Calculation Method and System for Inverting Carbon Emission Intensity of Strong Point Sources Based on UAV Sampling

Abstract

The present invention discloses a method and system for calculating the carbon emission intensity of strong point sources by inversion based on UAV sampling. The method includes the following steps: Step S1, collecting sampling data, where the sampling data includes position information, CO2 concentration, and meteorological data; Step S2, establishing a corresponding carbon emission diffusion model of strong point sources according to the collected sampling data; Step S3, solving the carbon emission intensity of power plants in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm; Step S4, evaluating the accuracy of the carbon emission intensity of power plants according to the verification dataset. The technical beneficial effect of the present invention is that it can effectively and quickly calculate the carbon emission intensity of strong point sources based on the measured data collected by UAV sampling equipment and combined with an adaptive quantitative evaluation model.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and specifically refers to a method and system for inverting the carbon emission intensity of strong point sources based on an unmanned aerial vehicle sampling device. Background Art

[0002] The total carbon emissions of strong point sources exceed one-third of the total carbon emissions in human production activities. Currently, the assessment of such carbon emissions can only be estimated with low accuracy and low temporal resolution through emission inventories. Since strong carbon emission sources are often caused by the combustion of fossil fuels and the like, their emission characteristics have large fluctuations. Therefore, how to quickly and accurately invert the carbon emissions of strong point sources has become an important part of China's carbon accounting work.

[0003] Nowadays, the "top-down" assessment method of emission sources through concentration data has gradually developed into a reliable means, with real-time and high accuracy. Satellite remote sensing, under specific detection orbits and meteorological conditions, can achieve quantitative calculation of point sources based on the obtained greenhouse gas concentrations. However, current greenhouse gas satellites are easily affected by the observation environment, such as clouds, aerosols, and solar radiation intensity, and cannot achieve directional monitoring of specific areas. The airborne remote sensing method can obtain large-scale carbon fluxes, but the technical requirements of the airborne remote sensing method are relatively high and the cost is high, and long-term continuous monitoring cannot be achieved. Ground-based equipment, fixed-station in-situ measurement instruments, and multi-sensor networking can evaluate the regional carbon flux, but the cost performance of establishing a sensor network for strong point source emissions is relatively low. The unmanned aerial vehicle gas sampling system has high flexibility and can collect the concentration distributions in different spaces, can fully obtain the CO2 diffusion characteristics caused by point source carbon emissions, and combined with the diffusion model and the inversion algorithm, can effectively reconstruct the carbon diffusion model and obtain the carbon emission intensity.

[0004] Therefore, this patent provides a set of solutions for quickly and accurately obtaining the carbon emissions of strong point sources based on an unmanned aerial vehicle sampling system and a stable carbon emission accounting method. The emission inversion framework recommended by this solution can greatly reduce the requirements for prior knowledge and fully achieve adaptive global optimal adjustment to ensure the least impact of the concentration sampling accuracy and meteorological acquisition accuracy in the actual application process. Summary of the Invention

[0005] To solve the problem in the prior art that it is impossible to monitor the emissions of strong point sources with high accuracy and high applicability, the present invention provides a method and system for inverting the carbon emission intensity of strong point sources based on an unmanned aerial vehicle sampling device.

[0006] The present invention provides a method for calculating the carbon emission intensity of strong point sources inverted based on unmanned aerial vehicle sampling, and the method includes the following steps:

[0007] Step S1: Collect sampling data, where the sampling data includes location information, CO2 concentration, and meteorological data;

[0008] Step S2: Based on the collected sampling data, establish a corresponding strong point source carbon emission diffusion model;

[0009] Step S3: Solve the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm optimization algorithm and the trust region algorithm;

[0010] Step S4: Evaluate the accuracy of the carbon emission intensity of the power plant according to the validation data set.

[0011] Preferably, collecting sampling data according to the unmanned aerial vehicle sampling device includes: designing a reasonable unmanned aerial vehicle flight sampling route according to the terrain characteristics and meteorological field characteristics of the strong point source to be evaluated, and obtaining the concentration diffusion characteristics of the carbon emission of the strong point source.

