Construction site carbon emission source inversion calculation method based on atmospheric diffusion model

Abstract

The invention relates to the technical field of carbon emission source inversion calculation, in particular to a construction site carbon emission source inversion calculation method based on an atmospheric diffusion model. According to the technical scheme, the method comprises the following steps that a spatial diffusion model is constructed, meteorological parameters of a construction site are obtained through meteorological monitoring equipment, pollutant concentration distribution is calculated based on an atmospheric diffusion model, and the model corrects vertical diffusion by introducing a mirror image source item; performing dynamic noise reduction on the data: performing filtering processing on the original monitoring data, and balancing noise suppression and real fluctuation retention through a dynamic adjustment algorithm; performing iterative inversion calculation; and comprehensive verification: combining theoretical simulation with actual detection. Through the dynamic coupling meteorological-emission model, the adaptive noise reduction algorithm and the multi-source verification mechanism, accurate, efficient and low-cost inversion of the carbon emission source of the construction site is realized, and reliable technical support is provided for real-time carbon management, carbon transaction and green construction.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission source inversion calculation, and particularly relates to a method for inverting and calculating carbon emission sources at construction sites based on an atmospheric diffusion model. Background Art

[0002] With the increasingly strict global control of carbon emissions, the monitoring and tracing of carbon emissions at construction sites have become an important topic in environmental management. Currently, the inversion of carbon emission sources mainly relies on the following methods:

[0003] Mass balance method: Estimate the carbon emission intensity based on a static emission factor library, but it ignores the impact of the dynamic working conditions of construction machinery (such as start-stop and load changes) on the emission factors, and cannot separate the multi-source superposition effect, resulting in significant errors in complex scenarios.

[0004] Inverse Lagrangian model: Invert the emission source through particle tracking, but it is necessary to simulate a super-large number of particles (such as the 10 6 th level) to ensure accuracy, resulting in an excessively long calculation time (>8 hours), and being highly sensitive to the accuracy of the meteorological field (for example, a wind speed error of 1 m / s can cause a positioning deviation of >200 m), making it difficult to meet the real-time requirements.

[0005] Bayesian inversion method: Optimize the emission source parameters relying on the prior distribution, but the prior distribution is highly subjective, especially in scenarios with diverse construction machinery types and complex emission characteristics, which is prone to introducing systematic biases.

[0006] Satellite remote sensing inversion: Has a low spatial resolution (usually >1 km), and is easily affected by cloud cover. The data missing rate is as high as 49% on cloudy or rainy days, and it cannot meet the high-precision monitoring requirements of construction sites.

[0007] Internet of Things sensing network: Requires a high-density layout of sensors (such as 100 nodes). In the long-term operation, the drift errors of the sensors accumulate, and the hardware cost is high (>2 million yuan), making it difficult to be applied on a large scale.

[0008] In addition, the carbon emissions at construction sites have the characteristics of multi-source dynamics (such as intermittent emissions from machinery such as excavators and cranes), significant local meteorological disturbances (such as strong winds and thermal buoyancy lifting effects), and complex terrain occlusion. Traditional methods have obvious limitations in terms of real-time performance, positioning accuracy, and multi-source analysis capabilities.

[0009] Therefore, this application proposes a method for inverting and calculating carbon emission sources at construction sites based on an atmospheric diffusion model. Summary of the Invention

[0010] The object of the present invention is to address the problem that traditional carbon emission source inversion methods in the background technology have obvious limitations in terms of real-time performance, positioning accuracy, and multi-source analysis ability, and to propose a carbon emission source inversion calculation method for construction sites based on an atmospheric diffusion model.

[0011] The technical solution of the present invention: A carbon emission source inversion calculation method for construction sites based on an atmospheric diffusion model, comprising the following steps:

[0012] S1. Construct a spatial diffusion model: Obtain the meteorological parameters of the construction site through meteorological monitoring equipment, and calculate the pollutant concentration distribution based on the atmospheric diffusion model. The model corrects the vertical diffusion by introducing an image source term.

[0013] S2. Dynamic data noise reduction: Filter the original monitoring data, and balance noise suppression and retention of real fluctuations through a dynamic adjustment algorithm.

[0014] S3. Iterative inversion calculation:

[0015] Initialize the emission intensity of potential emission sources;

[0016] Simulate the concentration based on the diffusion model and compare it with the measured value, and dynamically correct the emission intensity;

[0017] Terminate the calculation when the preset convergence condition is met;

[0018] S4. Comprehensive verification: Verify the reliability of the inversion result by combining theoretical simulation and actual detection.

