Agricultural machinery PM based on drone monitoring data 2.5 Pollution Emission Estimation Methods and Systems

Abstract

This application discloses an agricultural machinery PM based on drone monitoring data. 2.5 This invention addresses the technical challenges of non-overlapping monitoring points and emission sources, as well as poor local adaptability of emission factors in existing agricultural machinery emission monitoring methods and systems. It first utilizes unmanned aerial vehicles (UAVs) to carry PM2.5 data. 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area and preprocesses the data by performing missing data completion, anomaly removal, and duplicate correction. Secondly, based on a Gaussian plume diffusion model, combined with on-site meteorological parameters and agricultural machinery operating parameters, it inversely calculates the emission source intensity of agricultural machinery. Finally, it uses the analytic hierarchy process (AHP) to perform weighted analysis on multiple factors such as agricultural machinery age, operating mode, environmental humidity, and temperature, achieving localized and accurate calibration of emission factors. This application effectively solves the problem of spatial misalignment between monitoring points and emission sources, significantly improving the PM2.5 concentration of agricultural machinery. 2.5 The accuracy and reliability of emission estimates provide a quantitative basis for the compilation of regional emission inventories and the formulation of agricultural machinery renewal policies.

Description

Technical Field

[0001] This application relates to the field of air pollution monitoring and emission source intensity estimation technology, specifically to a method for detecting PM2.5 emissions from agricultural machinery based on UAV monitoring data. 2.5 Methods and systems for estimating pollution emissions. Background Technology

[0002] In the field of industrial equipment fault diagnosis, agricultural machinery, as an important component of non-road mobile sources, generates PM2.5 during regional concentrated operations. 2.5 Emissions have become a key issue in the prevention and control of agricultural non-point source pollution. Accurately estimating the emission levels of agricultural machinery under actual operating conditions is a prerequisite for local environmental protection departments to compile regional emission inventories, formulate pollution prevention and control policies, and develop agricultural machinery renewal plans.

[0003] Currently, the mainstream methods for monitoring agricultural machinery emissions mainly include emission inventory modeling, vehicle-specific power analysis, and on-board measurement using portable emission measurement systems (PEMS). Among these, PEMS can obtain real-time emission data of agricultural machinery under actual operating conditions, providing relatively strong data representativeness, but it has limitations such as high equipment cost, complex operation, and difficulty in large-scale deployment. Emission inventory modeling relies on universal emission factors, making it difficult to reflect regional differences in agricultural machinery age structure, operating modes, and environmental conditions, leading to significant discrepancies between estimated results and actual conditions.

[0004] In recent years, UAV remote sensing technology has been widely used in the field of air pollution monitoring due to its advantages of flexibility, efficiency, and ability to perform low-altitude, short-range monitoring. Some studies have attempted to use UAVs equipped with particulate matter monitoring equipment for monitoring agricultural machinery emissions, but the following technical challenges remain:

[0005] First, there's the issue of monitoring points not coinciding with emission sources. For example, PM2.5 emissions from agricultural machinery... 2.5 After pollutants diffuse through the atmosphere, the concentration data collected by drones is the superposition result of the diffusion, which cannot directly represent the original emission intensity, resulting in a lack of direct correspondence between monitoring data and emission sources.

[0006] Second, there is the problem of poor localization adaptability of emission factors. Existing national guidelines recommend emission factors mostly based on bench steady-state testing, while actual field operations of agricultural machinery involve various non-steady-state conditions such as starting, acceleration, turning, U-turns, and sudden load changes, resulting in significant differences in emission characteristics compared to bench testing. Furthermore, regional differences in the age structure of agricultural machinery, fuel quality, and climate conditions also make it difficult to directly apply universal emission factors.

[0007] Third, the multi-factor coupling mechanism is unclear. Agricultural machinery emissions are affected by a combination of factors such as vehicle age, operating mode, and ambient temperature and humidity. Existing technologies lack systematic quantitative analysis methods for these influencing factors, making it difficult to achieve accurate local calibration of emission factors. Summary of the Invention

[0008] To address the aforementioned technical challenges, this application proposes a three-pronged agricultural machinery PM (micro-monitoring) system integrating "UAV mobile monitoring + Gaussian plume inverse calculation + analytic hierarchy process localization calibration". 2.5 This emission estimation method aims to address the spatial misalignment between monitoring points and emission sources, achieving accurate localized calibration of emission factors and providing new technical support for county-level agricultural machinery emission assessment. The main objective of this application is to provide a method for assessing PM emissions from agricultural machinery based on UAV monitoring data. 2.5 The method for estimating pollution emissions includes the following steps:

[0009] Step S1: Obtain PM from the drone 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as meteorological environmental parameters and agricultural machinery operating parameters collected simultaneously.

[0010] Step S2: Preprocess the pollutant concentration data, including missing data completion, abnormal data removal and duplicate data correction, to obtain the preprocessed effective concentration data;

[0011] Step S3: Based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and using the effective concentration data, the PM2.5 concentration of the agricultural machinery is calculated in reverse. 2.5 Emission source strength;

[0012] Step S4: Calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery;

[0013] Step S5: Use the analytic hierarchy process (AHP) to perform weight analysis on the multiple factors affecting the emission factors, and perform localized calibration on the initial emission factors to obtain localized emission factors.

[0014] In one embodiment, the acquisition of UAV-borne PM 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as simultaneously collected meteorological environmental parameters and agricultural machinery operating parameters, including:

[0015] Before agricultural machinery operations, control the drone carrying PM 2.5 The monitoring system hovers at the preset monitoring points to collect background PM2.5 in farmland. 2.5 Concentration data, collected for a duration of no less than the first preset duration, is used to eliminate the interference of regional baseline concentrations on emission estimation;

[0016] After the agricultural machinery begins operation, control the drone carrying PM 2.5 The monitoring system hovers in automatic tracking mode at a first preset distance directly downwind of the agricultural machinery and at a second preset height above the machinery's exhaust outlet, simultaneously collecting PM2.5 levels throughout the entire operation of the agricultural machinery. 2.5 Concentration data should be collected for a duration of no less than the second preset duration, covering both effective operations and field turning conditions.

[0017] During the data collection process, wind speed, wind direction, temperature, humidity and atmospheric pressure data were collected by meteorological and environmental monitoring stations set up at the edge of the work area. The rated power, actual working time, fuel consumption, exhaust pipe physical height, flue gas outlet temperature and flue gas outlet velocity parameters of the agricultural machinery were obtained by inputting through the agricultural machinery vehicle recorder or ground workstation.

[0018] In one embodiment, the step of preprocessing the pollutant concentration data specifically includes:

[0019] For a single missing line, the average concentration values ​​at adjacent time points are used for completion; for two consecutive missing lines, bidirectional filling is used; and for three or more consecutive missing lines, linear interpolation is used for completion.

[0020] Outliers are identified based on the Raida criterion. When the number of outliers accounts for less than or equal to 5% of the total number of samples, the outliers are directly removed. When the number of outliers exceeds 5%, the relevant time period is investigated and then linear interpolation is used to complete the data.

[0021] For consecutively repeated data, cubic spline interpolation is used for replacement, and the interpolation interval covers the preset number of valid data before and after the repeated data.

[0022] In one embodiment, the PM2.5 concentration of the agricultural machinery is calculated by combining the Gaussian plume diffusion model with the meteorological environmental parameters and the operating parameters of the agricultural machinery, and by using the effective concentration data to inversely calculate the PM2.5 concentration of the agricultural machinery. 2.5 The steps for strengthening emission source control include:

[0023] Based on surface wind speed, solar radiation intensity, and cloud cover data from meteorological environmental parameters, the atmospheric stability level under the current operating conditions is determined according to the Pasquill-Gifford stability rating standard.

[0024] Based on the atmospheric stability level, the horizontal and vertical diffusion coefficients were calculated using the Briggs close-range diffusion model.

[0025] The system obtains the physical height of the exhaust pipe, the flue gas outlet temperature, the flue gas outlet velocity, and the inner diameter of the exhaust pipe outlet from the operating parameters of agricultural machinery, as well as the ambient temperature and ambient atmospheric pressure from the meteorological environmental parameters. Based on the segmented lifting logic, the flue gas lifting height is calculated, and the sum of the physical height of the exhaust pipe and the flue gas lifting height is taken as the effective height of the pollution source.

[0026] Substituting the preprocessed effective concentration data, the horizontal diffusion coefficient, the vertical diffusion coefficient, the effective height of the pollution source, and the average wind speed, crosswind distance, and monitoring height at the exhaust pipe height from the meteorological environmental parameters, into the Gaussian plume inverse model formula, the PM2.5 concentration of agricultural machinery is obtained by inversely solving the model. 2.5 Strong emission sources.