[0012] Preferably, establishing the corresponding strong point source carbon emission diffusion model includes: according to the carbon emission characteristics of the strong point source, combined with the spatial data collected by the unmanned aerial vehicle sampling device, selecting the Gaussian diffusion model to conduct mathematical modeling on the carbon emission of the power plant.

[0013] Preferably, establishing the corresponding strong point source carbon emission diffusion model includes:

[0014] S2.1: Establish a diffusion coordinate system according to the location of the strong emission source and the measured wind direction:

[0015] S2.2: Based on the diffusion coordinate system established in step S2.1, establish a strong point source carbon emission diffusion model.

[0016] Preferably, the strong point source carbon emission diffusion model is expressed as follows,

[0017]

[0018] where (x, y, z) are the coordinates of the measurement point of the unmanned aerial vehicle sampling device, C(x, y, z) is the CO2 concentration at the coordinates (x, y, z), q is the emission intensity, u is the wind speed, H is the effective emission height of the carbon emission of the power plant, σ y and σ z are the horizontal diffusion parameter and the vertical diffusion parameter respectively, B is the background concentration of CO2, α is the ground reflection coefficient, σ y =a·x b and σ z =c·x d , a, b are the horizontal diffusion coefficients, and c, d are the vertical diffusion coefficients.

[0019] Preferably, solving the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm optimization algorithm and the trust region algorithm further includes:

[0020] Solve the unknown parameters in the established carbon emission diffusion model, and then obtain the carbon emission intensity of the power plant. Among them, the unknown parameters are expressed as H, α, B, z, a, b, c, and d in the emission diffusion model.

[0021] Preferably, the solution of the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm includes:

[0022] According to the monitoring data collected at the downwind of the strong point source carbon emission and the established strong point source carbon emission diffusion model in S2, calculate the carbon emission intensity of the power plant based on the particle swarm algorithm and the trust region algorithm.

[0023] Preferably, the solution of the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm includes:

[0024] Collect the monitoring data at the downwind of the strong point source carbon emission and the established strong point source carbon emission diffusion model, and the solution of the carbon emission intensity of the power plant is implemented through the following steps:

[0025] S3.1. Classify the total data set collected by the UAV sampling system, randomly select 80% of the observation data volume in the total data set as the input set, input it into the strong point source carbon emission diffusion model, and obtain the corresponding parameters of the strong point source carbon emission diffusion model; the observation data volume represents CO2 concentration, measurement point location information, and meteorological data;

[0026] S3.2. Input the CO2 concentration data, meteorological data, and location information in the input set into the strong point source carbon emission diffusion model, based on the set parameter solution boundary, calculate 10,000 times repeatedly, and obtain the value closest to the optimal solution among the corresponding parameters; iterate through the trust region algorithm to solve the extreme point with the minimum objective function as the final optimal solution;

[0027] S3.3. Bring the CO2 concentration data, meteorological data, and location information of the input set into the strong point source carbon emission diffusion model again, use the minimum and maximum values of the corresponding parameters in the 10,000 calculation results in step S3.2 as the potential true value solution interval, use the average value of the corresponding parameters as the initial point, and accurately solve the parameter to be solved according to the optimized trust region algorithm, and its calculation result is used as the final parameter solution value.

[0028] Preferably, the evaluation of the accuracy of the calculated carbon emission result according to the verification data set includes:

[0029] Use the remaining 20% of the observation data in S3.1 as the validation set to evaluate the authenticity of the parameters solved in step S3; according to the parameters calculated in step S3 and the diffusion model, reconstruct the CO2 concentration values at the corresponding positions in the validation set; evaluate the authenticity of the parameters to be solved according to the correlation coefficient between the measured CO2 concentration and the simulated CO2 concentration.

[0030] Preferably, the evaluation of the authenticity of the parameters to be solved according to the correlation coefficient between the measured CO2 concentration and the simulated CO2 concentration is obtained through the following formula:

[0031]

[0032] In the formula, C v is the CO2 concentration value in the validation set, C' v is the CO2 concentration value corresponding to the validation set simulated according to the parameter value calculated in step S3 in the validation set, Cov(C' v , C v ) is the covariance of C' v and C v , Var(C' v ) is the variance of C' v , and Var(C v ) is the variance of C v ;

[0033] When R is greater than 0.9, the parameter solution value in step S3 will be regarded as the final true calculation result, which is the carbon emission intensity of the power plant; when R is less than 0.9, repeat steps S3 to S4 until R is greater than 0.9.