[0019] Optionally, the atmospheric diffusion model is an improved Gaussian diffusion model, specifically including:

[0020] Obtain the near-surface wind speed U, wind direction θ, and atmospheric boundary layer height h in real time;

[0021] Correct the vertical diffusion term through five groups of image source terms and the ground reflection attenuation factor β;

[0022] The pollutant concentration calculation formula of the improved Gaussian diffusion model is:

[0023]

[0024]

[0025] β = 0.88exp(-0.05x)

[0026] Δh = 0.015N 0.7

[0027] Where, C(x, y, z, z s ) is the pollutant concentration at the downwind point (x, y, z) of the pollution source; Cb is the background concentration; U is the near-surface wind speed; h is the height of the atmospheric boundary layer, determined according to the empirical formula; Q s is the emission intensity of the pollution source; β is the ground reflection attenuation factor; σ y , σ z are the horizontal diffusion coefficient and the vertical diffusion coefficient respectively, whose values are related to the horizontal distance x and are calculated using the Briggs formula; z s is the height of the plume axis from the ground; H is the physical height of the emission source; Δh is the thermal buoyancy lift height; W0 is the emission outlet velocity; N is the diesel engine speed.

[0028] Optionally, the dynamic adjustment algorithm is an adaptive Kalman filter, which realizes dynamic noise reduction through the time-varying gain K t and the historical memory factor α. The update formula of the adaptive Kalman filter is:

[0029]

[0030] where α is the historical memory factor, which has been verified by measured data to achieve an optimal balance between retaining the true fluctuations and suppressing random noise; K t is the time-varying Kalman gain, dynamically adjusting the confidence of the observed value; P t|t-1 is the state prediction covariance; R t is the dynamic observation noise variance, estimated in real time according to the sensor signal-to-noise ratio; H is taken as 1 to simplify the observation model.

[0031] Optionally, in step S3:

[0032] The initial emission intensity combines the OBD data of construction machinery to construct a prior distribution;

[0033] The dynamic correction adjusts the emission intensity through the inversion coefficient R n ,

[0034] The preset convergence condition is that the consistency index IOA≥0.8 and the emission intensity volatility of 3 consecutive iterations <5%;

[0035] The preset convergence condition is the consistency index:

[0036] The calculation formula of the inversion coefficient R n is:

[0037]

[0038] The dynamically corrected emission intensity is:

[0039] where M i and O iData obtained from model simulations and observations respectively; n is the number of possible construction activities that may emit, determined according to the construction layout and wind direction; N is the number of effective monitoring points; and are the mean values of the simulation values and the observed values respectively; and are the simulation values and the observed values corresponding to the monitoring points of enterprise n respectively.

[0040] Optionally, in step S4:

[0041] The theoretical simulation includes quantifying the overall uncertainty U of sensor error, model bias, and meteorological fluctuations using the Monte Carlo method total , and requires U total <25%;

[0042] The actual detection uses the 13CO2 isotope tracing technology to verify the model accuracy by detecting that the recovery rate η ≥ 85%;

[0043] The overall uncertainty U total The calculation formula is:

[0044]

[0045] where U sensor refers to the uncertainty of sensor measurement, that is, the accuracy range calibrated when the sensor leaves the factory; U model refers to the uncertainty of model bias; U meteo refers to the uncertainty of meteorological data, that is, the error introduced by inaccurate observation or prediction of the meteorological parameters input into the model.

[0046] Optionally, the method further includes docking with the BIM system to generate a carbon emission heat map and realizing the non-tamperable storage of data through blockchain technology.

[0047] Optionally, in the method, the meteorological monitoring device includes a Vaisala WXT536 weather station, the CO2 sensor uses a SenseAir K30, and the edge computing node is an NVIDIA Jetson AGX Xavier.

[0048] Optionally, in the isotope tracing verification, the concentration of the released 13CO2 is 0.1%, and the recovery rate is detected by a mass spectrometer.

[0049] Compared with the prior art, the present application includes at least one of the following beneficial technical effects:

[0050] By integrating mechanical condition data with an improved atmospheric diffusion model, the accuracy of emission source localization is significantly improved, and multi-source emissions in complex scenarios can be effectively distinguished.

[0051] Enhanced computational efficiency: By adopting an edge computing architecture optimization algorithm, the inversion time is significantly shortened to meet the requirements of high-frequency monitoring.