[0027] In one embodiment, the step of calculating the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery includes:

[0028] Multiplying the obtained emission source intensity by the actual operating time of the agricultural machinery, the PM2.5 concentration of the agricultural machinery during the actual operating period is calculated. 2.5 Total emissions;

[0029] Obtain the total fuel consumption of agricultural machinery during the actual operating period, and convert the PM... 2.5 The total emissions are divided by the total fuel consumption to calculate the initial emission factor based on fuel consumption;

[0030] The rated power of the agricultural machinery is obtained. The initial emission factor based on power is calculated by multiplying the initial emission factor based on fuel consumption by the total fuel consumption and then dividing by the product of the rated power and the actual operating time. This initial emission factor is then used as the initial emission factor for subsequent localization calibration.

[0031] In one embodiment, the step of using the analytic hierarchy process (AHP) to perform weighted analysis on the multiple factors affecting the emission factors and to localize the initial emission factors to obtain localized emission factors includes:

[0032] Determine the PM affecting agricultural machinery 2.5 The key factors of emission factors are used as the target layer, and the key factors are used as the criterion layer to construct a hierarchical analysis structure model.

[0033] By using expert scoring or questionnaires, pairwise comparisons are made between the key factors in the criteria layer to construct a judgment matrix, calculate the weight vector of each key factor, and perform a consistency check to obtain the weight value of each key factor.

[0034] The initial emission factor and the actual values ​​of each key factor are obtained. The initial emission factor, the weight values ​​of each key factor, and the actual values ​​are then weighted and summed to obtain the localized emission factor.

[0035] In one embodiment, the reverse calculation of PM for agricultural machinery 2.5 Emission source strength is determined using the following formula:

[0036]

[0037] Where Q represents agricultural machinery PM 2.5 Emission source strength, For corrected PM 2.5 Concentration, x is the downwind distance, y is the crosswind distance, z is the monitoring height, ΔH is the effective height of the pollution source, μ is the average wind speed at the height of the exhaust pipe, σ y σ is the horizontal diffusion coefficient. z is the vertical diffusion coefficient.

[0038] In one embodiment, the Gaussian plume diffusion model is based on the fact that pollutants are mainly transported by advection at ambient wind speed in the downwind direction, while the turbulent diffusion in the crosswind and vertical directions is independent and both conform to a normal distribution law.

[0039] In one embodiment, the key factors affecting emission factors include agricultural machinery age, operating mode, ambient humidity, and ambient temperature. The weights of each factor are determined using the analytic hierarchy process (AHP), and localized calibration is performed based on the following formula:

[0040]

[0041]

[0042] Among them, EF calc The total emissions under a certain operating mode, ∆T is the operating period, and M f The total fuel consumption under this operating mode is measured by the vehicle's fuel consumption recorder. (EF) local K1 represents the localized emission factors, K2 represents the agricultural machinery age factor, K3 represents the operation mode factor, K4 represents the humidity factor, and K5 represents the temperature factor.

[0043] A type of agricultural machinery PM based on drone monitoring data 2.5 Pollution emission estimation system, including:

[0044] The drone monitoring module is used to collect pollutant concentration data downwind of the agricultural machinery operation area, and simultaneously collect meteorological environmental parameters and agricultural machinery operating parameters.

[0045] The data preprocessing module, connected to the UAV monitoring module, is used to complete missing data, remove abnormal data, and correct duplicate data of the pollutant concentration data to obtain preprocessed effective concentration data.

[0046] The source strength inversion module, connected to the data preprocessing module, is used to inversely calculate the PM2.5 concentration of agricultural machinery based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and based on the effective concentration data. 2.5 Emission source strength;

[0047] The factor calculation module, connected to the source strength inversion module, is used to calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery.

[0048] The localization calibration module, connected to the factor calculation module, is used to perform weight analysis on multiple factors affecting emission factors using the analytic hierarchy process (AHP) and to perform localization calibration on the initial emission factors to obtain localized emission factors.

[0049] Therefore, this application has the following beneficial effects:

[0050] First, this application solves the technical challenge of spatial misalignment between monitoring points and emission sources. It deeply integrates UAV mobile monitoring technology with a Gaussian plume diffusion model, constructing a Gaussian plume inverse model to invert emission source strength from downwind monitoring point concentration, fundamentally resolving the problem of non-overlapping monitoring points and emission sources in traditional monitoring methods. Experimental verification shows that the correlation coefficient R² between the model-calculated values ​​and measured values ​​exceeds 0.97, significantly improving estimation accuracy.

[0051] Secondly, it achieves localized and precise calibration of emission factors. This application innovatively introduces the analytic hierarchy process (AHP) to conduct a systematic weighted analysis of multiple factors such as agricultural machinery age, operating mode, and environmental humidity and temperature, thereby achieving localized correction of initial emission factors. Experimental results show that PM2.5 levels in agricultural machinery in the Zibo area... 2.5 The localized emission factor is 1.54 times higher than the recommended value in the national guidelines, which truly reflects the actual characteristics of the region, such as the high proportion of old agricultural machinery and the decline in emission performance.

[0052] Third, it reveals the key influencing patterns of agricultural machinery emissions. Using the method described in this application, the PM2.5 emissions from agricultural machinery are clearly identified for the first time. 2.5 The emission factor is significantly negatively correlated with the rated power. The emission intensity of the 49kW low-power model is 2.14 times that of the 570kW high-power model. The old low-power agricultural machinery was accurately identified as the core contributor to regional emissions, providing a quantitative basis for the agricultural machinery renewal and elimination policy.

[0053] Fourth, it is easy to operate, economical, practical, and highly scalable. This application uses drones as a mobile monitoring platform, which, compared to the traditional PEMS vehicle-mounted measurement method, has lower equipment costs, is easier to operate, and does not require complex deployment. It is suitable for large-scale agricultural machinery emission surveys at the county level, providing a replicable technical path for local ecological and environmental departments to compile high spatiotemporal resolution emission inventories. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 Agricultural machinery PM based on drone monitoring data 2.5 System flowchart of pollution emission estimation method;

[0056] Figure 2 It is an airborne PM for drones 2.5 Comparison chart of concentration detection module and fixed detection module;

[0057] Figure 3 The corrected airborne module measured data and the fixed tower PM 2.5 Comparison chart of measured data from the concentration detection module;

[0058] Figure 4 Agricultural machinery PM based on drone monitoring data 2.5 PM2.5 Emission Estimation Methods 2.5 Concentration correlation analysis graph;

[0059] Figure 5 Agricultural machinery PM based on drone monitoring data 2.5 Emission factor diagram of pollution emission estimation methods;

[0060] Figure 6 Agricultural machinery PM based on drone monitoring data 2.5 A comparison chart of emission factors in pollution emission estimation methods and national standard emission factors. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0063] Addressing the PM of existing agricultural machinery 2.5 Pollution emission monitoring suffers from three major technical shortcomings: spatial misalignment between monitoring points and emission sources, poor local adaptability of emission factors, and unclear mechanisms of multi-factor coupling. This application proposes an innovative three-pronged technical solution: "UAV mobile monitoring + Gaussian plume inverse calculation + analytic hierarchy process (AHP) local calibration." This solution first constructs a UAV-based PM2.5 monitoring system. 2.5 The monitoring system mounts sensors on the top of the drone at a predetermined safe distance from the center of the fuselage, effectively avoiding interference from the propeller downwash airflow. Data acquisition employs a phased strategy: first, background concentration data from the farmland is collected to eliminate regional baseline concentration interference; then, during agricultural machinery operation, the drone is controlled to hover at a predetermined position directly downwind, simultaneously collecting emission concentration data. For the collected data, preprocessing methods such as missing data completion, outlier removal, and duplicate data correction are used to ensure data quality. Secondly, based on a Gaussian plume diffusion model, the agricultural machinery exhaust pipe outlet is simplified as a point source pollution source. Combining on-site meteorological environmental parameters and agricultural machinery operating parameters, the emission source strength is calculated through inversion. This method establishes a mathematical correlation between the downwind monitoring point concentration and the emission source, fundamentally solving the technical problem of spatial non-overlap between monitoring points and emission sources. The correlation coefficient R² between the model-calculated values ​​and the measured values ​​exceeds 0.97, significantly improving estimation accuracy. Finally, the analytic hierarchy process (AHP) was used to conduct weighted analysis on multiple factors, including agricultural machinery age, operating mode, environmental humidity, and temperature, constructing a hierarchical model. Expert scoring was used to determine the weights of each factor, achieving precise localized calibration of emission factors. Experimental results show that the localized emission factors are 1.54 times higher than the national guidelines recommended values, accurately reflecting the actual characteristics of a high proportion of old agricultural machinery and declining emission performance in the region. This application, through a three-pronged technical solution, achieves a closed-loop process from data collection and diffusion simulation to factor calibration, providing quantitative evidence for county-level emission inventory compilation and agricultural machinery renewal policy formulation, demonstrating significant technological advancement and application value.

[0064] This application provides an embodiment of agricultural machinery PM based on drone monitoring data. 2.5 The pollution emission estimation method includes steps S1 to S5, referring to... Figure 1 , Figure 1 Agricultural machinery PM based on drone monitoring data 2.5 System flowchart of pollution emission estimation method.

[0065] Step S1: Obtain PM from the drone 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as meteorological environmental parameters and agricultural machinery operating parameters collected simultaneously.