[0034] The present invention also provides an inversion strong point source carbon emission intensity calculation system based on unmanned aerial vehicle sampling, which is used for the above-mentioned inversion strong point source carbon emission intensity calculation method based on unmanned aerial vehicle sampling, including:

[0035] A data acquisition module for acquiring sampling data, where the sampling data includes position information, CO2 concentration, and meteorological data;

[0036] A data processing module for establishing a corresponding strong point source carbon emission diffusion model according to the acquired sampling data;

[0037] A data calculation module for solving the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm;

[0038] A data verification module for evaluating the accuracy of the calculated carbon emission intensity of the power plant according to the verification data set.

[0039] The technical effects and advantages of the present invention are:

[0040] Based on the measured data collected by the UAV sampling device and combined with an adaptive quantitative evaluation model, the present invention realizes an effective and rapid calculation of the carbon emission intensity of strong point sources. At the same time, based on the spatial carbon dioxide concentration data collected by the UAV, the present invention can realize a quantitative evaluation of the carbon emission intensity of strong point sources without prior knowledge constraints, at the 10-minute level, and with high precision. This solution has the characteristics of being not restricted by the terrain of the area to be monitored, having a strong self-correction function, and low cost. In the future, it can provide data support for environmental monitoring and carbon trading.

[0041] Other features and advantages of the present invention will be described in the following specification, and part of them will become obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures pointed out in the specification, claims, and drawings. Brief Description of the Drawings

[0042] Figure 1 is the general flow chart of the method for inverting the carbon emission intensity of strong point sources based on UAV sampling in the embodiment of the present invention;

[0043] Figure 2 is a schematic diagram of the coordinate system establishment of the emission diffusion model in the embodiment of the present invention;

[0044] Figure 3 is the flow chart for evaluating the accuracy of the carbon emission intensity of power plants in the embodiment of the present invention. Detailed Embodiments

[0045] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0046] A method for inverting the carbon emission intensity of strong point sources based on a UAV sampling device provided by an embodiment of the present invention has strong adaptability and high precision in quantitatively calculating the carbon emission intensity of strong point sources; specifically as Figure 1 shown, this method includes the following steps:

[0047] Step S1, collect sampling data, where the sampling data includes position information, CO2 concentration, and meteorological data.

[0048] Furthermore, in step S1, collecting sampling data includes: the UAV designs a reasonable UAV flight sampling route according to the terrain characteristics and meteorological field characteristics of the strong point source to be evaluated, and obtains the concentration diffusion characteristics of the carbon emission of the strong point source.

[0049] It should be noted that since the UAV sampling system has the ability of spatial collection and can freely obtain data samples at different positions. According to the topographic features and meteorological field characteristics of the strong point source to be evaluated, a reasonable UAV flight sampling route is designed to obtain the concentration diffusion characteristics of the carbon emissions from the strong point source.

[0050] Specifically, as shown in the appendix Figure 2 The UAV sampling equipment conducts high-frequency and high-precision data collection in the downwind area of the strong point source. The downwind area of the strong point source can be selected as the downwind opening of the chimney. The UAV sampling spirally ascends from the ground to a height (300 m), showing a spiral ascent on the route. While ascending, the UAV measures the CO2 concentration, position information, and meteorological data at the sampling points. The meteorological data includes wind speed, wind direction, atmospheric temperature, atmospheric humidity, and atmospheric pressure.

[0051] Among them, the CO2 concentration is sampled by the CO2 sampling system installed inside the UAV. The CO2 sampling system mainly includes a sampling tube, an air pump, and a greenhouse gas monitoring instrument. The greenhouse gas monitoring instrument includes, but is not limited to, cavity ring-down spectroscopy (CRDS) and off-axis integrated cavity output spectroscopy (OA-ICOS) greenhouse gas analyzers. The position information is recorded by a GPS instrument; the meteorological data is obtained by an on-vehicle meteorological station, and the meteorological data will participate in the calculation process of converting the CO2 concentration from ppm to g / m 3 of.