[0052] Strengthened anti-interference ability: Combined with intelligent filtering technology, environmental noise and sensor errors are effectively suppressed to ensure data reliability.

[0053] Expanded scenario adaptability: Supports monitoring under complex terrains and adverse weather conditions, enabling all-weather and multi-scenario applications.

[0054] Guaranteed data credibility: The model design based on physical mechanisms ensures the interpretability of results. Combined with blockchain technology, data traceability and anti-tampering are achieved to meet regulatory requirements.

[0055] Through the dynamic coupling of meteorological-emission models, adaptive noise reduction algorithms, and multi-source verification mechanisms, the present invention realizes the accurate, efficient, and low-cost inversion of carbon emission sources at construction sites, providing reliable technical support for real-time carbon management, carbon trading, and green construction. Description of the Drawings

[0056] Figure 1 It is a flowchart of the iterative calculation method in the embodiment. Detailed Embodiments

[0057] The technical solutions of the present invention will be further described below in conjunction with the drawings and specific embodiments.

[0058] The present invention proposes an inversion calculation method for carbon emission sources at construction sites based on an atmospheric diffusion model. By constructing a dynamically coupled meteorological-emission model system, accurate tracing of multi-source carbon emissions at the construction site is realized. This method uses the Gaussian diffusion model as the core calculation framework, introduces the environmental assessment module in the ADMS system of the UK Natural Environment Research Council for model optimization, and effectively solves the technical bottlenecks of fuzzy emission source positioning and difficult separation of multi-source contributions in traditional methods through the synergistic effect of real-time monitoring data and iterative inversion algorithms.

[0059] (1) Spatial Diffusion Model

[0060] The technical solution first establishes a three-dimensional spatial diffusion model, and key parameters such as the near-surface wind speed U, wind direction θ, and atmospheric boundary layer height h are obtained in real time through meteorological monitoring equipment deployed around the construction site. The pollutant transport process is described by an improved Gaussian model, and the multiple reflection effects of ground reflection and the top of the boundary layer are particularly considered in the model. By introducing five groups of mirror source terms, such as Equation (1), the vertical diffusion term is finely characterized.

[0061]

[0062] β = 0.88exp(-0.05x) (4)

[0063] Δh = 0.015N 0.7 (5)

[0064] where C(x, y, z, z s ) is the pollutant concentration at the downwind point (x, y, z) of the pollution source, with the unit of g·m -3 ; C b is the background concentration, with the unit of g·m -3 ; U is the near-surface wind speed, with the unit of m·s -1 ; h is the atmospheric boundary layer height, with the unit of m, which is determined according to the empirical formula; Q s is the emission intensity of the pollution source, with the unit of g·s -1 ; β is the ground reflection attenuation factor, dimensionless; σ y , σ z are the horizontal diffusion coefficient and the vertical diffusion coefficient respectively, with the unit of m, and their values are related to the horizontal distance x, and are calculated using the Briggs formula; z s is the height of the plume axis from the ground, with the unit of m; H is the physical height of the emission source, with the unit of m; Δh is the thermal buoyancy lift height, with the unit of m; W0 is the emission outlet velocity, with the unit of m / s; N is the diesel engine speed, with the unit of rpm.

[0065] According to the meteorological coupling mechanism, the Monin-Obukhov similarity theory is introduced to correct the diffusion parameters:

[0066]

[0067] where L is the Obukhov length, reflecting the atmospheric stability.

[0068] (2) Data preprocessing

[0069] First, deploy the model preprocessing unit based on the fog computing architecture, and develop an adaptive Kalman filter to denoise the original monitoring data.

[0070]

[0071] where α is the historical memory factor, which is verified by the measured data, and this value reaches the optimal balance between retaining the real fluctuations (such as the concentration step caused by the mechanical start and stop) and suppressing the random noise; K t is the time-varying Kalman gain, dynamically adjusting the trust degree of the observed value; P t|t-1 is the state prediction covariance (initialized by the construction machinery vibration spectrum analysis); R t is the dynamic observation noise variance, which is estimated in real time according to the sensor signal-to-noise ratio; H is taken as 1 to simplify the observation model (direct concentration measurement).

[0072] (3) Inversion calculation

[0073] Based on model construction and data processing, a multi-stage iterative inversion algorithm system was developed, as shown in Figure 1 .

