackers on PM2.

5. 2.5 To investigate the impact of concentration, the PM2.5 and CO concentration data for non-fireworks periods were calculated using the PM2.5 / CO ratio method. The ratio between the two was then calculated, and this ratio was multiplied by the CO concentration data for the fireworks period to obtain the PM2.5 regression data for the fireworks period. The measured PM2.5 concentration data for the fireworks period was then subtracted from the PM2.5 regression data to obtain the ratio concentration difference, which is used to determine the impact of fireworks on PM2.

5. The quantitative concentration difference was then compared with the ratio concentration difference.

2. The method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels according to claim 1, characterized in that, The acquisition of local meteorological factors and local PM2.5 2.5 Concentration data, including: The meteorological factors include local temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure, wherein temperature, humidity, precipitation, wind volume, wind speed, and surface air pressure are derived from local meteorological monitoring data and environmental monitoring data.

3. The method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels according to claim 2, characterized in that: The air volume vector is decomposed into u-components and v-components; The u component = wind speed * sin((wind direction - 180) × π / 180); The v component = wind speed * cos((wind direction - 180) × π / 180); The wind speed and direction data are from local meteorological monitoring data.

4. The method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels according to claim 3, characterized in that: The training set contains more data than the validation set.

5. The method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels according to claim 1, characterized in that, The method for establishing information on meteorological factors and PM based on the training set is described. 2.5 XGBoost models for concentration data include: The XGBoost model has several hyperparameters; The hyperparameters include the number of trees, learning rate, maximum tree depth, minimum loss function reduction required for training each tree and splitting a tree; The number of trees is used to represent the number of iterations; The learning rate is used to control the contribution of each tree to the final prediction; The maximum depth of the tree is used to control the complexity of the tree; The proportion of samples randomly selected during the training of each tree; The minimum loss function required for the splitting of the tree is reduced to control the tree generation process.

6. The method for quantifying the impact of fireworks and firecrackers on fine particulate matter levels according to claim 5, characterized in that: The XGBoost model uses Yeats optimization to select optimized hyperparameters in order to establish an optimized XGBoost model. The optimized XGBoost model fits multiple predicted PM values ​​based on the data in the training set. 2.5 Concentration data, based on multiple predicted PM2.5 concentrations. 2.5 Concentration data and corresponding local PM in the training set 2.5 The coefficient of determination R was calculated from the concentration data. 2 The root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv), with the coefficient of determination (R) being... 2 The root mean square error (RMSE), correlation coefficient (R), and coefficient of variation (cv) are used to reflect the stability and accuracy of the optimized XGBoost model.

Images