[0066] Step S2: Preprocess the pollutant concentration data, including missing data completion, abnormal data removal and duplicate data correction, to obtain the preprocessed effective concentration data;

[0067] Step S3: Based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and using the effective concentration data, the PM2.5 concentration of the agricultural machinery is calculated in reverse. 2.5 Emission source strength;

[0068] Step S4: Calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery;

[0069] Step S5: Use the analytic hierarchy process (AHP) to perform weight analysis on the multiple factors affecting the emission factors, and perform localized calibration on the initial emission factors to obtain localized emission factors.

[0070] Specifically, in this embodiment, the present application provides an agricultural machinery PM based on UAV monitoring data. 2.5 The proposed pollution emission estimation method addresses the technical challenges of spatial misalignment between monitoring points and emission sources, and poor localization adaptability of emission factors, by constructing a three-pronged technical system: "UAV mobile monitoring + Gaussian plume back-calculation + analytic hierarchy process (AHP) localization calibration." This method combines this with the following... Figure 1 The system flowchart shown below provides a detailed explanation of the specific implementation steps of this embodiment.

[0071] Step S1: Data Acquisition

[0072] This step first obtains the drone-borne PM. 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as meteorological environmental parameters and agricultural machinery operating parameters collected simultaneously.

[0073] Specifically, this embodiment uses the Matrice M300RTK UAV as the flight platform, equipped with an industrial-grade PM 2.5Sensors. To avoid interference from the downwash airflow of the drone propellers on the monitoring data, the sensors are designed vertically and mounted on the top of the drone, 47.5 cm from the center of the fuselage. This area has been verified as a stable airflow zone through aerodynamic testing. Before testing, standard PM2.5 concentrations of 50 μg / m³ and 200 μg / m³ were used. 2.5 The gas calibration sensor is used to ensure an error of ≤5%.

[0074] Data collection employs a phased strategy of background and task-based approaches: first, background PM data for farmland is collected. 2.5 Concentration data was collected for at least 20 minutes to eliminate the interference of regional baseline concentrations on emission estimates. Subsequently, after the agricultural machinery began operation, a drone was controlled to hover in automatic tracking mode at a position 8 meters directly downwind of the agricultural machinery and 3 meters above the exhaust outlet, simultaneously collecting PM2.5 concentration data throughout the entire operation. 2.5 Concentration data are collected for no less than 180 minutes for each operating mode, covering effective operations and field turning conditions.

[0075] Simultaneously, a meteorological and environmental monitoring station was set up 10m from the edge of the work area to collect data on wind speed, wind direction, temperature, humidity, and atmospheric pressure. Operating parameters such as rated power, actual working time, fuel consumption, exhaust pipe height, flue gas outlet temperature, and flue gas outlet velocity were obtained through the agricultural machinery's onboard recorder. All data were sampled at a frequency of 1Hz, and the ground workstation received and stored the transmitted data in real time.

[0076] Step S2: Data Preprocessing

[0077] This step preprocesses the collected pollutant concentration data, including missing data completion, outlier removal, and duplicate data correction, to obtain preprocessed effective concentration data. Specifically, it includes:

[0078] Single missing values ​​are filled in using the average concentration values ​​at adjacent time points; two consecutive missing values ​​are filled in using bidirectional methods; three or more consecutive missing values ​​are filled in using linear interpolation.

[0079] Outliers are identified based on the Raida criterion, with data exceeding the mean ± 3 standard deviations considered outliers. When the number of outliers accounts for less than or equal to 5% of the total sample, the outliers are directly removed. If the outlier percentage exceeds 5%, the relevant time period is investigated, and linear interpolation is used to complete the data.

[0080] For consecutively repeated data, cubic spline interpolation is used for replacement, with the interpolation interval covering 300 valid data points before and after the repeated data.

[0081] Step S3: Invert emission source strength using Gaussian plume backpropagation model

[0082] This step, based on the Gaussian plume diffusion model and combining meteorological environmental parameters with agricultural machinery operating parameters, reverse-engineers the PM2.5 concentration of agricultural machinery based on the pre-processed effective concentration data. 2.5 Strong emission sources.

[0083] First, atmospheric diffusion conditions are determined: based on on-site ground wind speed, solar radiation intensity, and cloud cover data, the atmospheric stability level is determined according to the Pasquill-Gifford stability classification standard. In this embodiment, most operating conditions are classified as Level B, indicating instability. The Briggs close-range diffusion model is used to calculate the horizontal diffusion coefficient σ. y With vertical diffusion coefficient σ z .

[0084] Next, the effective height of the pollution source is calculated: the physical height of the exhaust pipe, the flue gas outlet temperature, the flue gas outlet velocity, and the inner diameter of the exhaust pipe outlet are obtained from the operating parameters of agricultural machinery, as well as the ambient temperature and ambient atmospheric pressure from the meteorological environmental parameters. The flue gas rise height is calculated according to the segmented rise logic in the "Technical Guidelines for Environmental Impact Assessment - Atmospheric Environment" (HJ 2.2-2018). The sum of the physical height of the exhaust pipe and the flue gas rise height is taken as the effective height H of the pollution source.

[0085] Finally, emission source intensity inversion calculation is performed: the preprocessed effective concentration data and horizontal diffusion coefficient σ are used. y Vertical diffusion coefficient σ z The effective height H of the pollution source, along with the average wind speed μ at the height of the exhaust pipe, the crosswind distance y (in this embodiment, the UAV is located directly downwind, so y=0), and the monitoring height z from the meteorological environmental parameters, are substituted into the Gaussian plume inverse model formula to obtain the PM2.5 concentration of the agricultural machinery. 2.5 Emission source strength Q.

[0086] Step S4: Initial Emission Factor Calculation

[0087] This step calculates the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of agricultural machinery.

[0088] First, calculate the total emissions. Then, multiply the obtained emission source intensity Q by the actual operating time ΔT of the agricultural machinery to obtain the PM2.5 concentration of the agricultural machinery during the actual operating period. 2.5 Total emissions.

[0089] Next, the unit fuel emission factor is calculated to obtain the total fuel consumption M of agricultural machinery during the actual operating period. fuel The total emissions are divided by the total fuel consumption to obtain the initial emission factor based on fuel consumption (in g / kg fuel).

[0090] Finally, a unit power conversion is performed to obtain the rated power P_rated of the agricultural machinery. The initial emission factor based on fuel consumption is multiplied by the total fuel consumption and then divided by the product of the rated power and the actual operating time to obtain the initial emission factor EF based on power. calc (Unit: g / kWh), serving as the basis for subsequent localization calibration.

[0091] Step S5: Localization and calibration of the Analytic Hierarchy Process (AHP)

[0092] This step uses the analytic hierarchy process (AHP) to perform weight analysis on the multiple factors affecting emission factors, and performs localized calibration on the initial emission factors to obtain localized emission factors.

[0093] First, a hierarchical model is constructed to identify the PM factors affecting agricultural machinery. 2.5 The key factors of emission factors include agricultural machinery age, operation mode, environmental humidity and environmental temperature. A hierarchical analysis structure model is constructed with localized emission factors as the target layer and key factors as the criterion layer.

[0094] Next, the factor weights are calculated. Each key factor in the criterion layer is compared pairwise using expert scoring to construct a judgment matrix. The weight vectors of each factor are then calculated, and a consistency check is performed. In this embodiment, the weight of the agricultural machinery age factor is 0.378, the weight of the operating mode factor is 0.320, the weight of the humidity factor is 0.197, and the weight of the temperature factor is 0.110.

[0095] Finally, localization calibration is performed to obtain the initial emission factor and the actual values ​​of each key factor. The initial emission factor, along with the weighted values ​​and actual values ​​of each key factor, are then weighted and summed to obtain the localized emission factor. Through these steps, this embodiment successfully achieves PM2.5 emission control for agricultural machinery. 2.5 Accurate estimation of pollution emissions. Experimental verification showed that the correlation coefficient R² between the model-calculated values ​​and the measured values ​​exceeded 0.97. The localized emission factor was 1.54 times higher than the national guidelines recommended value, accurately reflecting the high emission characteristics of aging agricultural machinery in the region, and providing a quantitative basis for county-level emission inventory compilation and agricultural machinery renewal policy formulation.

[0096] In one embodiment, the acquisition of UAV-borne PM 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as simultaneously collected meteorological environmental parameters and agricultural machinery operating parameters, including:

[0097] Before agricultural machinery operations, control the drone carrying PM 2.5 The monitoring system hovers at the preset monitoring points to collect background PM2.5 in farmland. 2.5Concentration data, collected for a duration of no less than the first preset duration, is used to eliminate the interference of regional baseline concentrations on emission estimation;

[0098] After the agricultural machinery begins operation, control the drone carrying PM 2.5 The monitoring system hovers in automatic tracking mode at a first preset distance directly downwind of the agricultural machinery and at a second preset height above the machinery's exhaust outlet, simultaneously collecting PM2.5 levels throughout the entire operation of the agricultural machinery. 2.5 Concentration data should be collected for a duration of no less than the second preset duration, covering both effective operations and field turning conditions.

