A monte carlo method-based industrial park carbon metering uncertainty evaluation method

Abstract

The application discloses an industrial park carbon measurement uncertainty evaluation method based on a Monte Carlo method, and full-caliber carbon emission accounting models containing carbon increase, carbon reduction, carbon export and green electricity energy storage special carbon are constructed; combined with multi-type variable uncertainty evaluation, multi-distribution random sampling, Monte Carlo iteration simulation and two-dimensional sensitivity analysis, global accurate evaluation of industrial park carbon measurement uncertainty is completed. Main steps include: S100, demarcating the carbon measurement accounting boundary of the industrial park and determining the carbon emission measurement model of the industrial park; S200, identifying and analyzing uncertainty sources; S300, establishing a Monte Carlo method MCM uncertainty evaluation mathematical model; S400, calculating the standard uncertainty of input variables; S500, calculating the random sampling of input variables; S600, iteratively calculating the total carbon emission of the industrial park; S700, statistically calculating the uncertainty of output; S800, calculating the sensitivity of key uncertainty sources; S900, verifying the convergence of MCM calculation. The application mainly aims at the systematic Monte Carlo uncertainty evaluation of full-caliber carbon income and expenditure (containing carbon increase, carbon reduction, carbon export and green electricity energy storage special carbon) of the industrial park, realizes high-precision uncertainty quantification of a new park composite carbon measurement scene, and provides a technical scheme for scientific, standardized and accurate carbon measurement of the industrial park.

Description

Technical Field

[0001] This invention relates to the field of carbon emission measurement and its uncertainty quantification technology, and in particular to a method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method. Background Technology

[0002] Reducing carbon emissions to address climate change is gradually becoming a global consensus, and energy conservation and carbon reduction are key means. Industrial parks, as concentrated areas of energy consumption and carbon emissions, directly impact the scientific validity and effectiveness of regional assessments, enterprise carbon measurement (or carbon verification), carbon trading accounting, and carbon emission control decisions, playing a crucial role in achieving carbon reduction targets. Uncertainty, as an important indicator for measuring the accuracy of carbon emission measurement, requires a highly effective assessment method.

[0003] Current carbon measurement uncertainty assessments generally employ the traditional GUM uncertainty assessment method. This method, based on the assumptions of linear error propagation and normal distribution, is only suitable for simple, linear, and low-variable traditional carbon emission accounting scenarios and has significant technical shortcomings: First, it cannot adapt to new nonlinear coupled accounting variables such as green electricity, energy storage, peak shaving, and carbon transfer, and linear approximation will cause bias in the assessment results; second, it only supports a single normal distribution assumption and cannot adapt to uniformly distributed or triangularly distributed variables such as emission factors, loss coefficients, and peak shaving coefficients; third, it cannot quantify the interference of newly added carbon reduction and carbon transfer variables on the overall uncertainty and cannot accurately locate the core error source; fourth, it only outputs a single extended uncertainty value and cannot reflect the true probability distribution characteristics of total carbon emissions.

[0004] Currently, the industry lacks a systematic Monte Carlo uncertainty assessment scheme for the full range of carbon revenue and expenditure (including carbon increase, carbon reduction, carbon relocation, and carbon for green energy storage projects) in industrial parks. This makes it impossible to achieve high-precision uncertainty quantification in new composite carbon measurement scenarios in industrial parks, thus hindering the standardization and accuracy of carbon measurement in industrial parks. Therefore, this invention proposes a Monte Carlo method for assessing the uncertainty of carbon measurement in industrial parks. Summary of the Invention

[0005] The purpose of this invention is to address the technical problems of existing technologies, such as the inability of the traditional GUM method to adapt to new composite carbon metering scenarios in industrial parks, including green electricity, energy storage, peak shaving, and internal and external carbon transfer. These problems include limitations in the linear assumptions of the model, poor adaptability of variable distribution, low accuracy in uncertainty assessment, inability to accurately identify core error sources, and inability to quantify the overall carbon budget. This invention provides a Monte Carlo method for assessing the uncertainty of carbon metering in industrial parks, which can accurately analyze the uncertainty of carbon emissions in industrial parks and improve the accuracy and reliability of carbon metering.

[0006] To achieve the aforementioned objectives, this invention provides a method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method. It delineates the boundaries of carbon measurement accounting for industrial parks and constructs a comprehensive carbon emission accounting model encompassing carbon increase, carbon reduction, carbon relocation, and carbon emissions from green energy storage projects. By combining multi-type variable uncertainty assessment, multi-distribution random sampling, Monte Carlo iterative simulation, and two-dimensional sensitivity analysis, it achieves a comprehensive and accurate assessment of the uncertainty in carbon measurement within industrial parks. The core technical solution is as follows:

[0007] A method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method (MCM) includes the following steps: S100 defines the boundaries for carbon accounting in industrial parks and determines the carbon emission measurement model for industrial parks. S200 identifies and analyzes sources of uncertainty; S300, Establish a mathematical model for evaluating the uncertainty of the Monte Carlo method MCM; S400, Calculation of standard uncertainty of input variables; S500, random sampling calculation of input variables; S600, iterative calculation of total carbon emissions of industrial parks; S700, statistical calculation of uncertainty of output quantity; S800, sensitivity calculation of key uncertainty sources; Convergence verification of S900 and MCM calculations.

[0008] The carbon measurement method for industrial parks mainly includes: establishing a carbon measurement model, collecting all the data required for carbon measurement (energy consumption data, energy production data, carbon emission factor data, etc.), calculating carbon emissions using the carbon measurement model, and assessing the uncertainty of carbon emissions. The measurement system may include carbon emission devices in the industrial park, various metering instruments, and edge carbon meters that receive carbon emission factors from the industrial park's carbon emission devices and read data from various metering instruments to calculate carbon emissions. This technical solution does not elaborate on the carbon measurement methods and systems, but only describes the content related to the assessment of carbon emission uncertainty.

[0009] 1. Define the carbon accounting boundaries for industrial parks and determine the carbon emission measurement model for industrial parks:

[0010] Taking into account factors such as energy input, energy conversion efficiency, energy output, and carbon emission factors of industrial parks, an industrial park carbon measurement model is established that can accurately measure the carbon emissions of industrial parks and their constituent units over any time period. In the formula C total For the total carbon emissions of the industrial park, Cdirect For total direct carbon emissions, C grid C represents the indirect carbon emissions generated from purchasing electricity and heat from the power grid and heat network. green_indirect C represents the indirect carbon emissions during the green electricity generation process. storage For the additional carbon emissions generated during green energy storage processes, C other For other carbon emissions, C peak_shaving The reduction in carbon emissions due to peak shaving by green electricity. grid_out The indirect carbon emissions generated from selling electricity and heat to the power grid and heat network, C green_out To reduce carbon emissions by providing green electricity outside the park, C other_out Carbon emissions borne by other transfers.