[0074] In the initial stage, theoretical emission intensities were assigned to each potential emission source according to the distribution of construction machinery Innovatively combined with on-board diagnostic (OBD) data of construction machinery (such as diesel engine intake air flow and hydraulic pressure) to construct a prior distribution of emission intensity Among them, μ is dynamically assigned with reference to the "Emission Factor Handbook for Non-road Mobile Machinery". At the same time, real-time wind direction data is combined to screen effective influencing sources (emission sources upwind of the monitoring point).

[0075] A three-dimensional space diffusion model was used to simulate the concentration of emission sources using the initial intensity. By comparing the simulated concentration of each emission source with the measured value at the monitoring point the inversion coefficient R was calculated n , and the emission intensity was dynamically corrected accordingly When the index of agreement (IOA) ≥ 0.8 and the volatility of three consecutive iterations < 5%, the calculation is terminated.

[0076]

[0077] Among them, M i and O i are the data obtained from model simulation and observation respectively; n is the number of possible construction activities with emissions determined according to the construction layout and wind direction; N is the number of effective monitoring points; and are the mean values of the simulated values and the observed values respectively; and are the simulated values and the observed values corresponding to the nth enterprise at the monitoring point respectively.

[0078] (4) Verification system

[0079] To verify the reliability of the model, the system integrates a dual verification mechanism.

[0080] At the theoretical level, Monte Carlo simulation was used to quantify the overall uncertainty brought by sensor errors, model biases, and meteorological fluctuations, ensuring an accuracy requirement of U total < 25%.

[0081]

[0082] Among them, U sensor refers to the uncertainty of sensor measurement, that is, the accuracy range calibrated when the sensor leaves the factory; U modelRefers to the uncertainty of the model deviation, that is, the output deviation caused by theoretical simplification or parameter errors in the atmospheric diffusion model and optimization algorithm used in the inversion calculation. The model deviation in complex terrain is generally 15.3%; U meteo Refers to the uncertainty of meteorological data, that is, the error introduced by inaccurate observation or prediction of meteorological parameters (wind speed U, wind direction θ, atmospheric stability, etc.) input into the model.

[0083] At the engineering application level, the CO2 isotope tracer technology is innovatively introduced 13 to directly verify the inversion accuracy of the model by releasing a tracer gas with a known intensity and detecting its spatial distribution. The model accuracy is verified by the recovery rate detected by mass spectrometry. When η≥85%, the model is determined to be reliable.

[0084]

[0085] Among them, is the concentration of the released carbon dioxide tracer, which is 0.1%; is the concentration of the detected carbon dioxide tracer; Q model is the emission intensity of the simulated emission source, g·s -1 ; Q actual is the actual emission intensity of the emission source, g·s -1 .

[0086] The technical extensibility of the present invention is reflected in the adaptability to multi-dimensional application scenarios. By docking with the BIM system, a three-dimensional construction progress model integrating a carbon emission heat map can be generated, providing data support for the optimization of green construction plans. In the field of carbon trading, the spatio-temporal scheduling emission data output by the system can be directly docked with the MRV verification system, and its blockchain evidence storage module realizes the tamper-proof storage of carbon data through the Hyperledger architecture. Tests show that this method can reduce the carbon emission accounting cost of large construction sites to less than 40% of traditional manual audits, and at the same time support high-frequency dynamic monitoring updated every 15 minutes, providing technical support for real-time carbon management decisions at the construction site.

[0087] Effect verification

[0088] (1) The atmospheric diffusion model adopts the mirror source phase correction technology, adding a ground reflection attenuation factor β in Equation (1) to effectively reduce the overestimation of concentration caused by multiple reflections on hard ground surfaces.

[0089] (2) In the calculation of the atmospheric diffusion model, the thermal buoyancy lift height Δh is innovatively linked with the construction machinery working condition database, and the mapping relationship between the diesel engine speed N (rpm) and Δh is established.

[0090] (3) The multi-dimensional performance improvement is shown in the table:

[0091]

[0092]

[0093] (4) The application effects are as follows:

[0094] The multi-source parsing ability is greatly improved, and independent emission sources with a minimum contribution rate of 5% can be identified;

[0095] The calculation efficiency is greatly improved, and the inversion speed is 15 times higher than that of the traditional CFD method;

[0096] The positioning accuracy is improved, and the error is <50m under strong wind conditions (level 6 wind);

[0097] The average convergence speed is fast, and the average number of iterations is less than 30;

[0098] For the inversion method based on point monitoring data, the spatial resolution has been effectively improved, and the data efficiency in cloudy weather has increased from 51% to 89% compared with the satellite remote sensing inversion method.