[0099] During the data collection process, wind speed, wind direction, temperature, humidity and atmospheric pressure data were collected by meteorological and environmental monitoring stations set up at the edge of the work area. The rated power, actual working time, fuel consumption, exhaust pipe physical height, flue gas outlet temperature and flue gas outlet velocity parameters of the agricultural machinery were obtained by inputting through the agricultural machinery vehicle recorder or ground workstation.

[0100] Specifically, in this embodiment, the PM carried by the UAV is obtained. 2.5 The specific steps for the monitoring system to collect pollutant concentration data downwind of the agricultural machinery operation area, as well as the simultaneous collection of meteorological environmental parameters and agricultural machinery operating parameters, are as follows:

[0101] First, background concentration data is collected before agricultural machinery operations. Then, the drone carrying PM2.5 is controlled. 2.5 The monitoring system hovers at a pre-set monitoring point, located upwind of the work area or at the edge of the work area, to avoid interference from agricultural machinery emissions. Once the drone has stabilized, the PM2.5 system is activated. 2.5 The sensor continuously collects background concentration data from farmland for a period of not less than a first preset duration, which is 20 minutes in this embodiment. This background concentration data is used to subtract the regional baseline concentration in subsequent emission estimations, eliminating the influence of environmental background values ​​on the calculation results and ensuring the accuracy of the retrieved source strength.

[0102] Secondly, emissions data collection is performed after the agricultural machinery begins operation. This involves controlling the PM2.5 carried by the drone. 2.5 The monitoring system switches to automatic tracking mode, automatically following the movement of the agricultural machinery and hovering at a first preset distance directly downwind of the machinery, and at a second preset height above the machinery's exhaust outlet. In this embodiment, the first preset distance is 8 meters, and the second preset height is 3 meters. This position, verified by aerodynamic testing, effectively receives the diffused pollutant plume while avoiding direct interference from the propeller downwash airflow. The drone continuously collects PM2.5 data from this position throughout the entire agricultural machinery operation process. 2.5Concentration data are collected for a duration of no less than the second preset duration. In this embodiment, the second preset duration is 180 minutes to ensure that it covers complete operational conditions such as effective operations, field turns, and sudden load changes.

[0103] During the aforementioned background and operational data collection, environmental parameters and agricultural machinery operating parameters were collected simultaneously. Specifically, a meteorological environmental monitoring station was set up 10 meters from the edge of the operational area, avoiding areas prone to interference from agricultural machinery operations. Wind speed, wind direction, temperature, humidity, and atmospheric pressure data were continuously collected at a sampling frequency of 1Hz. Simultaneously, operating parameters such as the rated power, actual operating time, fuel consumption, exhaust pipe height, flue gas outlet temperature, and flue gas outlet velocity of the agricultural machinery were acquired through onboard recorders on the agricultural machinery or manual input from a ground workstation. All of the above data was synchronously received, stored, and time-correlated by the ground workstation, providing a complete set of input parameters for subsequent Gaussian plume inverse model calculations.

[0104] In one embodiment, the step of preprocessing the pollutant concentration data specifically includes:

[0105] For a single missing line, the average concentration values ​​at adjacent time points are used for completion; for two consecutive missing lines, bidirectional filling is used; and for three or more consecutive missing lines, linear interpolation is used for completion.

[0106] Outliers are identified based on the Raida criterion. When the number of outliers accounts for less than or equal to 5% of the total number of samples, the outliers are directly removed. When the number of outliers exceeds 5%, the relevant time period is investigated and then linear interpolation is used to complete the data.

[0107] For consecutively repeated data, cubic spline interpolation is used for replacement, and the interpolation interval covers the preset number of valid data before and after the repeated data.

[0108] Specifically, in this embodiment, the preprocessing steps for pollutant concentration data are as follows:

[0109] First, complete the missing data. (Drone-borne PM) 2.5 During data acquisition, the monitoring system may experience data loss due to signal transmission interruptions, sensor momentary failures, or other reasons. This embodiment employs a differentiated completion strategy based on the continuity of the missing data: for a single missing data point, the arithmetic mean of the concentration values ​​at the two adjacent time points is used for completion; for two consecutive missing data points, a bidirectional filling method using two valid data points before and after the missing time period is used; for three or more consecutive missing data points, linear interpolation is used, i.e., interpolating at equal intervals based on the changing trends of valid data before and after the missing time period to fill the missing interval. This completion strategy ensures data continuity while preserving the fluctuation characteristics of the original data to the greatest extent possible.

[0110] Secondly, outlier data removal and processing are performed. Based on the Raida criterion (also known as the 3σ criterion), outlier determination is performed on the preprocessed concentration data. This involves calculating the mean and standard deviation of the sample data, and identifying data exceeding the mean ± three times the standard deviation as outliers. When the number of outliers is less than or equal to 5% of the total sample size, it is considered random error and the outlier data is directly removed. When the number of outliers exceeds 5%, it is preliminarily determined that there may be instrument malfunction, sensor drift, or strong external interference. In this case, data processing is paused, and the monitoring equipment for the relevant time period is investigated for faults. After confirming data distortion, linear interpolation is used to complete the outlier intervals. This processing method preserves the integrity of normal data while effectively eliminating the interference of outlier data on subsequent inversion calculations.

[0111] Finally, duplicate data correction is performed. 2.5 When sensors are monitoring low concentrations or continuously for extended periods, they may exhibit continuous repetitive data due to limited measurement accuracy. This embodiment, referencing EU FILTER quality control requirements and US QARTOD durability testing standards, sets a threshold for determining continuous repetitive data: if approximately 500 consecutive identical values ​​appear within a 4-hour monitoring period, it is considered repetitive data. For such repetitive data, cubic spline interpolation is used for replacement, with the interpolation interval covering 300 valid data points before and after the repetitive data. A smooth interpolation curve is constructed using the changing trends of the valid data before and after the repetition, making the corrected data more consistent with the physical laws of actual concentration changes. Through these three preprocessing steps, noise and anomalies in the original monitoring data are effectively eliminated, providing a high-quality valid concentration dataset for subsequent Gaussian plume inverse modeling.

[0112] In one embodiment, the PM2.5 concentration of the agricultural machinery is calculated by combining the Gaussian plume diffusion model with the meteorological environmental parameters and the operating parameters of the agricultural machinery, and by using the effective concentration data to inversely calculate the PM2.5 concentration of the agricultural machinery. 2.5 The steps for strengthening emission source control include:

[0113] Based on surface wind speed, solar radiation intensity, and cloud cover data from meteorological environmental parameters, the atmospheric stability level under the current operating conditions is determined according to the Pasquill-Gifford stability rating standard.

[0114] Based on the atmospheric stability level, the horizontal and vertical diffusion coefficients were calculated using the Briggs close-range diffusion model.

[0115] The system obtains the physical height of the exhaust pipe, the flue gas outlet temperature, the flue gas outlet velocity, and the inner diameter of the exhaust pipe outlet from the operating parameters of agricultural machinery, as well as the ambient temperature and ambient atmospheric pressure from the meteorological environmental parameters. Based on the segmented lifting logic, the flue gas lifting height is calculated, and the sum of the physical height of the exhaust pipe and the flue gas lifting height is taken as the effective height of the pollution source.

[0116] Substituting the preprocessed effective concentration data, the horizontal diffusion coefficient, the vertical diffusion coefficient, the effective height of the pollution source, and the average wind speed, crosswind distance, and monitoring height at the exhaust pipe height from the meteorological environmental parameters, into the Gaussian plume inverse model formula, the PM2.5 concentration of agricultural machinery is obtained by inversely solving the model. 2.5 Strong emission sources.

[0117] Specifically, in this embodiment, the PM2.5 concentration of agricultural machinery is calculated by reverse calculation based on the Gaussian plume diffusion model. 2.5 The steps for identifying emission source intensity are as follows:

[0118] First, atmospheric stability is determined. Based on surface wind speed, solar radiation intensity, and cloud cover data collected from meteorological and environmental monitoring stations, the atmospheric stability level of the current operating conditions is determined according to the Pasquill-Gifford stability classification standard. This standard comprehensively considers three dimensions—solar radiation intensity, cloud cover, and surface wind speed—and classifies atmospheric stability into six levels, from A to F. In this embodiment, based on measured data, most operating conditions are classified as level B, while some strong radiation and low wind speed conditions are classified as level A.

[0119] Next, the diffusion coefficient is calculated. Based on the determined stability level, the horizontal diffusion coefficient σ is calculated using the Briggs close-range diffusion model. y With vertical diffusion coefficient σ z By selecting the formula corresponding to near-distance diffusion in the Briggs model, the horizontal and vertical diffusion coefficients can be obtained.