[0011] Total direct carbon emissions C direct Indirect carbon emissions C grid Indirect carbon emissions C during green electricity generation green_indirect Additional carbon emissions C generated during green energy storage processes storage Other carbon emissions C other Carbon emissions reduced by green electricity peak shaving C peak_shaving Indirect carbon emissions C from selling electricity and heat to the power grid and heat grid grid_out Carbon emissions reduced by providing green electricity outside the park (C) green_out Other carbon emissions transferred other_out The carbon scalar model is as follows:

[0012] (1) Total direct carbon emissions C direct Computational model: Among them, C comb For carbon emissions from the combustion of fossil fuels, AD i For the activity level of fossil fuel i, EF i The carbon emission factor for fossil fuel i. NCV i For the consumption of various fossil fuels, FC i The average lower heating value of fuel i, CC iwei The carbon content per unit calorific value of fuel i, OF i Let be the carbon oxidation rate of fuel i. Among them, C proc C represents carbon emissions from industrial production processes. 溶剂、 C 电极、 C 原料、 C 其它 P represents the carbon emissions generated from solvent consumption, electrode material consumption, carbon-containing raw material consumption, and other substance consumption during industrial production processes. i P 电极 These represent the activity level (consumption) of solvent i / raw material i, respectively, EF i EF 电极 These are the carbon emission factors of solvent i and raw material i, respectively, and the carbon emission factor of the electrode.

[0013] (2) Indirect carbon emission measurement C grid Computational model: Among them, C grid_e Indirect carbon emissions from purchasing electricity from the grid, E grid_e β represents the electricity purchased by the park from the external power grid. grid_e Carbon emission factor of power grid, C grid_WG E is the indirect carbon emissions from externally purchased heat. grid_WG For the heat power purchased from external sources for the park, β grid_WG It is a thermal carbon emission factor.

[0014] (3) Indirect carbon emissions C during green electricity generation green_indirect Computational model: Among them, C green_indirect For indirect carbon emissions during green electricity generation, E green γ represents the amount of green electricity generated within the park, and γ is the indirect carbon emission coefficient of green electricity generation.

[0015] (4) Additional carbon emissions C generated during green energy storage process storage Computational model: Among them, C storage For the additional carbon emissions generated during the green energy storage process, E storage η represents the amount of electricity charged into the green energy storage system, η represents the charging and discharging efficiency of the energy storage device, and β represents the energy storage capacity. storage This refers to the carbon emission factor corresponding to the energy loss of energy storage devices.

[0016] (5) Other carbon emissions Cother Computational model: Among them, C other For other carbon emissions, M other,i The amount of the i-th substance that causes other carbon emissions, EF other,i Let be the carbon emission factor of the i-th substance that causes other carbon emissions.

[0017] (6) Carbon emissions reduced by green electricity peak shaving C peak_shaving Computational model: Among them, C peak_shaving E reduces carbon emissions for peak shaving by green electricity peak_shaving β represents the amount of conventional energy generation that is replaced during green electricity peak shaving. conventional β is the carbon emission factor for conventional energy power generation. green The carbon emission factor for green electricity generation (a comprehensive factor considering indirect carbon emissions).

[0018] (7) Indirect carbon emissions C from selling electricity and heat to the power grid and heat grid grid_out Computational model: Among them, C grid_out C represents the indirect carbon emissions generated from selling electricity and heat to the power grid and heat network. grid_e_out E represents the indirect carbon emissions transferred from the sale of electricity from industrial parks to the grid. grid_e_out β represents the amount of electricity sold from the industrial park to the grid. grid_e_out Industrial parks sell electricity to the grid, carbon emission factor C grid_WG_out E represents the indirect carbon emissions generated by the industrial park selling heat to the heating network. grid_WG_out For the heat sold by the industrial park to the heating network, β grid_WG_out The industrial park sells carbon emission factors from the heating network.

[0019] (8) Carbon emissions reduced by providing green electricity outside the park C green_out Computational model: Among them, C green_out To reduce carbon emissions by providing green electricity outside the park, E green_out To provide green electricity to areas outside the park, β grid_e Carbon emissions from electricity supplied by the power grid to industrial parks.

[0020] (9) Other carbon emissions transferred to other countries other_out Computational model: Among them, C other_out M will bear the carbon emissions from other industries that are shifted out of industrial parks. other_out，i For the amount of other material i transferred out of the industrial park, EF other，i For other substances i that are transferred out of the industrial park, the carbon emission factor is...

[0021] 2. Identify and analyze sources of uncertainty:

[0022] Based on the total carbon emissions C of the industrial park total Analysis of the computational model, total carbon emissions C total Uncertainty u(C) total ) from the output sample set {C total,1 C total,2 ,…, C total,n (n is the number of samples, e.g., n=10) 5 The standard uncertainty is obtained.

[0023] Uncertainty about total direct carbon emissions u(C) direct Uncertainty about indirect carbon emissions from electricity purchased from the grid, u(C) grid Uncertainty about indirect carbon emissions during green electricity generation, u(C) green_indirect The uncertainty of additional carbon emissions u(C) generated by green energy storage processes storage Other carbon emission uncertainties u(C) othert Uncertainty about the amount of carbon emissions reduced by green electricity peak shaving, u(C) peak_shaving Uncertainty regarding the indirect carbon emissions u(C) generated from selling electricity and heat to the power grid and heat network. grid_out The uncertainty of carbon emission reduction u(C) by providing green electricity outside the park green_out The uncertainty of other carbon emissions transferred outside the industrial park, u(C) other_out The analysis is as follows:

[0024] (1) Based on the total direct carbon emissions C direct Analysis of the computational model, total direct carbon emissions C direct Uncertainty u(C) direct The uncertainty sources of ) mainly include the uncertainty of carbon emissions u(C) from the combustion of fossil fuels. comb The uncertainty of carbon emissions from industrial production processes, u(C) proc The uncertainty of carbon emissions from fossil fuel combustion, u(C).comb The main influencing factors include the consumption of various fossil fuels, calorific value, carbon content per unit calorific value, carbon oxidation rate, and carbon emission factor. The uncertainty of carbon emissions from industrial production processes, u(C), is... proc The main factors influencing the amount of raw materials consumed in the industrial production process and the carbon emission factors of those raw materials are as follows: or

[0025] (2) Based on the measurement of indirect carbon emissions C grid Analysis of the computational model, indirect carbon emission measurement C grid Uncertainty u(C) grid The uncertainty components of ) mainly include: the uncertainty of indirect carbon emissions from purchased electricity u(C) grid_e Uncertainty regarding indirect carbon emissions from purchased heat, u(C) grid_WG ).