[0099] 1. Hardware deployment

[0100] Arrange a monitoring network at the construction site:

[0101] Weather station: Install 1 unit (model: Vaisala WXT536) every 200m along the site boundary, with an installation height of 10m, and collect real-time wind speed U (0 - 60m / s, accuracy ±0.3m / s), wind direction θ (0 - 360°, accuracy ±3°), and temperature (-40 to 70°C) data, with a sampling frequency of 1Hz.

[0102] CO2 sensor: Arrange at intervals of 50m downstream of the prevailing wind direction (model: SenseAir K30), with a height of 1.5m from the ground, a measurement range of 0 - 5000ppm, and an accuracy of ±(50ppm + 3% reading).

[0103] Vibration monitor: Install a three-axis accelerometer (model: PCB356A15) at the base of large machinery (excavator, crane) to monitor vibration signals in the range of 4 - 2000Hz.

[0104] Edge computing node: Every 500m 2 Deploy 1 unit (NVIDIA Jetson AGX Xavier), with a built-in 4G communication module.

[0105] 2. Data preprocessing and noise reduction

[0106] (1) Vibration noise suppression

[0107] The accelerometer collects vibration signals x(t) in real time and performs a 1024-point FFT transform to obtain the spectrum X(f);

[0108] Identify the mechanical characteristic frequency (such as the impact frequency of the excavator \(f_c = 35\pm5\) Hz), and generate a band-stop filter:

[0109]

[0110] Filter the original CO2 data \(C\) raw (t).

[0111] (2) Kalman dynamic noise reduction

[0112] Initialize the parameters: \(P0 = 10R0\) (\(R0 = 50\) ppm 2 ), \(\alpha = 0.7\), and the sliding window \(N = 10\).

[0113] Iterative calculation: Take the dynamic estimation \(R\) of the variance of the nearest 10 samples t .

[0114] (3) Gaussian diffusion model construction

[0115] Calculate the diffusion coefficient \(\sigma\) according to the Briggs formula y , \(\sigma\) z ;

[0116] Introduce turbulence compensation and calculate

[0117] The plume height correction \(z\) s ;

[0118] Implement the mirror source correction \(\beta\) and add an attenuation factor to each mirror source term;

[0119] (4) Inversion calculation

[0120] Select an initial calculation time, and obtain the hourly regional wind speed, wind direction, and pollutant concentration through real-time monitoring equipment. Determine the upwind emission sources through the hourly wind direction data, and first assign an initial emission intensity to all upwind emission sources (such as construction activity \(n\)) Use Equation (1) to superimpose and calculate the influence of the upwind emission sources to obtain the simulated value at the downwind monitoring point of construction activity \(n\) Compare the simulated value with the observed value to obtain the inversion coefficient \(R\) n , and use the inversion coefficient to correct the initial emission intensity of the source. Input the corrected source emission intensity again for calculation. Using the index of agreement (IOA) as the judgment criterion, if \(IOA\geq0.8\) and the volatility of 3 consecutive iterations is \(<5\%\), terminate the loop to obtain the emission intensity of construction activity \(n\).

[0121] (5) Isotope tracing verification

[0122] Release at the suspected source point 13 CO2 (concentration 0.1%, flow rate 0.5 L / min), for 30 min continuously.

[0123] Collect gas samples at the downwind sampling point every 5 min, and analyze the δ 13 C value using an isotope mass spectrometer (Picarro G2301).

[0124] Calculate the recovery rate η. If it is greater than or equal to 85, it passes the verification, and determine the emission intensity of the final emission source.

[0125] The above specific embodiments are only several alternative embodiments of the present invention. Based on the technical solution of the present invention and the relevant revelations of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A method for inverse calculation of carbon emission sources at construction sites based on an atmospheric diffusion model, characterized in that, It includes the following steps: S1. Construct a spatial diffusion model: Obtain the meteorological parameters of the construction site through meteorological monitoring equipment, calculate the pollutant concentration distribution based on the atmospheric diffusion model, and the model corrects the vertical diffusion by introducing an image source term; S2. Data dynamic noise reduction: Filter the original monitoring data, and balance noise suppression and retention of real fluctuations through a dynamic adjustment algorithm; S3. Iterative inversion calculation: Initialize the emission intensity of potential emission sources; Simulate the concentration based on the diffusion model and compare it with the measured value, and dynamically correct the emission intensity; Terminate the calculation when the preset convergence condition is met; S4. Comprehensive verification: Verify the reliability of the inversion result by combining theoretical simulation and actual detection.