[0120]

[0121] In the formula, a and b are the parameters for calculating the diffusion coefficient, and the calculation of the parameters is shown in the table below:

[0122] Diffusion coefficient parameter table

[0123] Stability level <![CDATA[σ y ]]> <![CDATA[σ z ]]> A a=0.22, b=0.0001 a=0.20, b=0 B a=0.16, b=0.0001 a=0.12, b=0.0003

[0124] Next, the effective height of the pollution source is calculated. The physical height of the exhaust pipe, flue gas outlet temperature, flue gas outlet velocity, and exhaust pipe outlet inner diameter are obtained from the agricultural machinery operating parameters, as well as the ambient temperature and atmospheric pressure from the meteorological environmental parameters. Based on the segmented lift logic in the "Technical Guidelines for Environmental Impact Assessment—Atmospheric Environment" (HJ 2.2-2018), the flue gas heat release rate and buoyancy flux are first calculated, followed by the flue gas lift height. The Holland formula is used for calculations under windy conditions, while a special treatment method is used for calm conditions. Finally, the sum of the exhaust pipe physical height and the flue gas lift height is taken as the effective height of the pollution source.

[0125] Finally, the emission source strength was calculated by inversion. The preprocessed effective concentration data, horizontal diffusion coefficient, vertical diffusion coefficient, effective height of the pollution source, and meteorological environmental parameters including average wind speed at the exhaust pipe height, crosswind distance, and monitoring height were substituted into the Gaussian plume inverse calculation model formula. The crosswind distance was set to 0, and the monitoring height to 2.5 meters. The PM2.5 concentration of agricultural machinery was then calculated using an iterative solution method. 2.5 Emission source strength. Experimental verification shows that the correlation coefficient R² between the emission source strength obtained using the above method and the on-board measured value is above 0.97, with the error controlled within 5%, effectively solving the technical problem of spatial non-overlap between monitoring points and emission sources.

[0126] A fixed observation tower was selected as the reference platform to ensure that the horizontal distance between the UAV and the anemometer exceeded the safe distance and that the detection modules were at the same height, allowing for synchronous data recording. Two groups of 725 samples each were acquired under different environmental conditions. Moving averages were performed at 25-group intervals, and the results are as follows: Figure 2 As shown. Another dataset containing 725 samples was taken, and the measured values ​​of the UAV were substituted into the relational expression. The corrected data was then compared with the measured data from the fixed tower. The results are as follows. Figure 3 As shown, based on the constructed Gaussian plume inverse model, the emission source strength is accurately inverted through Pasquill-Gifford stability determination, Holland effective source height calculation, and Briggs diffusion coefficient correction. PM2.5 is simultaneously collected using a vehicle-mounted measurement method. 2.5 The emission levels at the source intensity point were verified. The results are as follows: Figure 4 As shown, the R² of the model-calculated value and the measured value is 0.97, and the error is controlled within 5%.

[0127] In one embodiment, the step of calculating the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery includes:

[0128] Multiplying the obtained emission source intensity by the actual operating time of the agricultural machinery, the PM2.5 concentration of the agricultural machinery during the actual operating period is calculated. 2.5 Total emissions;

[0129] Obtain the total fuel consumption of agricultural machinery during the actual operating period, and convert the PM... 2.5 The total emissions are divided by the total fuel consumption to calculate the initial emission factor based on fuel consumption;

[0130] The rated power of the agricultural machinery is obtained. The initial emission factor based on power is calculated by multiplying the initial emission factor based on fuel consumption by the total fuel consumption and then dividing by the product of the rated power and the actual operating time. This initial emission factor is then used as the initial emission factor for subsequent localization calibration.

[0131] Specifically, in this embodiment, the steps for calculating the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of agricultural machinery are as follows:

[0132] First, the total emissions are calculated. The emission source strength Q obtained from the Gaussian plume inverse model in step S3 is multiplied by the actual operating time ∆T of the agricultural machinery to calculate the PM2.5 concentration during the actual operating period. 2.5 Total emissions. In this embodiment, taking agricultural machinery No. 1 (corn planting mode) as an example, the emission source strength Q obtained by inversion is 0.00117 g / s, the actual operating time ∆T is 1800 seconds, and the total emissions are calculated to be 2.106 g. This total emission reflects the PM emitted into the atmosphere by the agricultural machinery during the operating period. 2.5 Total pollutant levels are the foundational data for subsequent emission factor calculations.

[0133] Secondly, calculate the initial emission factor based on fuel consumption. Obtain the total fuel consumption M of the agricultural machinery during the actual operating period. fuel This data was obtained through actual measurements using an onboard fuel consumption recorder for agricultural machinery. The PM2.5 calculated above... 2.5 The total emissions are divided by the total fuel consumption to obtain the initial emission factor based on fuel consumption, expressed in g / kg fuel. In this example, the total fuel consumption of agricultural machinery No. 1 is 0.59 kg, and the calculated initial emission factor based on fuel consumption is 3.57 g / kg fuel. This emission factor reflects the PM emissions per unit of fuel consumption. 2.5 Pollutant quality is one of the commonly used forms of expression for emission factors internationally.

[0134] Finally, the initial emission factor is converted to a power-based value. The rated power P_rated of the agricultural machinery is obtained from the machinery's nameplate or technical manual. The initial emission factor based on fuel consumption is multiplied by the total fuel consumption to obtain another expression of the total emissions. This is then divided by the product of the rated power and the actual operating time to calculate the initial emission factor based on power, expressed in g / kWh. In this embodiment, the rated power of agricultural machinery No. 1 is 49.0 kW, and the calculated initial emission factor based on power is 2.51 g / kWh. This unit is consistent with the emission factor unit recommended in the National Technical Guidelines for the Compilation of Air Pollutant Emission Inventories from Non-Road Mobile Sources, facilitating subsequent comparison with national standards and the localization of the analytic hierarchy process (AHP). Through the above three steps, the complete conversion from source strength to initial emission factor is completed, providing standardized basic data for subsequent multi-factor weighted analysis.

[0135] In one embodiment, the step of using the analytic hierarchy process (AHP) to perform weighted analysis on the multiple factors affecting the emission factors and to localize the initial emission factors to obtain localized emission factors includes:

[0136] Determine the PM affecting agricultural machinery 2.5 The key factors of emission factors are used as the target layer, and the key factors are used as the criterion layer to construct a hierarchical analysis structure model.

[0137] By using expert scoring or questionnaires, pairwise comparisons are made between the key factors in the criteria layer to construct a judgment matrix, calculate the weight vector of each key factor, and perform a consistency check to obtain the weight value of each key factor.

[0138] The initial emission factor and the actual values ​​of each key factor are obtained. The initial emission factor, the weight values ​​of each key factor, and the actual values ​​are then weighted and summed to obtain the localized emission factor.

[0139] Specifically, in this embodiment, the steps of using the analytic hierarchy process (AHP) to perform weight analysis on the multiple factors affecting emission factors and to localize the initial emission factors are as follows:

[0140] First, a hierarchical analysis model is constructed. This determines the PM (particulate matter) affecting agricultural machinery. 2.5 Key factors in emission assessment include agricultural machinery age, operating mode, ambient humidity, and ambient temperature. A hierarchical analysis model is constructed using localized emission factors as the target layer and the four key factors mentioned above as the criterion layer. Specifically, the agricultural machinery age factor reflects the impact of the machinery's service life on emission performance degradation; the longer the machinery is, the lower the combustion efficiency and the higher the emission factor. The operating mode factor reflects the differences in load intensity of agricultural machinery under different operating modes such as corn planting, wheat harvesting, and corn silage; the higher the load intensity, the lower the emission factor per unit power. The ambient humidity and temperature factors reflect the impact of meteorological conditions on engine combustion efficiency and emission characteristics; high humidity easily leads to engine carbon buildup, while low temperature affects combustion completeness.

[0141] Secondly, the weights of each factor are calculated. A judgment matrix is ​​constructed by comparing each key factor in the criterion layer pairwise using an expert scoring method. Multiple experts in the field are invited to assign values ​​to each factor based on its importance to emissions using a 1-9 scale. Taking the agricultural machinery age factor and the operating mode factor as an example, if the age factor is considered slightly more important than the operating mode factor, a value of 3 is assigned; if the age factor is considered significantly more important than the humidity factor, a value of 5 is assigned. After pairwise comparisons of all factors, a 4×4 judgment matrix is ​​constructed. The eigenvector corresponding to the largest eigenvalue of this matrix is ​​calculated, and after normalization, the weight vector of each factor is obtained. A consistency test is performed, and the consistency ratio CR is calculated. If CR < 0.1, the judgment matrix is ​​considered to have satisfactory consistency. In this embodiment, the weight of the agricultural machinery age factor is 0.378, the weight of the operating mode factor is 0.320, the weight of the humidity factor is 0.197, the weight of the temperature factor is 0.110, and the CR value is 0.021, which meets the consistency requirements.

[0142] Finally, localized calibration calculations are performed. The initial emission factor and the actual values ​​of each key factor are obtained. The initial emission factor is the power-based initial emission factor calculated in step S4; in this embodiment, it is 2.51 g / kWh for agricultural machinery No. 1. The machinery age factor is determined based on the actual service life, with 1.15 for 6 years of service; the operation mode factor is 1.05 based on the corn planting mode; the humidity factor is 1.02 based on the measured humidity of 64.6%; and the temperature factor is 0.98 based on the measured temperature of 299.44K. The initial emission factor, along with the weights and actual values ​​of each key factor, are weighted and summed. The localized emission factor equals the initial emission factor multiplied by the sum of the products of each factor's weight and its value. The calculated localized emission factor is 2.68 g / kWh, an increase of approximately 6.8% compared to the initial emission factor, accurately reflecting the high emission characteristics of older agricultural machinery in the region. Through these three steps, accurate calibration from the initial emission factor to the localized emission factor is completed, providing basic data consistent with local conditions for the compilation of the regional emission inventory.