[0026] (3) Based on the indirect carbon emissions C during the green electricity generation process green_indirect Analysis of the computational model, the indirect carbon emissions C during the green electricity generation process green_indirect Uncertainty u(C) green_indirect The uncertainty components of ) mainly include: the uncertainty of green electricity generation within the park, u(E) green The uncertainty of the indirect carbon emission coefficient u(γ) in the green electricity generation process.

[0027] (4) Based on the additional carbon emissions C generated by the green energy storage process storage Analysis of the computational model shows the additional carbon emissions C generated by the green energy storage process. storage Uncertainty u(C) storage The uncertainty components of ) mainly include: the uncertainty of the charging capacity of the green energy storage system, u(E) storage The uncertainty of the charge / discharge efficiency of the energy storage device, u(η), and the uncertainty of the carbon emission factor corresponding to the energy loss of the energy storage device, u(β). storage ).

[0028] (5) Based on other carbon emissions C other Analysis of the calculation model, other carbon emissions C other Uncertainty u(C) other The uncertainty components of ) mainly include: the uncertainty u(M) of the amount of the i-th substance that causes other carbon emissions. other,i The carbon emission factor uncertainty u(EF) of the i-th substance that causes other carbon emissions other,i ).

[0029] (6) Based on the reduction of carbon emissions C due to peak shaving of green electricity peak_shaving Analysis of the computational model shows that the carbon emissions C reduced by green electricity peak shaving peak_shaving Uncertainty u(C) peak_shaving The uncertainty components of ) mainly include: the uncertainty of conventional energy generation replaced by green electricity during peak shaving, u(E) peak_shaving The uncertainty of carbon emission factor u(β) for conventional energy power generation conventional The uncertainty u(β) of the carbon emission factor (a comprehensive factor considering indirect carbon emissions) of green electricity generation green ).

[0030] (7) Indirect carbon emissions C generated from selling electricity and heat to the grid and heat network grid_out Analysis of the calculation model shows the indirect carbon emissions C generated from selling electricity and heat to the power grid and heat network. grid_out Uncertainty component u(C) grid_out The main uncertainty includes: u(C) of indirect carbon emissions from selling electricity to the grid. grid_e_out Uncertainty regarding indirect carbon emissions from selling heat to the heating network, u(C) grid_WG_out ).

[0031] (8) Based on the reduction of carbon emissions C from providing green electricity outside the park green_out Analysis of the calculation model shows that providing green electricity outside the park reduces carbon emissions C. green_out Uncertainty u(C) green_out The uncertainty components mainly include: the uncertainty of the amount of green electricity supplied to the outside of the park, u(E). green_out The uncertainty of the carbon emission factor u(β) of electricity supplied by the power grid to the industrial parkgrid_e ).

[0032] (9) Carbon emissions C based on other transfers other_out Analysis of the calculation model, and the carbon emissions C transferred elsewhere. other_out Uncertainty u(C) other_out The uncertainty components mainly include: the uncertainty u(M) of the amount of substance i transferred out of the industrial park. other_out，i Other substances i that are transferred out of the industrial park have carbon emission factors u(EF). other，i ).

[0033] In summary, based on the analysis, the sources of uncertainty can be mainly categorized as follows: Uncertainty introduced by energy consumption measurement errors: The accuracy limitations of energy metering equipment in industrial parks lead to errors in the measurement of energy consumption data, such as errors in electricity meters, gas meters, heat meters, coal metering devices, as well as errors in the measurement of calorific value, carbon content, and carbon oxidation rate. Uncertainty introduced by material consumption measurement error: Measurement error of production raw materials used and consumed in the industrial park, such as solvent consumption error, electrode material consumption error, carbon-containing raw material consumption error and other substance consumption error, etc. Uncertainty introduced by carbon emission factor errors: Carbon emission factor data for different energy sources contain errors and are subject to uncertainty due to various factors such as energy quality and combustion efficiency. For example, the carbon emission factor of coal may fluctuate between (2.5~3.0) tCO2 / t depending on the origin and quality. Errors also exist in the carbon emission factor data for different materials. Errors also exist in the carbon emission factor data for electricity generation. Uncertainties introduced by errors related to green electricity ancillary services: green electricity generation error, green electricity carbon emission factor error; green electricity energy storage system charging error, discharging efficiency error, green electricity energy storage system energy loss carbon emission factor error; green electricity peak shaving substitution generation error, conventional energy generation carbon emission factor error, green electricity generation carbon emission factor error.

[0034] According to the parameter types in the carbon accounting model, the sources of uncertainty in carbon emissions are all mapped to the input variables, which can be mainly categorized into the following four types. Measurement uncertainty: The primary source, caused by the accuracy of measuring instruments, measurement methods, and environmental interference, such as measurement errors in energy consumption (electricity meters, gas meters, weighing instruments) and material quantity (flow meters, electronic scales), which meet the accuracy class standards of measuring instruments; Parameter uncertainty: Caused by deviations in the values ​​of parameters such as emission factors, oxidation rates, material conversion rates, and carbon content, such as differences between default emission factors and actual local values ​​in the park, and fluctuations in emission factor values ​​in different literature / standards; Model uncertainty: Caused by the selection of accounting models, boundary delineation, and accounting methods, such as the estimation model for fugitive emissions (empirical formulas / measured models), and the selection of accounting boundaries in range 3. This type of uncertainty can be converted into parameter uncertainty through scenario setting; Statistical and human-related uncertainty: Caused by data statistics, entry, and summarization, such as statistical deviations in monthly / annual energy data of enterprises, errors in manual entry, and omissions / misreporting of data by enterprises in the park. This type of uncertainty can be quantified through deviation analysis of historical data.

[0035] 3. Establish an uncertainty mathematical model:

[0036] For each source of uncertainty, a corresponding probability distribution model and uncertainty mathematical model are constructed based on its historical data, equipment manual, or relevant standards.

[0037] 4. Calculation of standard uncertainty of input variables

[0038] The standard uncertainty of the input variable is the core input parameter for uncertainty assessment in Monte Carlo MCM, and can be obtained through Type A and Type B assessments. In industrial parks, over 90% of the variables' uncertainties are assessed using Type B due to the lack of continuous measured data. The general formula for Type B standard uncertainty assessment is: In the formula, 'a' represents the uncertainty half-width of the variable, and 'k' is the coverage factor, which is determined by the type of probability distribution of the variable. For a normal distribution, the coverage factor k = 1.96 with a 95% coverage probability and k = 2.58 with a 99% coverage probability; for a uniform distribution, the coverage factor k ≈ 1.732; and for a triangular distribution, the coverage factor k ≈ 2.449.