[0052] Step S2: Establish a corresponding carbon emission diffusion model for the strong point source according to the collected sampling data;

[0053] In the step S2, according to the carbon emission characteristics of the strong point source and combined with the data characteristics collected by the UAV equipment, the corresponding carbon emission diffusion model is established as follows:

[0054] The CO2 released by the carbon emissions of the strong point source can be considered as stable and continuous emissions. Combining the spatial data collected by the UAV sampling system, the Gaussian diffusion model can be used to mathematically model the carbon emissions of the power plant.

[0055] Step S2.1: Establish a diffusion coordinate system according to the position of the strong emission source and the measured wind direction:

[0056] Take the coordinates of the emission equipment as the origin, the wind direction as the X-axis, the direction perpendicular to the wind direction in the horizontal direction as the Y-axis, and the direction perpendicular to the XOY plane as the Z-axis to establish a coordinate system, as shown in the appendix Figure 2 ;

[0057] Step S2.2: Establish a carbon emission diffusion model for the strong point source according to the coordinate system established in step S2.1: The carbon emission diffusion model for the strong point source is expressed as follows,

[0058]

[0059] Among them, (x, y, z) are the coordinates of the measurement points of the UAV sampling device, C(x, y, z) is the CO2 concentration at the coordinates (x, y, z), q is the emission intensity, u is the wind speed, H is the effective emission height of the power plant carbon emission, and σ y and σ z are the horizontal diffusion parameter and the vertical diffusion parameter respectively, B is the background concentration of CO2, α is the ground reflection coefficient, and σ y = a·x b and σ z = c·x d where a, b are the horizontal diffusion coefficients, and c, d are the vertical diffusion coefficients.

[0060] Step S3. Solve the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm.

[0061] The calculation of the strong point source carbon emission intensity based on the particle swarm algorithm and the optimized trust region algorithm includes the following steps:

[0062] The UAV sampling system obtains a large number of monitoring data at the downwind of the strong point source carbon emission. Based on the collected data and the Gaussian diffusion model established in step S2, the unknown parameters (H, α, B, z, a, b, c, d) in the strong point source carbon emission diffusion model can be solved based on the particle swarm algorithm and the trust region algorithm. For details, refer to Figure 3 as shown, and the specific steps are as follows:

[0063] Step S3.1 Classify the total data set collected by the UAV sampling system, and randomly select 80% of the observed data volume (CO2 concentration, measurement point location information, and meteorological data) as the input set to realize the solution of the parameters (q, H, α, B, z, a, b, c, d).

[0064] Step S3.2 Perform preliminary calculations on the parameters to be solved in step S3.1 according to the particle swarm algorithm.

[0065] In this embodiment, first, set the boundaries of the solution parameters: among them, the upper and lower boundaries of the wind speed and wind direction are set according to the measurement accuracy of the actual meteorological instrument; the upper and lower boundaries of the effective emission height of the emission source are set to 0 - 100m; the setting boundary of the ground reflection coefficient α is 0 - 1; the lower limits of the remaining parameters to be solved are all 0, and no upper limit is set. The set parameter boundaries are used as the logical domain of the particle swarm algorithm and are solved according to formula 2.

[0066] v i = v i + c1×rand×randc2×(pbest i - x i) + c2 × rand × (pbest i - x i ) (2)

[0067] x i = x i + v (3)

[0068] where i = 1, 2, 3... N, N is the total number of particles in the particle swarm, v i is the velocity of the particle, rand is a random number between (0, 1), and x i represents the position of the particle, here representing the positions of various parameters within the set logical domain; c1 and c2 represent learning factors. Set the objective function Loss.

[0069]

[0070] C X (x j , y j , z j ), j = 1, 2, 3... m, are the simulated values of CO2 concentrations at different positions in the input set using iterative parameters, and m is the number of CO2 collection points in the input set.