[0143] In one embodiment, the reverse calculation of PM for agricultural machinery in step 3 2.5 Emission source strength is determined using the following formula:

[0144]

[0145] Where Q represents agricultural machinery PM 2.5 Emission source strength, For corrected PM 2.5 Concentration, x is the downwind distance, y is the crosswind distance, z is the monitoring height, ΔH is the effective height of the pollution source, μ is the average wind speed at the height of the exhaust pipe, σ y σ is the horizontal diffusion coefficient. z is the vertical diffusion coefficient.

[0146] Specifically, in this embodiment, the reverse calculation of PM for agricultural machinery described in step 3... 2.5 The emission source strength is calculated using a Gaussian plume inverse model formula. This formula is based on the fundamental physical assumptions of pollutant diffusion: pollutants are primarily transported downstream by advection at ambient wind speed, while turbulent diffusion in the crosswind and vertical directions is independent and both conform to a normal distribution. The exhaust pipe outlet of agricultural machinery is simplified as a point source pollution source, and its emitted PM2.5... 2.5 The concentration of a smoke plume at any spatial point during its leeward diffusion can be described by this formula. The meanings and methods of obtaining the parameters in the formula are as follows:

[0147] Q represents the agricultural machinery PM to be solved. 2.5 Emission source strength, expressed in g / s, is the target value for the inversion calculation;

[0148] C c PM after step 2 preprocessing and air-to-ground comparison correction 2.5 Concentration, in μg / m³, was obtained from actual measurements by onboard sensors of the UAV;

[0149] x is the downwind distance, that is, the horizontal distance between the monitoring point and the emission source in the wind direction. In this embodiment, the drone hovers 8m downwind of the agricultural machinery, so x=8m;

[0150] y is the crosswind distance, which is the vertical distance of the monitoring point from the center axis of the wind direction. In this embodiment, the UAV is located on the center axis of the plume, so y=0.

[0151] z represents the monitoring height, which is the vertical height of the drone's PM2.5 sensor above the ground. In this embodiment, it is taken as 2.5m.

[0152] μ is the average wind speed at the height of the exhaust pipe, in m / s, which is obtained by actual measurement from the meteorological and environmental monitoring station set up at the edge of the work area;

[0153] σ y σ is the horizontal diffusion coefficient. z The vertical diffusion coefficient is calculated using the Briggs short-range diffusion model based on the Pasquill-Gifford atmospheric stability level determined in step 3.1.

[0154] H is the effective height of the pollution source, in meters, which is equal to the sum of the physical height of the exhaust pipe h and the rise height of the flue gas ΔH, and is calculated according to the segmented rise logic in step 3.2.

[0155] This embodiment uses agricultural machinery No. 1 (corn planting mode) as an example for illustration: Field measurement C c=125μg / m³, x=8m, y=0, z=2.5m, μ=1.91m / s; Atmospheric stability is determined to be level B, and σ is calculated. y =0.93, σ z =0.58; H = 3.4m was calculated after flue gas lift. Substituting the above parameters into the formula, the emission source strength Q = 0.00117g / s was obtained through iterative solution.

[0156] To verify the accuracy of the inversion results, this embodiment uses an onboard PEMS system to simultaneously measure PM at the exhaust pipe of agricultural machinery. 2.5 The emission source strength was measured to be 0.00121 g / s. The relative error between the inverted value and the measured value was 3.3%, and the correlation coefficient R² was above 0.97, verifying the applicability and accuracy of the Gaussian plume inverse model in agricultural machinery emission monitoring scenarios, and effectively solving the technical problem of spatial non-overlap between monitoring points and emission sources.

[0157] In one embodiment, the Gaussian plume diffusion model described in step 3 is based on the fact that pollutants are mainly transported by advection at ambient wind speed in the downwind direction, while the turbulent diffusion in the crosswind and vertical directions is independent of each other and both conform to the normal distribution law.

[0158] Specifically, in this embodiment, the Gaussian plume diffusion model used in step S3 is based on the following core assumptions: pollutants are mainly transported by advection at ambient wind speed in the downwind direction, and the turbulent diffusion in the crosswind direction and vertical direction is independent of each other and both conform to the normal distribution law.

[0159] When agricultural machinery emits PM 2.5 After pollutants enter the atmosphere, in the downwind direction (x direction), the transport of pollutants is mainly controlled by the average environmental wind speed, and the contribution of turbulent diffusion is relatively small. Therefore, downwind diffusion can be ignored, and only advection transport is considered. In the crosswind direction (y direction) and vertical direction (z direction), the pollutant concentration distribution follows a Gaussian distribution (normal distribution), and the turbulent diffusion in the two directions is independent of each other. That is, the diffusion in the crosswind direction does not affect the diffusion in the vertical direction, and vice versa.

[0160] Based on the above assumptions, the exhaust pipe outlet of agricultural machinery can be simplified as a point source pollution source, and the PM emitted from it... 2.5 During the leeward diffusion of the smoke plume, the concentration at any spatial point can be described by the Gaussian smoke plume model formula. The validity of this assumption has been widely verified in the field of atmospheric environment, and it is particularly suitable for short-range, close-range monitoring of agricultural machinery emissions in the field. In this embodiment, the drone hovers 8m leeward from the agricultural machinery, i.e., at a crosswind distance y=0, located on the central axis of the smoke plume. This is combined with the average wind speed μ at the exhaust pipe height measured on-site and the horizontal diffusion coefficient σ determined based on atmospheric stability. y With vertical diffusion coefficient σ zThe effective height H of the pollution source is calculated, and the above parameters are substituted into the Gaussian plume inverse model formula to obtain the emission source strength Q by inverse solution.

[0161] In one embodiment, the key factors affecting emission factors in step 5 include agricultural machinery age, operating mode, ambient humidity, and ambient temperature. The weights of each factor are determined using the analytic hierarchy process (AHP), and localized calibration is performed based on the following formula:

[0162]

[0163]

[0164] Among them, EF calc The total emissions under a certain operating mode, ∆T is the operating period, and M f The total fuel consumption under this operating mode is measured by the vehicle's fuel consumption recorder. (EF) local K1 represents the localized emission factors, K2 represents the agricultural machinery age factor, K3 represents the operation mode factor, K4 represents the humidity factor, and K5 represents the temperature factor.

[0165] Specifically, in this embodiment, the key factors affecting emission factors in step 5 include agricultural machinery age factor, operation mode factor, environmental humidity factor and environmental temperature factor, and the weight of each factor is determined by the analytic hierarchy process, and localized calibration is achieved by weighted summation.

[0166] First, this embodiment identifies four key influencing factors: the agricultural machinery age factor K1 reflects the impact of the agricultural machinery's service life on emission performance degradation. The longer the vehicle's age, the more severe the engine wear, the lower the combustion efficiency, and the higher the emission factor. The operation mode factor K2 reflects the differences in the intensity of different operation loads. The corn planting mode has large load fluctuations and a high proportion of non-steady-state conditions, while the wheat harvesting mode has relatively stable loads, and the corn stalk silage mode is mostly high-power continuous operation. The ambient humidity factor K3 reflects that high humidity can easily lead to carbon deposits in the engine, affecting the completeness of combustion. The ambient temperature factor K4 reflects poor fuel atomization and decreased combustion efficiency under low-temperature conditions.

[0167] This embodiment uses the analytic hierarchy process (AHP) to determine the weights of each factor. First, an AHP structure model is constructed, with localized emission factors as the target layer and four key factors as the criterion layer. Ten experts in agricultural machinery and atmospheric environment were invited to conduct pairwise comparisons of each factor in the criterion layer using a 1-9 scaling method, constructing a 4×4 judgment matrix. The eigenvector corresponding to the largest eigenvalue of the matrix is ​​calculated, normalized, and the weight vector of each factor is obtained, followed by a consistency check. The calculated weights are: agricultural machinery age factor w1 = 0.378, operating mode factor w2 = 0.320, humidity factor w3 = 0.197, temperature factor w4 = 0.110, and the consistency ratio CR = 0.021 < 0.1, meeting the consistency requirements.

[0168] In the localized calibration calculation, the initial emission factor EF is first calculated through step 4. calc Taking the No. 1 agricultural machine (corn planting mode, 49.0kW) as an example, EF calc =2.51g / kWh. Then, the actual values ​​of each factor were obtained: the agricultural machinery age factor K1 was 1.15 based on 6 years of service; the operation mode factor K2 was 1.05 based on the corn planting mode; the environmental humidity factor K3 was 1.02 based on the measured humidity of 64.6%; and the environmental temperature factor K4 was 0.98 based on the measured temperature of 299.44K. The calculated localized emission factor was 2.71g / kWh, an increase of approximately 8.0% compared to the initial emission factor of 2.51g / kWh. This accurately reflects the actual characteristics of the region, including a high proportion of old agricultural machinery and declining emission performance, providing basic data consistent with local conditions for the compilation of the regional emission inventory.