[0039] The absolute standard uncertainty of each input variable in the carbon measurement model is calculated separately. Measurement-related input variables include those caused by the accuracy of measuring instruments, measurement methods, and environmental interference, such as measurement errors in energy consumption (electricity meters, gas meters, weighing instruments) and material quantity (flow meters, electronic scales), conforming to the accuracy class standards of the measuring instruments. Parameter-related input variables include parameters such as emission factors, oxidation rates, material conversion rates, and carbon content, such as the difference between the default emission factor and the actual local value in the park, and the fluctuation of different emission factor values. Model-related input variables include parameters related to model selection, boundary delineation, and calculation methods. Statistical and human-related input variables include deviations caused by data statistics, entry, and summarization, such as statistical deviations in monthly / annual energy data of enterprises, errors in manual entry, and omissions / misreports of data by enterprises in the park. These uncertainties can be quantified through deviation analysis of historical data.

[0040] The mean, standard uncertainty, and distribution type of all variables are summarized to form the final input parameter table for MCM, as follows: Natural gas consumption <![CDATA[Q1]]> Natural gas emission factors <![CDATA[EF1]]> diesel consumption <![CDATA[Q2]]> Diesel emission factors <![CDATA[EF2]]> Electricity consumption <![CDATA[Q3]]> Electricity emission factors <![CDATA[EF3]]>

[0041] 5. Random sampling calculation of input variables

[0042] For each input variable, generate a large number of random sample values ​​according to its probability distribution. The sampling must ensure the independence of each variable, and the sample value of one variable should not affect other variables.

[0043] Set sampling parameters: number of samples, random number seed, and range of sampled values.

[0044] The number of samples n is generally taken as (10) 4 ~10 6 In this embodiment, n=10. 5 10 times. Generate 10 for each variable. 5 Each sample value is used to form a sample set {x} 11 , x 12 , ..., x 1N (n sampled values ​​of the first variable), all sampled values ​​are calculated using the distribution-specific formula, without empirical estimation. The random number seed is fixed at 12345 to ensure the calculation results can be repeatedly verified. The range of sampled values ​​is constrained by "mean ± 3 × standard uncertainty".

[0045] The first sample values ​​of each variable in the carbon econometrics model are summarized to form a single sample set. A total of n sample sets need to be generated, as shown in the table below: First sample value First sample value … The nth sample value

[0046] 6. Iterative calculation of total carbon emissions in industrial parks

[0047] The generated sample set is then substituted into the park's carbon accounting model to calculate the total carbon emissions for each sampling period, forming the output sample set {C}. total,1 C total,2 ,…, C total,n}(n is the number of samples, which can be 10) 4 10 5 10 6 wait).

[0048] 7. Statistical calculation of uncertainty of output quantity

[0049] By statistically analyzing the output sample set of the total emissions from the sampling number, uncertainty indices (including but not limited to the best estimate, standard uncertainty, relative standard uncertainty, 95% quantile, 95% expanded uncertainty, and 95% coverage interval) are calculated to comprehensively quantify the uncertainty characteristics of carbon measurement in the park. Simulated mean (best estimate): ; Standard uncertainty: ; Relative standard uncertainty: ; 95th percentile C 0.025 After sorting the sample set in ascending order, the nth × 0.025 = 2500th value is obtained. 95th percentile C 0.975 After sorting the sample set in ascending order, the nth × 0.975 = 97500th value is obtained. 95% expanded uncertainty: ; 95% coverage area: [C 0.025 C 0.975 ].

[0050] The final assessment conclusion is: the best estimate of the total carbon emissions of the industrial park, with an expanded uncertainty under a 95% inclusion probability and a relative standard uncertainty, indicates that the actual carbon emissions have a 95% probability of falling within the inclusion range.

[0051] 8. Sensitivity calculation of key uncertainty sources

[0052] Industrial parks have numerous input variables, and the contributions of different variables to the uncertainty of total emissions vary significantly. This step identifies key variables through quantitative calculations, providing a targeted direction for subsequent uncertainty optimization. The Pearson correlation coefficient method and the variance contribution method are the most commonly used methods.

[0053] The formula for calculating the Pearson correlation coefficient is as follows:

[0054] The results of the Pearson correlation coefficient calculation should be filled in the table below: <![CDATA[Pearson correlation coefficient r j >

[0055] The formula for calculating the variance contribution method is as follows:

[0056] The variance contribution calculation results should be filled in the table below: <![CDATA[Variance contribution degree C j >

[0057] 9. Convergence verification of MCM calculation

[0058] To ensure the reliability of MCM calculation results, the sampling frequency is increased to check the rate of change of the indicators, verify the convergence of the sampling frequency, and avoid calculation deviations caused by insufficient sampling. In engineering, the simulated mean and standard uncertainty are used as convergence indicators. A sampling frequency of 10 can be selected. 3 10 4 10 5 wait.

[0059] The relative rate of change R of a certain indicator (mean or standard uncertainty) in two consecutive samples (n2 > n1) is R = |I n2 -I n1 | / |I n1 |*100%, where I is the simulated mean C or standard uncertainty u(C). Convergence is determined when R<0.1%.

[0060] The convergence verification results should be filled in the table below: 10³ <![CDATA[10 4 ]]> <![CDATA[10 5 ]]> <![CDATA[10 6 ]]>

[0061] When the relative rates of change of the mean and standard uncertainty are both less than 0.1% after a certain number of samplings, the convergence requirement is met, and the number of samplings is reasonable, with no need to increase it. However, if the relative rates of change are large, the number of samplings needs to be increased.

[0062] This application has achieved beneficial technical effects:

[0063] Improving the accuracy of carbon measurement: By comprehensively considering the various complex factors in carbon emission theory and green electricity ancillary services, and accurately assessing the uncertainties brought about by these factors, the accuracy of carbon emission measurement in industrial parks can be effectively improved, providing reliable data support for carbon emission reduction decisions.

[0064] Quantifying uncertainty: The MCM assessment method is used to quantify the uncertainty in the carbon measurement process, and to express the possible fluctuation range of carbon emissions with a clear numerical value, which helps enterprises and regulatory authorities to better understand the reliability of carbon measurement results.