[0071] The particle swarm algorithm finds the optimal solution that minimizes the loss function Loss through optimal iteration This process is represented by the following formula.

[0072]

[0073] Bring the CO2 concentration data, position information, and meteorological data of the input set into the carbon emission diffusion model, solve the boundary based on the set parameters, and repeat the calculation 10,000 times.

[0074] Due to the optimal solution obtained by the particle swarm algorithm being only the value closest to the optimal solution in the global feasible region, it is necessary to iterate through the trust region method to solve for the extreme point that minimizes the objective function f(X) as the final optimal solution

[0075] Step S3.3 performs precise calculation on the parameters to be solved based on the optimized trust region; again, bring the CO2 concentration data, meteorological data, and position information of the input set into the carbon emission diffusion model, use the minimum and maximum values of the corresponding parameters in the 10,000 calculation results in step S3.2 as the potential true value solution interval, use the average value of the corresponding parameters as the initial point, and precisely solve the parameters to be solved according to the optimized trust region algorithm, and its calculation result is used as the final parameter solution value. Specifically as follows:

[0076] Solve through trust region iteration First, it is necessary to define f(X) at the k-th iteration, and f(X) is represented by the following formula 6.

[0077]

[0078]

[0079] Solve for ΔX through formulas 7 and 8 k To further minimize the objective function f(X), obtain ΔX k+1 = X k + ΔX k .

[0080]

[0081]

[0082] where τ k can be represented by formula 9, can be represented by formula 10, can be represented by formula 11, Δ k is defined as the trust region radius ||ΔX k || < Δ k .

[0083]

[0084]

[0085] where, X1 = q, X2 = a, X3 = b, X4 = c, X5 = d, X6 = B, X7 = α, X8 = z. When ||f(X k+1 ) - f(X k )|| < ε, where ε is the preset iteration stop condition. When the stop condition is satisfied, where the first element of is the CO2 emission to be solved.

[0086] Step S4: Evaluate the accuracy of the calculated carbon emission results according to the validation dataset.

[0087] The evaluation of the authenticity of the calculated carbon emission results according to the validation dataset is as follows: Use the remaining 20% of the observed data in S3.1 as the validation set to evaluate the authenticity of the parameters solved in step S3. According to the solved values of the unknown parameters in the diffusion model calculated in step S3 and the diffusion model, reconstruct the CO2 concentration values at the corresponding positions in the validation set. Evaluate the authenticity of the parameters to be solved based on the correlation coefficient between the measured CO2 concentration and the simulated CO2 concentration.

[0088]

[0089] C′ v (x i , y i , z i ), i = 1, 2, 3…n, is the simulated CO2 concentration value at the corresponding position in the validation set according to the diffusion model and the solved parameter values. q, H, α, B, z, a, b, c, and d are the parameter values calculated in step S3. The correlation coefficient is calculated as follows:

[0090]

[0091] C v is the CO2 concentration value in the validation set, and C′ v is the CO2 concentration value corresponding to the validation set simulated according to the parameter values calculated in step S3. Cov(C′ v , C v ) is the covariance of C′ v and C v , Var(C′ v ) is the variance of C′ v , and Var(C v ) is the variance of C v .

[0092] When R is greater than 0.9, the parameter solution value in step S3 will be regarded as the final true calculation result, and the strong point source carbon emission intensity and other parameters can be obtained. When R is less than 0.9, repeat steps S3 to S4 until R is greater than 0.9.

[0093] The embodiment of the present invention also provides an inversion strong point source carbon emission intensity calculation system based on UAV sampling, which is used to implement the above-mentioned inversion strong point source carbon emission intensity calculation method based on UAV sampling, and is characterized in that it includes

[0094] A data acquisition module, which is used to acquire sampling data, and the sampling data includes position information, CO2 concentration, and meteorological data;

[0095] A data processing module, which is used to establish a corresponding strong point source carbon emission diffusion model according to the acquired sampling data;

[0096] A data calculation module, which is used to solve the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm;

[0097] A data verification module, which is used to evaluate the accuracy of the calculated carbon emission intensity of the power plant according to the verification data set.