[0169] The operating modes of the agricultural machinery include at least one of corn planting, wheat harvesting, and corn silage. These three operating modes cover the main types of agricultural machinery operations in the wheat-corn rotation area of ​​central Shandong, accounting for more than 78% of the total agricultural machinery operations in the region. Corn planting involves multiple processes such as sowing, fertilizing, and covering the soil simultaneously, resulting in large load fluctuations and frequent engine operation in unsteady conditions. Wheat harvesting has a relatively stable operating load and a more stable engine condition. Corn silage is mostly high-power continuous operation with high combustion efficiency. These three modes represent different load intensities and operating conditions, comprehensively reflecting the differences in agricultural machinery emissions across the region.

[0170] Meanwhile, the operating parameters of the agricultural machinery include the rated power, actual operating time, fuel consumption, exhaust pipe physical height, flue gas outlet temperature, and flue gas outlet velocity. The rated power is obtained from the agricultural machinery nameplate or technical manual and is used for subsequent conversion of the emission factor per unit power. The actual operating time is determined by recording the start and end times of the drone tracking operation using a ground workstation. Fuel consumption is obtained through actual measurement using an onboard fuel consumption recorder, reflecting the actual fuel consumption during the operating period. The exhaust pipe physical height is obtained through on-site measurement and is used to calculate the effective height of the pollution source. The flue gas outlet temperature and flue gas outlet velocity are obtained through actual measurement using a temperature sensor and flow meter installed at the exhaust pipe outlet, and are key parameters for calculating the flue gas rise height. Figure 5 As shown, two points are evident: the emission intensity varies significantly among different modes; corn planting (Mode 1) has the highest emissions, wheat harvesting (Mode 2) is in the middle, and corn silage (Mode 3) has the lowest. The 49kW planter model has the highest emission factor at 2.51 g / kWh. Emissions are strongly negatively correlated with power: the higher the rated power of the agricultural machinery, the lower the emission factor. The emission intensity of the 49kW model is 2.14 times that of the 570kW high-power model, primarily due to the higher combustion efficiency and superior exhaust aftertreatment technology of the high-power model, highlighting the high emission characteristics of older, low-power agricultural machinery in the region.

[0171] like Figure 6 As shown, the localized emission factor is significantly negatively correlated with the rated power of agricultural machinery: the emission factor of the 49kW low-power model reaches 2.51g / kWh, while that of the 570kW high-power model is only 1.17g / kWh, the former being 2.14 times that of the latter. Compared with the national "Technical Guidelines for the Compilation of Air Pollutant Emission Inventories for Non-Road Mobile Sources," the localized emission factor in this study is significantly higher. Taking the 75-130kW power range as an example, the national guidelines state that the emission factor for this range is 0.8-0.95g / kWh, while the measured value for the same range in this study is 1.26-1.86g / kWh, which is 1.54 times higher than the national guidelines. The differences stem from: firstly, field operations involve a large number of non-steady-state conditions, resulting in emission intensity higher than that of bench steady-state tests; secondly, the over-service of regional agricultural machinery leads to a decline in emission performance, with the emission factor of the National III emission standard models produced in 2017 being 40% higher than that of new machines; and thirdly, regional differences in fuel quality. Compared with the emission factors of construction machinery, the PM emission factor of 75-130kW construction machinery reached 3.36g / kWh, which is higher than that of agricultural machinery in the same power range in this study (1.26-1.86g / kWh). The difference is due to the more severe working conditions of construction machinery and the fact that the in-service equipment is mainly of the National I / II stage.

[0172] A type of agricultural machinery PM based on drone monitoring data 2.5 Pollution emission estimation system, including:

[0173] The drone monitoring module is used to collect pollutant concentration data downwind of the agricultural machinery operation area, and simultaneously collect meteorological environmental parameters and agricultural machinery operating parameters.

[0174] The data preprocessing module, connected to the UAV monitoring module, is used to complete missing data, remove abnormal data, and correct duplicate data of the pollutant concentration data to obtain preprocessed effective concentration data.

[0175] The source strength inversion module, connected to the data preprocessing module, is used to inversely calculate the PM2.5 concentration of agricultural machinery based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and based on the effective concentration data. 2.5 Emission source strength;

[0176] The factor calculation module, connected to the source strength inversion module, is used to calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery.

[0177] The localization calibration module, connected to the factor calculation module, is used to perform weight analysis on multiple factors affecting emission factors using the analytic hierarchy process (AHP) and to perform localization calibration on the initial emission factors to obtain localized emission factors.

[0178] Specifically, the agricultural machinery PM based on UAV monitoring data provided in this embodiment 2.5 The pollution emission estimation system, in which its various modules work collaboratively, is implemented as follows:

[0179] The drone monitoring module is the data acquisition front-end of this system, consisting of the drone platform and PM. 2.5 It consists of sensors, a meteorological and environmental monitoring station, and a ground workstation. The UAV platform uses the Matrice M300RTK and is equipped with an industrial-grade PM2.5T engine. 2.5 The sensor, featuring a vertical design and validated through indoor aerodynamic testing, is mounted on the top of the drone at a distance of 47.5 cm from the center of the fuselage. This area offers stable airflow, effectively avoiding interference from the propeller downwash. Before agricultural machinery operations, the drone hovers 20 m upwind of the work area to collect background PM2.5 levels in the farmland. 2.5 Concentration data was collected over a 20-minute period to eliminate interference from the regional baseline concentration. After the agricultural machinery began operation, the drone switched to automatic tracking mode, hovering 8 meters downwind of the machinery and 3 meters above the exhaust vent, simultaneously collecting PM2.5 concentration data throughout the entire operation. 2.5Concentration data is collected for at least 180 minutes for each operating mode, covering effective operation and field turning conditions. Simultaneously, a meteorological monitoring station set up 10m from the edge of the operating area continuously collects wind speed, wind direction, temperature, humidity, and atmospheric pressure data at a frequency of 1Hz. The agricultural machinery's onboard recorder records real-time operating parameters such as rated power, actual operating time, fuel consumption, exhaust pipe height, flue gas outlet temperature, and flue gas outlet velocity. The ground workstation simultaneously receives and stores all of the above data, establishing a time-correlated dataset.

[0180] The data preprocessing module is connected to the UAV monitoring module. After receiving the raw monitoring data, it performs three preprocessing steps: First, missing data is filled in. A single missing data is filled in using the average of adjacent time points. Two consecutive missing data points are filled in using bidirectional interpolation. Three or more consecutive missing data points are filled in using linear interpolation. Second, outliers are identified based on the Laida criterion. When the proportion of outliers is ≤5%, they are directly removed. When the proportion is >5%, the fault is investigated and linear interpolation is used to fill in the outliers. Finally, continuous duplicate data are replaced using cubic spline interpolation. The interpolation interval covers 300 valid data points before and after the duplicate data to ensure that the data quality meets the requirements of subsequent inversion.

[0181] The source strength inversion module is connected to the data preprocessing module, and inversely calculates the emission source strength based on the Gaussian plume diffusion model. This module first determines the atmospheric stability level according to the Pasquill-Gifford standard based on surface wind speed, solar radiation intensity, and cloud cover data, and then calculates the horizontal diffusion coefficient σ using the Briggs near-field diffusion model. y With vertical diffusion coefficient σ z Then, based on the segmented lift logic in the "Technical Guidelines for Environmental Impact Assessment - Atmospheric Environment" (HJ 2.2-2018), and combining parameters such as the physical height of the exhaust pipe, flue gas outlet temperature, flue gas outlet velocity, ambient temperature, and atmospheric pressure, the flue gas lift height is calculated to obtain the effective height H of the pollution source. Finally, the pre-processed effective concentration data, diffusion coefficient, effective height, average wind speed, crosswind distance (0 in this embodiment), and monitoring height (2.5m) are substituted into the Gaussian plume inverse calculation model formula to obtain the emission source strength Q by reverse calculation.

[0182] The factor calculation module is connected to the source strength inversion module, calculating the initial emission factor based on the emission source strength. First, the emission source strength Q is multiplied by the actual operating duration ΔT to obtain the PM2.5 emission factor. 2.5 Total emissions; then divide total emissions by total fuel consumption M. fuel The initial emission factor based on fuel consumption is obtained; finally, the initial emission factor based on fuel consumption is multiplied by the total fuel consumption and then divided by the product of rated power and actual operating time to convert it into the initial emission factor EF based on power. calc (Unit: g / kWh) for easy comparison with national emission standards.