[0065] High adaptability: This method can be adapted to different types of industrial parks. By adjusting the carbon measurement model and the probability distribution model of uncertainty sources, it can be applied to industrial parks with various energy consumption structures and green electricity ancillary service modes, and has broad application prospects. Attached Figure Description

[0066] Figure 1 A flowchart of a method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method. Detailed Implementation

[0067] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0068] An embodiment of the present invention provides a method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method, comprising the following steps: S100, determine the carbon emission measurement model for the industrial park; S200 identifies and analyzes sources of uncertainty; S300, establish a mathematical model for evaluating the uncertainty of MCM; S400, Calculation of standard uncertainty of input variables; S500, random sampling calculation of input variables; S600, iterative calculation of total carbon emissions of industrial parks; S700, statistical calculation of uncertainty of output quantity; S800, sensitivity calculation of key uncertainty sources; Convergence verification of S900 and MCM calculations.

[0069] The carbon measurement method for industrial parks mainly includes: establishing a carbon measurement model, collecting all the data required for carbon measurement (energy consumption data, energy production data, carbon emission factor data, etc.), calculating carbon emissions using the carbon measurement model, and assessing the uncertainty of carbon emissions. The measurement system may include carbon emission devices in the industrial park, various metering instruments, and edge carbon meters that receive carbon emission factors from the industrial park's carbon emission devices and read data from various metering instruments to calculate carbon emissions. This technical solution does not elaborate on the carbon measurement methods and systems, but only describes the content related to the assessment of carbon emission uncertainty.

[0070] 1. Define the carbon accounting boundaries for industrial parks and determine the carbon emission measurement model for industrial parks:

[0071] Taking into account factors such as energy input, energy conversion efficiency, energy output, and carbon emission factors of industrial parks, an industrial park carbon measurement model is established that can accurately measure the carbon emissions of industrial parks and their constituent units over any time period. In the formula C total For the total carbon emissions of the industrial park, C direct For total direct carbon emissions, C grid C represents the indirect carbon emissions generated from purchasing electricity and heat from the power grid and heat network. green_indirect C represents the indirect carbon emissions during the green electricity generation process. storage For the additional carbon emissions generated during green energy storage processes, C other For other carbon emissions, C peak_shaving The reduction in carbon emissions due to peak shaving by green electricity. grid_out The indirect carbon emissions generated from selling electricity and heat to the power grid and heat network, C green_out To reduce carbon emissions by providing green electricity outside the park, C other_out Carbon emissions borne by other transfers.

[0072] Total direct carbon emissions C direct Indirect carbon emissions C grid Indirect carbon emissions C during green electricity generation green_indirect Additional carbon emissions C generated during green energy storage processes storage Other carbon emissions C other Carbon emissions reduced by green electricity peak shaving C peak_shaving Indirect carbon emissions C from selling electricity and heat to the power grid and heat grid grid_out Carbon emissions reduced by providing green electricity outside the park (C) green_out Other carbon emissions transferred other_out The carbon scalar model is as follows:

[0073] (1) Total direct carbon emissions C direct Computational model: Among them, C comb For carbon emissions from the combustion of fossil fuels, AD i For the activity level of fossil fuel i, EF i The carbon emission factor for fossil fuel i. NCV i For the consumption of various fossil fuels, FCi The average lower heating value of fuel i, CC iwei The carbon content per unit calorific value of fuel i, OF i Let be the carbon oxidation rate of fuel i. Among them, C proc C represents carbon emissions from industrial production processes. 溶剂、 C 电极、 C 原料、 C 其它 P represents the carbon emissions generated from solvent consumption, electrode material consumption, carbon-containing raw material consumption, and other substance consumption during industrial production processes. i P 电极 These represent the activity level (consumption) of solvent i / raw material i, respectively, EF i EF 电极 These are the carbon emission factors of solvent i and raw material i, respectively, and the carbon emission factor of the electrode.

[0074] (2) Indirect carbon emission measurement C grid Computational model: Among them, C grid_e Indirect carbon emissions from purchasing electricity from the grid, E grid_e β represents the electricity purchased by the park from the external power grid. grid_e Carbon emission factor of power grid, C grid_WG E is the indirect carbon emissions from externally purchased heat. grid_WG For the heat power purchased from external sources for the park, β grid_WG It is a thermal carbon emission factor.

[0075] (3) Indirect carbon emissions C during green electricity generation green_indirect Computational model: Among them, C green_indirect For indirect carbon emissions during green electricity generation, E green γ represents the amount of green electricity generated within the park, and γ is the indirect carbon emission coefficient of green electricity generation.

[0076] (4) Additional carbon emissions C generated during green energy storage process storage Computational model: Among them, C storage For the additional carbon emissions generated during the green energy storage process, E storage η represents the amount of electricity charged into the green energy storage system, η represents the charging and discharging efficiency of the energy storage device, and β represents the energy storage capacity. storage This refers to the carbon emission factor corresponding to the energy loss of energy storage devices.

[0077] (5) Other carbon emissions C other Computational model: Among them, C other For other carbon emissions, M other,i The amount of the i-th substance that causes other carbon emissions, EF other,i Let be the carbon emission factor of the i-th substance that causes other carbon emissions.

[0078] (6) Carbon emissions reduced by green electricity peak shaving C peak_shaving Computational model: Among them, C peak_shaving E reduces carbon emissions for peak shaving by green electricity peak_shaving β represents the amount of conventional energy generation that is replaced during green electricity peak shaving. conventional β is the carbon emission factor for conventional energy power generation. green The carbon emission factor for green electricity generation (a comprehensive factor considering indirect carbon emissions).

[0079] (7) Indirect carbon emissions C from selling electricity and heat to the power grid and heat grid grid_out Computational model: Among them, C grid_out C represents the indirect carbon emissions generated from selling electricity and heat to the power grid and heat network. grid_e_out E represents the indirect carbon emissions transferred from the sale of electricity from industrial parks to the grid. grid_e_out β represents the amount of electricity sold from the industrial park to the grid. grid_e_out Industrial parks sell electricity to the grid, carbon emission factor C grid_WG_out E represents the indirect carbon emissions generated by the industrial park selling heat to the heating network. grid_WG_out For the heat sold by the industrial park to the heating network, β grid_WG_out The industrial park sells carbon emission factors from the heating network.

[0080] (8) Carbon emissions reduced by providing green electricity outside the park C green_out Computational model: Among them, C green_out To reduce carbon emissions by providing green electricity outside the park, E green_out To provide green electricity to areas outside the park, β grid_e Carbon emissions from electricity supplied by the power grid to industrial parks.

[0081] (9) Other carbon emissions transferred to other countries other_out Computational model: Among them, C other_out M will bear the carbon emissions from other industries that are shifted out of industrial parks. other_out，i For the amount of other material i transferred out of the industrial park, EF other，i For other substances i that are transferred out of the industrial park, the carbon emission factor is...