[0098] It can be understood that the inversion strong point source carbon emission intensity calculation system based on UAV sampling provided by the present invention corresponds to the inversion strong point source carbon emission intensity calculation method based on UAV sampling provided by the foregoing embodiments. The relevant technical features of the inversion strong point source carbon emission intensity calculation system based on UAV sampling can refer to the relevant technical features of the inversion strong point source carbon emission intensity calculation system based on UAV sampling, which will not be elaborated here.

[0099] In summary, the method and system for inverting the carbon emission intensity of strong point sources based on UAV sampling equipment provided by the present invention can quickly and accurately obtain the carbon emission solutions of strong point sources based on the UAV sampling system and the use of a stable carbon emission accounting method. The emission inversion framework adopted by this technical solution can greatly reduce the requirements for prior knowledge and fully realize adaptive global optimal adjustment to ensure the minimum influence of the concentration sampling accuracy and meteorological acquisition accuracy in the actual application process.

[0100] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0101] The protection scope of the present invention is not limited to the above examples. Obviously, those skilled in the art can make various changes and deformations to the present invention without departing from the scope and spirit of the present invention. If these changes and deformations fall within the scope of the claims of the present invention and their equivalent technologies, the intention of the present invention also includes these changes and deformations.

Claims

1. A calculation method for inverting the carbon emission intensity of strong point sources based on UAV sampling, the method comprising the following steps: Step S1, collect sampling data, the sampling data including location information, CO2 concentration and meteorological data; Step S2, establish a corresponding carbon emission diffusion model of strong point sources according to the collected sampling data; Step S3, solve the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm; including: collect the monitoring data at the downwind of the strong point source carbon emission and the established carbon emission diffusion model of strong point sources, and solve the carbon emission intensity of the power plant through the following steps: S3.1, classify the total data set collected by the UAV sampling system, randomly select 80% of the observation data volume in the total data set as the input set, input it into the carbon emission diffusion model of strong point sources, and obtain the corresponding parameters of the carbon emission diffusion model of strong point sources; S3.2, input the CO2 concentration data, meteorological data and location information in the input set into the carbon emission diffusion model of strong point sources, based on the set parameter solution boundaries, calculate 10,000 times repeatedly, and obtain the value closest to the optimal solution among the corresponding parameters; iterate through the trust region algorithm to solve the extreme point with the minimum objective function as the final optimal solution; S3.3, bring the CO2 concentration data, meteorological data and location information of the input set into the carbon emission diffusion model of strong point sources again, take the minimum value and the maximum value of the corresponding parameters in the 10,000 calculation results in step S3.2 as the solution interval of the potential true value, take the average value of the corresponding parameters as the initial point, accurately solve the parameter to be solved according to the optimized trust region algorithm, and take the calculation result as the final parameter solution value; Step S4, evaluate the accuracy of the carbon emission intensity of the power plant according to the verification data set.

2. The calculation method for inverting the carbon emission intensity of strong point sources based on UAV sampling according to claim 1, characterized in that: Collecting sampling data according to the UAV sampling equipment includes: designing a reasonable UAV flight sampling route according to the terrain characteristics and meteorological field characteristics of the strong point source to be evaluated, and obtaining the concentration diffusion characteristics of the strong point source carbon emission.

3. The calculation method for inverting the carbon emission intensity of strong point sources based on UAV sampling according to claim 1, characterized in that: The establishment of the corresponding carbon emission diffusion model of strong point sources includes: according to the carbon emission characteristics of strong point sources, combined with the spatial data collected by the UAV sampling equipment, select the Gaussian diffusion model to conduct mathematical modeling on the carbon emission of the power plant.

4. A method for calculating the carbon emission intensity of an inversion strong point source based on UAV sampling according to claim 3, characterized in that: The establishment of the corresponding carbon emission diffusion model of strong point sources includes: S2.1, establish a diffusion coordinate system according to the location of the strong emission source and the measured wind direction; S2.2, establish a carbon emission diffusion model of strong point sources according to the diffusion coordinate system established in step S2.

1.