2. The construction site carbon emission source inversion calculation method based on the atmospheric diffusion model according to claim 1, characterized in that The atmospheric diffusion model is an improved Gaussian diffusion model, specifically including: Obtain the near-surface wind speed U, wind direction θ and atmospheric boundary layer height h in real time; Correct the vertical diffusion term through five groups of image source terms and the ground reflection attenuation factor β; The pollutant concentration calculation formula of the improved Gaussian diffusion model is: β = 0.88exp(-0.05x) Δh = 0.015N 0.7 Among them, C(x, y, z, z s ) is the pollutant concentration at the downwind point (x, y, z) of the pollution source; C b is the background concentration; U is the near-surface wind speed; h is the height of the atmospheric boundary layer, which is determined according to the empirical formula; Q s is the emission intensity of the pollution source; β is the ground reflection attenuation factor; σ y , σ z are the horizontal diffusion coefficient and the vertical diffusion coefficient respectively, and their values are related to the horizontal distance x and are calculated using the Briggs formula; z s is the height of the plume axis from the ground; H is the physical height of the emission source; Δh is the thermal buoyancy lift height; W0 is the emission outlet velocity; N is the diesel engine speed.

3. A method for inverse calculation of carbon emission sources at construction sites based on an atmospheric diffusion model according to claim 1, characterized in that, The dynamic adjustment algorithm is an adaptive Kalman filter, which realizes dynamic noise reduction through a time-varying gain K t and a historical memory factor α. The update formula of the adaptive Kalman filter is as follows: Among them, α is the historical memory factor. Verified by actual measurement data, this value reaches an optimal balance between retaining real fluctuations and suppressing random noise; K t is the time-varying Kalman gain, dynamically adjusting the confidence in the observed value; P t|t-1 is the state prediction covariance; R t is the dynamic observation noise variance, estimated in real time according to the signal-to-noise ratio of the sensor; H is taken as 1 to simplify the observation model.

4. A method for inversely calculating carbon emission sources at a construction site based on an atmospheric diffusion model according to claim 1, characterized in that, In step S3: The initialized emission intensity constructs a prior distribution by combining the OBD data of construction machinery; The dynamic correction adjusts the emission intensity by inverting the coefficient R n ​ The preset convergence condition is that the index of agreement IOA ≥ 0.8 and the emission intensity volatility of three consecutive iterations < 5%; The preset convergence condition is the consistency index: Inversion coefficient R n The calculation formula is as follows: The dynamic emission intensity correction is as follows: Among them, M i and O i are the data obtained from model simulation and observation respectively; n is the number of possible construction activities that may be emitted determined according to the construction layout and wind direction; N is the number of effective monitoring points; and are the mean values of the simulation values and the mean values of the observed values respectively; and are the simulation values and the observed values of the monitoring points corresponding to enterprise n respectively.

5. A method for inverse calculation of carbon emission sources at construction sites based on an atmospheric diffusion model according to claim 1, characterized in that, In step S4: The theoretical simulation includes using the Monte Carlo method to quantify the overall uncertainty U of sensor errors, model biases, and meteorological fluctuations total , and requires that U total < 25%; The actual detection uses 13 the CO2 isotope tracing technique, and the model accuracy is verified by detecting that the recovery rate η≥85%; Overall uncertainty U total The calculation formula is as follows: Among them, U sensor refers to the uncertainty measured by the sensor, that is, the accuracy range calibrated when the sensor leaves the factory; U model refers to the uncertainty of the model deviation; U meteo refers to the uncertainty of meteorological data, that is, the error introduced by inaccurate observation or prediction of the meteorological parameters input into the model.

6. The method for inverse calculation of carbon emission sources at construction sites based on an atmospheric diffusion model according to claim 1, characterized in that, The method further includes docking with the BIM system to generate a carbon emission heat map, and realizing tamper-proof data storage through blockchain technology.

7. A method for inversely calculating carbon emission sources at a construction site based on an atmospheric diffusion model according to claim 1, characterized in that In the method, the meteorological monitoring equipment includes a Vaisala WXT536 weather station, the CO2 sensor uses a SenseAir K30, and the edge computing node is an NVIDIA Jetson AGX Xavier.

8. A method for inverse calculation of carbon emission sources at construction sites based on an atmospheric diffusion model according to claim 1, characterized in that, In the isotope tracer verification, the released 13CO2 concentration is 0.1%, and the recovery rate is detected by a mass spectrometer.

Images