[0082] 2. Identify and analyze sources of uncertainty:

[0083] Based on the total carbon emissions C of the industrial park total Analysis of the computational model, total carbon emissions C total Uncertainty u(C) total ) from the output sample set {C total,1 C total,2 ,…, C total,n (n is the number of samples, e.g., n=10) 5 The standard uncertainty is obtained.

[0084] Uncertainty about total direct carbon emissions u(C) direct Uncertainty about indirect carbon emissions from electricity purchased from the grid, u(C) grid Uncertainty about indirect carbon emissions during green electricity generation, u(C) green_indirect The uncertainty of additional carbon emissions u(C) generated by green energy storage processes storage Other carbon emission uncertainties u(C) othert Uncertainty about the amount of carbon emissions reduced by green electricity peak shaving, u(C) peak_shaving Uncertainty regarding the indirect carbon emissions u(C) generated from selling electricity and heat to the power grid and heat network. grid_out The uncertainty of carbon emission reduction u(C) by providing green electricity outside the park green_out The uncertainty of other carbon emissions transferred outside the industrial park, u(C) other_out The analysis is as follows:

[0085] (1) Based on the total direct carbon emissions C directAnalysis of the computational model, total direct carbon emissions C direct Uncertainty u(C) direct The uncertainty sources of ) mainly include the uncertainty of carbon emissions u(C) from the combustion of fossil fuels. comb The uncertainty of carbon emissions from industrial production processes, u(C) proc The uncertainty of carbon emissions from fossil fuel combustion, u(C). comb The main influencing factors include the consumption of various fossil fuels, calorific value, carbon content per unit calorific value, carbon oxidation rate, and carbon emission factor. The uncertainty of carbon emissions from industrial production processes, u(C), is... proc The main factors influencing the amount of raw materials consumed in the industrial production process and the carbon emission factors of those raw materials are as follows: or

[0086] (2) Based on the measurement of indirect carbon emissions C grid Analysis of the computational model, indirect carbon emission measurement C grid Uncertainty u(C) grid The uncertainty components of ) mainly include: the uncertainty of indirect carbon emissions from purchased electricity u(C) grid_e Uncertainty regarding indirect carbon emissions from purchased heat, u(C) grid_WG ).

[0087] (3) Based on the indirect carbon emissions C during the green electricity generation process green_indirect Analysis of the computational model, the indirect carbon emissions C during the green electricity generation process green_indirect Uncertainty u(C) green_indirect The uncertainty components of ) mainly include: the uncertainty of green electricity generation within the park, u(E) green The uncertainty of the indirect carbon emission coefficient u(γ) in the green electricity generation process.

[0088] (4) Based on the additional carbon emissions C generated by the green energy storage process storage Analysis of the computational model shows the additional carbon emissions C generated by the green energy storage process. storageUncertainty u(C) storage The uncertainty components of ) mainly include: the uncertainty of the charging capacity of the green energy storage system, u(E) storage The uncertainty of the charge / discharge efficiency of the energy storage device, u(η), and the uncertainty of the carbon emission factor corresponding to the energy loss of the energy storage device, u(β). storage ).

[0089] (5) Based on other carbon emissions C other Analysis of the calculation model, other carbon emissions C other Uncertainty u(C) other The uncertainty components of ) mainly include: the uncertainty u(M) of the amount of the i-th substance that causes other carbon emissions. other,i The carbon emission factor uncertainty u(EF) of the i-th substance that causes other carbon emissions other,i ).

[0090] (6) Based on the reduction of carbon emissions C due to peak shaving of green electricity peak_shaving Analysis of the computational model shows that the carbon emissions C reduced by green electricity peak shaving peak_shaving Uncertainty u(C) peak_shaving The uncertainty components of ) mainly include: the uncertainty of conventional energy generation replaced by green electricity during peak shaving, u(E) peak_shaving The uncertainty of carbon emission factor u(β) for conventional energy power generation conventional The uncertainty u(β) of the carbon emission factor (a comprehensive factor considering indirect carbon emissions) of green electricity generation green ).

[0091] (7) Indirect carbon emissions C generated from selling electricity and heat to the grid and heat network grid_out Analysis of the calculation model shows the indirect carbon emissions C generated from selling electricity and heat to the power grid and heat network. grid_out Uncertainty component u(C) grid_out The main uncertainty includes: u(C) of indirect carbon emissions from selling electricity to the grid. grid_e_out Uncertainty regarding indirect carbon emissions from selling heat to the heating network, u(C) grid_WG_out ).

[0092] (8) Based on the reduction of carbon emissions C from providing green electricity outside the park green_out Analysis of the calculation model shows that providing green electricity outside the park reduces carbon emissions C. green_out Uncertainty u(C) green_out The uncertainty components mainly include: the uncertainty of the amount of green electricity supplied to the outside of the park, u(E). green_out The uncertainty of the carbon emission factor u(β) of electricity supplied by the power grid to the industrial park grid_e ).

[0093] (9) Carbon emissions C based on other transfers other_out Analysis of the calculation model, and the carbon emissions C transferred elsewhere. other_out Uncertainty u(C) other_out The uncertainty components mainly include: the uncertainty u(M) of the amount of substance i transferred out of the industrial park. other_out，i Other substances i that are transferred out of the industrial park have carbon emission factors u(EF). other，i ).

[0094] In summary, based on the analysis, the sources of uncertainty can be mainly categorized as follows: Uncertainty introduced by energy consumption measurement errors: The accuracy limitations of energy metering equipment in industrial parks lead to errors in the measurement of energy consumption data, such as errors in electricity meters, gas meters, heat meters, coal metering devices, as well as errors in the measurement of calorific value, carbon content, and carbon oxidation rate. Uncertainty introduced by material consumption measurement error: Measurement error of production raw materials used and consumed in the industrial park, such as solvent consumption error, electrode material consumption error, carbon-containing raw material consumption error and other substance consumption error, etc. Uncertainty introduced by carbon emission factor errors: Carbon emission factor data for different energy sources contain errors and are subject to uncertainty due to various factors such as energy quality and combustion efficiency. For example, the carbon emission factor of coal may fluctuate between (2.5~3.0) tCO2 / t depending on the origin and quality. Errors also exist in the carbon emission factor data for different materials. Errors also exist in the carbon emission factor data for electricity generation. Uncertainties introduced by errors related to green electricity ancillary services: green electricity generation error, green electricity carbon emission factor error; green electricity energy storage system charging error, discharging efficiency error, green electricity energy storage system energy loss carbon emission factor error; green electricity peak shaving substitution generation error, conventional energy generation carbon emission factor error, green electricity generation carbon emission factor error.