5. A method for calculating the carbon emission intensity of an inversion strong point source based on UAV sampling according to claim 1, characterized in that: The carbon emission diffusion model of strong point sources is expressed as follows, Among them, ( x , y , z ) are the coordinates of the measurement points of the UAV sampling device, is the CO2 concentration at the coordinates of ( x , y , z ), q is the carbon emission intensity of the power plant, u is the wind speed, H is the effective emission height of the carbon emission of the power plant, and are the horizontal diffusion parameter and the vertical diffusion parameter respectively, B is the background concentration of CO2, α is the ground reflection coefficient, , , a , b is the horizontal diffusion coefficient, c , d is the vertical diffusion coefficient.

6. The calculation method for inverting the carbon emission intensity of strong point sources based on UAV sampling according to claim 5, characterized in that: The solution of the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm further includes: Solve the unknown parameters in the established carbon emission diffusion model, and then obtain the carbon emission intensity of the power plant. Among them, the unknown parameters are expressed as those in the emission diffusion model H, α, B, z, a, b, c, d .

7. A method for calculating the carbon emission intensity of an inversion strong point source based on drone sampling according to claim 1, characterized in that: The evaluation of the accuracy of the calculated carbon emission results according to the verification data set includes: Use the remaining 20% of the observed data in S3.1 as the validation set to evaluate the authenticity of the parameters solved in step S3; according to the parameters calculated in step S3 and the diffusion model, reconstruct the CO2 concentration values at the corresponding positions in the validation set; evaluate the authenticity of the parameters to be solved according to the correlation coefficient between the measured CO2 concentration and the simulated CO2 concentration.

8. A method for calculating the carbon emission intensity of an inversion strong point source based on UAV sampling according to claim 7, characterized in that: The evaluation of the authenticity of the parameters to be solved according to the correlation coefficient between the measured CO2 concentration and the simulated CO2 concentration is obtained through the following formula: Wherein, is the CO2 concentration value in the validation set, is the CO2 concentration value corresponding to the validation set simulated by the parameter value calculated according to step S3 in the validation set, Cov( , ) is and covariance, Var( ) is variance, Var( ) is variance; When R is greater than 0.9, the parameter solution value in step S3 will be regarded as the final true calculation result, which is the carbon emission intensity of the power plant; when R is less than 0.9, repeat steps S3 to S4 until R is greater than 0.

9.

9. An inversion strong point source carbon emission intensity calculation system based on UAV sampling, which is used to implement an inversion strong point source carbon emission intensity calculation method according to any one of claims 1 to 8, and is characterized in that: including a data acquisition module for acquiring sampling data, where the sampling data includes position information, CO2 concentration, and meteorological data; a data processing module for establishing a corresponding strong point source carbon emission diffusion model according to the acquired sampling data; a data calculation module for solving the carbon emission intensity of the power plant in the established carbon emission diffusion model based on the particle swarm algorithm and the trust region algorithm; including: acquiring the monitoring data at the downwind of the strong point source carbon emission and the established strong point source carbon emission diffusion model, and the solution of the carbon emission intensity of the power plant is implemented through the following steps: S3.

1. Classify the total data set collected by the UAV sampling system, randomly select 80% of the observed data volume in the total data set as the input set, input it into the strong point source carbon emission diffusion model, and obtain the corresponding parameters of the strong point source carbon emission diffusion model; S3.

2. Input the CO2 concentration data, meteorological data, and position information in the input set into the strong point source carbon emission diffusion model, and repeat the calculation 10,000 times based on the set parameter solution boundary to obtain the value closest to the optimal solution among the corresponding parameters; iterate through the trust region algorithm to solve the extreme point with the minimum objective function as the final optimal solution; S3.

3. Again, input the CO2 concentration data, meteorological data, and position information of the input set into the strong point source carbon emission diffusion model. Take the minimum and maximum values of the corresponding parameters in the 10,000 calculation results in step S3.2 as the potential true value solution interval, take the average value of the corresponding parameters as the initial point, and accurately solve the parameter to be solved according to the optimized trust region algorithm, and take the calculation result as the final parameter solution value; a data verification module for evaluating the accuracy of calculating the carbon emission intensity of the power plant according to the verification data set.

Images