[0183] The localization calibration module is connected to the factor calculation module, and uses the analytic hierarchy process (AHP) to localize the initial emission factors. This module first identifies the agricultural machinery age factor, operating mode factor, environmental humidity factor, and environmental temperature factor as key influencing factors, and constructs an AHP structure model with the localized emission factors as the target layer. A judgment matrix is ​​constructed using expert scoring, and the weights of each factor are calculated (in this embodiment, 0.378, 0.320, 0.197, and 0.110), followed by consistency checks. Finally, the initial emission factors and the actual values ​​of each factor are obtained; for example, the vehicle age factor is set to 1.15 based on service life, and the operating mode factor is set to 1.05 based on corn planting mode. The localized emission factors are obtained through weighted summation. Through the collaborative work of these five modules, this system successfully constructs a closed-loop estimation system from data acquisition to result output, effectively solving the problem of spatial misalignment between monitoring points and emission sources, and achieving accurate localization calibration of emission factors.

[0184] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0185] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0186] It should be particularly noted that, through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, or of course, by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An agricultural machinery PM based on drone monitoring data 2.5 The method for estimating pollution emissions is characterized by, Includes the following steps: Step S1: Obtain UAV-borne PM 2.5 The monitoring system collects pollutant concentration data collected by the monitoring system in the downwind direction of the agricultural machinery operation area, and synchronously collects meteorological environmental parameters and agricultural machinery working condition parameters. Step S2: Preprocess the pollutant concentration data, including missing data completion, abnormal data removal and duplicate data correction, to obtain the preprocessed effective concentration data; Step S3: Based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and using the effective concentration data, the PM2.5 concentration of the agricultural machinery is calculated in reverse. 2.5 Emission source strength; Step S4: Calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery; Step S5: Use the analytic hierarchy process (AHP) to perform weight analysis on the multiple factors affecting the emission factors, and perform localized calibration on the initial emission factors to obtain localized emission factors.

2. The estimation method according to claim 1, characterized in that, The acquisition of drone-borne PM 2.5 The monitoring system collects pollutant concentration data downwind of the agricultural machinery operation area, as well as simultaneously collected meteorological environmental parameters and agricultural machinery operating parameters, including: Before agricultural machinery operations, control the drone carrying PM 2.5 The monitoring system hovers at the preset monitoring points to collect background PM2.5 in farmland. 2.5 Concentration data, collected for a duration of no less than the first preset duration, is used to eliminate the interference of regional baseline concentrations on emission estimation; After the agricultural machinery begins operation, control the drone carrying PM 2.5 The monitoring system hovers in automatic tracking mode at a first preset distance directly downwind of the agricultural machinery and at a second preset height above the machinery's exhaust outlet, simultaneously collecting PM2.5 levels throughout the entire operation of the agricultural machinery. 2.5 Concentration data should be collected for a duration of no less than the second preset duration, covering both effective operations and field turning conditions. During the data collection process, wind speed, wind direction, temperature, humidity and atmospheric pressure data were collected by meteorological and environmental monitoring stations set up at the edge of the work area. The rated power, actual working time, fuel consumption, exhaust pipe physical height, flue gas outlet temperature and flue gas outlet velocity parameters of the agricultural machinery were obtained by inputting through the agricultural machinery vehicle recorder or ground workstation.

3. The estimation method according to claim 1, characterized in that, The preprocessing steps for the pollutant concentration data specifically include: For a single missing line, the average concentration values ​​at adjacent time points are used for completion; for two consecutive missing lines, bidirectional filling is used; and for three or more consecutive missing lines, linear interpolation is used for completion. Outliers are identified based on the Raida criterion. When the number of outliers accounts for less than or equal to 5% of the total number of samples, the outliers are directly removed. When the number of outliers exceeds 5%, the relevant time period is investigated and then linear interpolation is used to complete the data. For consecutively repeated data, cubic spline interpolation is used for replacement, and the interpolation interval covers the preset number of valid data before and after the repeated data.

4. The estimation method according to claim 1, characterized in that, The method uses a Gaussian plume diffusion model, combined with meteorological environmental parameters and agricultural machinery operating parameters, and inversely calculates the PM2.5 concentration of the agricultural machinery based on the effective concentration data. 2.5 The steps for strengthening emission source control include: Based on surface wind speed, solar radiation intensity, and cloud cover data from meteorological environmental parameters, the atmospheric stability level under the current operating conditions is determined according to the Pasquill-Gifford stability rating standard. Based on the atmospheric stability level, the horizontal and vertical diffusion coefficients were calculated using the Briggs close-range diffusion model. The system obtains the physical height of the exhaust pipe, the flue gas outlet temperature, the flue gas outlet velocity, and the inner diameter of the exhaust pipe outlet from the operating parameters of agricultural machinery, as well as the ambient temperature and ambient atmospheric pressure from the meteorological environmental parameters. Based on the segmented lifting logic, the flue gas lifting height is calculated, and the sum of the physical height of the exhaust pipe and the flue gas lifting height is taken as the effective height of the pollution source. Substituting the preprocessed effective concentration data, the horizontal diffusion coefficient, the vertical diffusion coefficient, the effective height of the pollution source, and the average wind speed, crosswind distance, and monitoring height at the exhaust pipe height from the meteorological environmental parameters, into the Gaussian plume inverse model formula, the PM2.5 concentration of agricultural machinery is obtained by inversely solving the model. 2.5 Strong emission sources.

5. The estimation method according to claim 1, characterized in that, The step of calculating the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of agricultural machinery includes: Multiplying the obtained emission source intensity by the actual operating time of the agricultural machinery, the PM2.5 concentration of the agricultural machinery during the actual operating period is calculated. 2.5 Total emissions; Obtain the total fuel consumption of agricultural machinery during the actual operating period, and convert the PM... 2.5 The total emissions are divided by the total fuel consumption to calculate the initial emission factor based on fuel consumption; The rated power of the agricultural machinery is obtained. The initial emission factor based on power is calculated by multiplying the initial emission factor based on fuel consumption by the total fuel consumption and then dividing by the product of the rated power and the actual operating time. This initial emission factor is then used as the initial emission factor for subsequent localization calibration.

6. The estimation method according to claim 1, characterized in that, The steps of using the analytic hierarchy process (AHP) to perform weighted analysis on the multiple factors affecting emission factors, and to localize the initial emission factors to obtain localized emission factors, include: Determine the PM affecting agricultural machinery 2.5 The key factors of emission factors are used as the target layer, and the key factors are used as the criterion layer to construct a hierarchical analysis structure model. By using expert scoring or questionnaires, pairwise comparisons are made between the key factors in the criteria layer to construct a judgment matrix, calculate the weight vector of each key factor, and perform a consistency check to obtain the weight value of each key factor. The initial emission factor and the actual values ​​of each key factor are obtained. The initial emission factor, the weight values ​​of each key factor, and the actual values ​​are then weighted and summed to obtain the localized emission factor.

7. The estimation method according to claim 4, characterized in that, The reverse calculation of PM2.5 emission source strength of agricultural machinery in step 3 uses the following formula: ； Where Q represents agricultural machinery PM 2.5 Emission source strength, For corrected PM 2.5 Concentration, x is the downwind distance, y is the crosswind distance, z is the monitoring height, ΔH is the effective height of the pollution source, μ is the average wind speed at the height of the exhaust pipe, σ y σ is the horizontal diffusion coefficient. z is the vertical diffusion coefficient.

8. The estimation method according to claim 4, characterized in that, The Gaussian plume diffusion model described in step 3 is based on the fact that pollutants are mainly transported by advection at ambient wind speed in the downwind direction, while the turbulent diffusion in the crosswind and vertical directions is independent and both conform to the normal distribution law.

9. The estimation method according to claim 6, characterized in that, The key factors affecting emission factors mentioned in step 5 include agricultural machinery age, operating mode, ambient humidity, and ambient temperature. The weights of each factor are determined using the analytic hierarchy process (AHP), and localized calibration is performed based on the following formula: ； ； Among them, EF calc The total emissions under a certain operating mode, ∆T is the operating period, and M f The total fuel consumption under this operating mode is measured by the vehicle's fuel consumption recorder. (EF) local K1 represents the localized emission factors, K2 represents the agricultural machinery age factor, K3 represents the operation mode factor, K4 represents the humidity factor, and K5 represents the temperature factor.

10. An agricultural machinery PM based on UAV monitoring data 2.5 Pollution emission estimation system, characterized in that, include: The drone monitoring module is used to collect pollutant concentration data downwind of the agricultural machinery operation area, and simultaneously collect meteorological environmental parameters and agricultural machinery operating parameters. The data preprocessing module, connected to the UAV monitoring module, is used to complete missing data, remove abnormal data, and correct duplicate data of the pollutant concentration data to obtain preprocessed effective concentration data. The source strength inversion module, connected to the data preprocessing module, is used to inversely calculate the PM2.5 concentration of agricultural machinery based on the Gaussian plume diffusion model, combined with the meteorological environmental parameters and the agricultural machinery operating parameters, and based on the effective concentration data. 2.5 Emission source strength; The factor calculation module, connected to the source strength inversion module, is used to calculate the initial emission factor based on the emission source strength and the actual operating time and fuel consumption of the agricultural machinery. The localization calibration module, connected to the factor calculation module, is used to perform weight analysis on multiple factors affecting emission factors using the analytic hierarchy process (AHP) and to perform localization calibration on the initial emission factors to obtain localized emission factors.

Images