[0095] According to the parameter types in the carbon accounting model, the sources of uncertainty in carbon emissions are all mapped to the input variables, which can be mainly categorized into the following four types. Measurement uncertainty: The primary source, caused by the accuracy of measuring instruments, measurement methods, and environmental interference, such as measurement errors in energy consumption (electricity meters, gas meters, weighing instruments) and material quantity (flow meters, electronic scales), which meet the accuracy class standards of measuring instruments; Parameter uncertainty: Caused by deviations in the values ​​of parameters such as emission factors, oxidation rates, material conversion rates, and carbon content, such as differences between default emission factors and actual local values ​​in the park, and fluctuations in emission factor values ​​in different literature / standards; Model uncertainty: Caused by the selection of accounting models, boundary delineation, and accounting methods, such as the estimation model for fugitive emissions (empirical formulas / measured models), and the selection of accounting boundaries in range 3. This type of uncertainty can be converted into parameter uncertainty through scenario setting; Statistical and human-related uncertainty: Caused by data statistics, entry, and summarization, such as statistical deviations in monthly / annual energy data of enterprises, errors in manual entry, and omissions / misreporting of data by enterprises in the park. This type of uncertainty can be quantified through deviation analysis of historical data.

[0096] 3. Establish an uncertainty mathematical model:

[0097] For each source of uncertainty, a corresponding probability distribution model and uncertainty mathematical model are constructed based on its historical data, equipment manual, or relevant standards.

[0098] 4. Calculation of standard uncertainty of input variables

[0099] The standard uncertainty of the input variable is the core input parameter for uncertainty assessment in Monte Carlo MCM, and can be obtained through Type A and Type B assessments. In industrial parks, over 90% of the variables' uncertainties are assessed using Type B due to the lack of continuous measured data. The general formula for Type B standard uncertainty assessment is: In the formula, 'a' represents the uncertainty half-width of the variable, and 'k' is the coverage factor, which is determined by the type of probability distribution of the variable. For a normal distribution, the coverage factor k = 1.96 with a 95% coverage probability and k = 2.58 with a 99% coverage probability; for a uniform distribution, the coverage factor k ≈ 1.732; and for a triangular distribution, the coverage factor k ≈ 2.449.

[0100] The absolute standard uncertainty of each input variable in the carbon measurement model is calculated separately. Measurement-related input variables include those caused by the accuracy of measuring instruments, measurement methods, and environmental interference, such as measurement errors in energy consumption (electricity meters, gas meters, weighing instruments) and material quantity (flow meters, electronic scales), conforming to the accuracy class standards of the measuring instruments. Parameter-related input variables include parameters such as emission factors, oxidation rates, material conversion rates, and carbon content, such as the difference between the default emission factor and the actual local value in the park, and the fluctuation of different emission factor values. Model-related input variables include parameters related to model selection, boundary delineation, and calculation methods. Statistical and human-related input variables include deviations caused by data statistics, entry, and summarization, such as statistical deviations in monthly / annual energy data of enterprises, errors in manual entry, and omissions / misreports of data by enterprises in the park. These uncertainties can be quantified through deviation analysis of historical data.

[0101] The mean, standard uncertainty, and distribution type of all variables are summarized to form the final input parameter table for MCM, as follows: Natural gas consumption <![CDATA[Q1]]> <![CDATA[800*10 4 m³]]> <![CDATA[2.0408*10 4 m³]]> normal distribution Natural gas emission factors <![CDATA[EF1]]> 2.1654 0.0231 Uniform distribution diesel consumption <![CDATA[Q2]]> 150 t 0.1531 t normal distribution Diesel emission factors <![CDATA[EF2]]> 3.156 0.0245 triangular distribution Electricity consumption <![CDATA[Q3]]> <![CDATA[1200 *10 4 kWh]]> <![CDATA[3.0612 *10 4 kWh]]> normal distribution Electricity emission factors <![CDATA[EF3]]> 0.568 0.0123 triangular distribution

[0102] 5. Random sampling calculation of input variables

[0103] For each input variable, generate a large number of random sample values ​​according to its probability distribution. The sampling must ensure the independence of each variable, and the sample value of one variable should not affect other variables.

[0104] Set sampling parameters: number of samples, random number seed, and range of sampled values.

[0105] The number of samples n is generally taken as (10) 4 ~10 6 In this embodiment, n=10. 5 10 times. Generate 10 for each variable. 5 Each sample value is used to form a sample set {x} 11 , x 12 , ..., x 1N (n sampled values ​​of the first variable), all sampled values ​​are calculated using the distribution-specific formula, without empirical estimation. The random number seed is fixed at 12345 to ensure the calculation results can be repeatedly verified. The range of sampled values ​​is constrained by "mean ± 3 × standard uncertainty".

[0106] The first sample values ​​of each variable in the carbon econometrics model are summarized to form a single sample set, and a total of sample sets need to be generated for a total of sampling times. First sample value 800.2894 2.1494 150.0217 3.1623 1200.4341 0.5712 First sample value 800.8894 2.1291 151.2125 3.2522 1207.3365 0.5884 … <![CDATA[10th 5 sampling value]]> 799.8597 2.1272 149.9125 3.1572 1202.2376 0.5781

[0107] 6. Iterative calculation of total carbon emissions in industrial parks

[0108] Substituting the generated sample set into the park's carbon accounting model one by one, we obtain n=10.5 The total carbon emissions of each industrial park form an output sample set {C} total,1 C total,2 ,…, C total,n}(n=10 5 In one embodiment, the output sample set contains values ​​of {2879.24, 2885.67, 2868.91, ...,2890.12, 2872.35}.

[0109] 7. Statistical calculation of uncertainty of output quantity

[0110] By sampling 10 5 A sample set of outputs of total emissions {C total,1 C total,2 ,…, C total,n}(n=10 5 Statistical analysis was conducted to calculate uncertainty indices (including but not limited to the best estimate, standard uncertainty, relative standard uncertainty, 95% quantile, 95% expanded uncertainty, and 95% coverage interval), comprehensively quantifying the uncertainty characteristics of carbon measurement in the park. Simulated mean (best estimate): ; Standard uncertainty: ; Relative standard uncertainty: ; 95th percentile C 0.025 After sorting the sample set in ascending order, the nth × 0.025 = 2500th value is obtained. 95th percentile C 0.975 After sorting the sample set in ascending order, the nth × 0.975 = 97500th value is obtained. 95% expanded uncertainty: ; 95% coverage area: [C 0.025 C 0.975 ].

[0111] The final assessment conclusion is as follows: the best estimate of the total carbon emissions of the industrial park is 2880.56 tCO2, the standard uncertainty of the total emissions is 68.32 tCO2, the relative standard uncertainty is 2.37%, the expanded uncertainty at a 95% coverage probability is 132.41 tCO2, and the actual carbon emissions have a 95% probability of falling within the coverage range [2748.15, 3012.97] tCO2.

[0112] 8. Sensitivity calculation of key uncertainty sources

[0113] The industrial park has numerous input variables, and the contribution of different variables to the uncertainty of total emissions varies significantly. This step identifies key variables through quantitative calculations, providing a targeted direction for subsequent uncertainty optimization.

[0114] The results calculated based on the Pearson correlation coefficient are shown in the table below: <![CDATA[Pearson correlation coefficient r j > 0.89 0.68 0.18 0.23 0.76 0.15

[0115] The results of the variance contribution calculation are shown in the table below: <![CDATA[Variance contribution degree C j > 45.2% 12.8% 3.1% 4.7% 32.6% 1.6%

[0116] The core sources of uncertainty in carbon metering in this industrial park in this embodiment are: natural gas consumption Q1 (contribution 45.2%) and electricity consumption Q3 (contribution 32.6%), which together contribute 77.8%. Subsequent optimization should prioritize these two variables.

[0117] 9. Convergence verification of MCM calculation

[0118] To ensure the reliability of MCM calculation results, the sampling frequency is increased to check the rate of change of the index, verify the convergence of the sampling frequency, and avoid calculation deviations caused by insufficient sampling. In engineering, the simulated mean and standard uncertainty are used as convergence indices. In this embodiment, the sampling frequency is 10. 3 10 4 10 5 5*10 5 The convergence verification results are shown in the table below: 10³ 2878.21 69.15 - - <![CDATA[10 4 ]]> 2880.15 68.57 0.067% 0.84% <![CDATA[10 5 ]]> 2880.56 68.32 0.014% 0.36% <![CDATA[5×10 5 ]]> 2880.61 68.30 0.002% 0.029%

[0119] When the number of samplings reaches 10 5 At this time, the relative rates of change of the mean and standard uncertainty are both <0.1%, which meets the convergence requirement. In this case, 10 is selected. 5 The current sampling is reasonable, and there is no need to increase the number of samplings.

[0120] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0121] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

[0122] The embodiments of the Monte Carlo method-based carbon measurement uncertainty assessment method for industrial parks provided by this invention have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this invention, and the descriptions of the embodiments above are only for the purpose of helping to understand the core ideas of this invention. It should be noted that those skilled in the art can make several improvements and modifications to this invention without departing from the principles of this invention, and these improvements and modifications also fall within the protection scope of the claims of this invention.

Claims

1. A Monte Carlo method for assessing the uncertainty of carbon measurement in industrial parks. This method constructs a comprehensive carbon emission accounting model encompassing carbon increase, carbon reduction, carbon relocation, and carbon emissions from green energy storage projects. It combines multi-variable uncertainty assessment, multi-distribution random sampling, Monte Carlo iterative simulation, and two-dimensional sensitivity analysis to achieve a comprehensive and accurate assessment of the uncertainty in carbon measurement within industrial parks. The method includes the following steps: S100 defines the boundaries for carbon accounting in industrial parks and determines the carbon emission measurement model for industrial parks. S200 identifies and analyzes sources of uncertainty; S300, establish a mathematical model for evaluating the uncertainty of MCM; S400, Calculation of standard uncertainty of input variables; S500, random sampling calculation of input variables; S600, iterative calculation of total carbon emissions of industrial parks; S700, statistical calculation of uncertainty of output quantity; S800, sensitivity calculation of key uncertainty sources; Convergence verification of S900 and MCM calculations.

2. The method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method according to claim 1, characterized in that, In step S200, the sources of uncertainty include seven categories: instrument measurement error, emission factor default deviation, data statistical error, equipment operation and maintenance fluctuation, randomness of green electricity output, energy storage loss fluctuation, and carbon transfer accounting boundary deviation. Among them, energy consumption, electricity / heat purchase and sale data, and production capacity statistics are assessed using Type A, and the standard uncertainty is obtained by calculating the sample standard deviation based on long-term measured repeated data. Emission factor, energy storage loss coefficient, peak shaving and emission reduction coefficient, and empirical correction coefficient are assessed using Type B, and the standard uncertainty is calculated based on metrological specifications, industry standards, and upper and lower limits of error.

3. The method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method according to claim 1, characterized in that, The variable distribution fitting rules in step S300 are as follows: for continuous variables measured by instruments, a normal distribution is fitted; for industry default parameters without optimal values, a uniform distribution is fitted; and for correction parameters with optimal values ​​and upper and lower limit constraints, a triangular distribution is fitted.

4. The method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method according to claim 1, characterized in that, In step S500, the Monte Carlo simulation parameter setting rule is: when the number of variables is less than 30, the number of samples is set to 10. 5 When the number of variables is 30 to 60, the sampling number is set to 5 × 10. 5 The convergence criterion is that the relative rate of change of the mean and standard uncertainty of multiple consecutive batches of sampling is less than 0.1%.

5. The method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method according to claim 1, characterized in that, In step S800, the sensitivity analysis adopts a dual judgment of Pearson correlation coefficient and variance contribution. By calculating the correlation coefficient between each input variable and the total output carbon emissions, and the proportion of univariate variance contribution, the core uncertainty interference variables are sorted and identified, providing a targeted basis for optimizing the accuracy of carbon measurement.

6. The method for assessing the uncertainty of carbon measurement in industrial parks based on the Monte Carlo method according to claim 1, characterized in that, The accounting logic for each carbon emission variable is as follows: direct carbon emissions C direct It is obtained by summing the emissions from fossil fuel combustion, industrial processes, and waste disposal within the park; the purchased indirect carbon emissions C grid The carbon emissions C from green electricity are obtained by summing the products of the purchased electricity and purchased heat corresponding to the regional power grid and the thermal emission factor, respectively. green_indirect The carbon emissions from green energy storage are calculated based on indirect emissions from the entire lifecycle of photovoltaic and wind power equipment production, operation, maintenance, and cleaning; C is the additional carbon emissions from green energy storage. storage Carbon emissions are calculated from energy storage battery charging and discharging losses, equipment operation and maintenance, and temperature control energy consumption; green electricity peak shaving emission reduction C peak_shaving Equivalent carbon reduction is calculated based on green electricity peak-shaving replacement of thermal power generation and thermal power baseline emission factors; carbon emissions transferred from external electricity and heat sales (C2) grid_out The carbon transfer amount is calculated based on the electricity and heat transmitted, as well as the corresponding emission factors; the carbon reduction C from green electricity supplied externally. green_out Regional emission reductions are calculated based on the amount of green electricity supplied from outside the region and the emission factors that replace traditional thermal power.

Images