A carbon emission calculation method for a distributed power random output power distribution network

Abstract

This invention provides a method for calculating carbon emissions in a distributed generation (DG) stochastic output distribution network. The method includes: acquiring raw data of the DG stochastic output distribution network; inputting the raw data into a preset node admittance matrix to determine the distribution probability of parameters at each node in the distribution network, and performing deterministic power flow calculation based on the determined distribution probability to obtain the randomly distributed generator carbon emission intensity; calculating the corresponding judgment coefficient based on the obtained randomly distributed generator carbon emission intensity; determining whether convergence is required based on the judgment coefficient; if convergence is required, redetermining the distribution probability of each node parameter and re-performing the deterministic power flow calculation until convergence is no longer required; if convergence is not required, outputting the obtained randomly distributed generator carbon emission intensity and power flow results as the final calculation result. This invention lays a solid foundation for the comprehensive analysis and evaluation of carbon emissions under various operating conditions and meets the requirements for carbon flow calculation in distribution networks containing DG.

Description

Technical Field

[0001] This invention relates to the field of power system automation technology, and in particular to a method for calculating carbon emissions in a distributed power generation random output distribution network. Background Technology

[0002] Energy crisis and environmental pollution are major threats facing humanity today and significant obstacles to social progress. Building a low-carbon society and reducing the over-exploitation of fossil fuels are crucial pathways to achieving sustainable development. To comprehensively analyze and assess system carbon emissions, it is necessary to establish carbon emission indicators and develop an accurate calculation method. To date, relevant research has been conducted on carbon emission flows. Among these, the carbon emission calculation method based on macroscopic data analysis is simple, practical, and highly applicable in engineering; however, it cannot determine the specific distribution of carbon emissions and cannot achieve emergency control in areas with severe carbon pollution. The carbon emission calculation method based on power flow data can solve in detail for the carbon emissions generated and absorbed by system transmission lines, voltage sources, and loads. Simultaneously, it can accurately calculate the carbon flow density of each branch, exhibiting strong theoretical basis and good applicability.

[0003] With the vigorous construction of smart grids, distributed generation technology is constantly innovating and developing. Distributed power sources, represented by wind turbines, photovoltaics, and energy storage, are widely used in microgrid grid-connected systems due to their clean, flexible, and sustainable characteristics. Distributed power generation systems are affected by sunlight, temperature, and climate, resulting in intermittent, random, and fluctuating power output, leading to random carbon emissions. Under these conditions, system voltage, current, and power cannot remain constant and satisfy a certain probability distribution, making it impossible to solve for carbon emissions based on conventional power flow data. Therefore, the carbon flow calculation method for distribution networks containing distributed power sources requires further research. Thus, this invention proposes a method for calculating carbon emission flows in distribution networks that takes into account the random output of distributed power sources. Summary of the Invention

[0004] The purpose of this invention is to propose a carbon emission calculation method for distributed power generation random output distribution networks, thereby solving the technical problem that existing methods cannot meet the carbon flow calculation requirements of distribution networks containing distributed power generation.

[0005] On the one hand, a method for calculating carbon emissions in distributed generation stochastic output distribution networks is provided, including:

[0006] Obtain raw data of the distributed generation random output distribution network;

[0007] The raw data is input into a preset node admittance matrix to determine the distribution probability of the parameters of each node in the distribution network, and deterministic power flow calculation is performed based on the determined distribution probability to obtain the generator carbon emission intensity distributed randomly.

[0008] Calculate the corresponding judgment coefficient based on the randomly distributed generator carbon emission intensity; determine whether convergence is required based on the judgment coefficient. If convergence is required, redetermine the distribution probability of each node parameter and recalculate the deterministic power flow until convergence is no longer required; if convergence is not required, output the randomly distributed generator carbon emission intensity and power flow results as the final calculation results.

[0009] Preferably, determining the distribution probability of parameters at each node of the distribution network specifically includes:

[0010] The raw data is input into a preset node admittance matrix to determine the type of parameters of each node in the distribution network;

[0011] Calculate the corresponding random number based on the type of parameters of each node in the distribution network, and determine the distribution probability of each node parameter based on the obtained random number;

[0012] The types of the parameters of each node are at least discrete random variables and continuous random variables.

[0013] Preferably, when the type of distribution network node parameters is discrete random variables, the probability distribution is calculated according to the following formula:

[0014] P{X=x k}=p k k = 1, 2, 3…

[0015]

[0016] Where, P{X=x k} represents the probability of X taking each possible value, k represents the sequence number of each possible value, and E(X) represents the random number corresponding to the random variable X; x k p represents the possible values ​​of a random variable. k This indicates the probability of taking this value.

[0017] Preferably, when the type of distribution network node parameters is a continuous random variable, the probability distribution is calculated according to the following formula:

[0018]

[0019]

[0020] Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters.

[0021] Preferably, obtaining the generator carbon emission intensity according to a random distribution specifically includes:

[0022] The active power flow vector of each node in the lossless network is determined using the following formula:

[0023] A u ·P g =P G

[0024]

[0025] Among them, P g A represents the active power flow vector of each node after being converted into an equivalent lossless network. u Represents an n x n matrix [A] u ] ij P G Indicates the injected power of the generator. Let c be the set of nodes that directly flow into the active power flow from node i. ij P is the power sharing factor. j-i Let P be the active power flowing from node j to node i. j Let be the active power flowing through node j.

[0026] Preferably, it further includes calculating the equivalent active power flow of the branch and the equivalent active load of the node according to the following formula:

[0027]

[0028]

[0029] Among them, |P (g) i-j | represents the equivalent active power flow of branch ij, [A u -1 ] ik Let P represent an n x n matrix. i (g) P represents the active power of node i in the equivalent lossless network. GK P represents the power output of the equivalent generator node. i Let P be the active power flowing through node i. i-j Let P be the active power flowing from node i to node j, n represent the total number of nodes, k represent the equivalent generator node, and P' be the active power. Li This represents the actual load of node i per unit time. This represents the equivalent active load of node i.

[0030] Preferably, the method further includes calculating the nodal carbon potential, branch carbon flux density, and branch carbon flux rate of the distribution network according to the following formulas:

[0031]

[0032] R L =ρL ·P L

[0033] ρ loadi =e Ni ·P load

[0034] ρ source =e G ·P source

[0035] Where, ρ s E represents the carbon flux density of branch s. Gi P represents the carbon emission intensity of voltage source i. BS P represents the active power injected into the branch node. Gi ρ represents the active power injected into node i by the voltage source. L P represents the branch carbon flux density. L e represents the active power flowing out of the branch node. Ni ρ represents the carbon potential at the node where the load is located. loadi P represents the carbon emissions directly generated by the load. load ρ represents the active power consumed by the load. source This indicates the amount of carbon emissions directly generated by the generator; e G Indicates the carbon emission intensity of the generator; P source This indicates the active power injected into the generator.

[0036] Preferably, the step of calculating the corresponding judgment coefficient based on the obtained randomly distributed generator carbon emission intensity specifically includes:

[0037] Determine the types of nodal carbon potential, branch carbon flux density, and branch carbon flux rate of the calculated distribution network.

[0038] When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a discrete random variable, the probability distribution is calculated according to the following formula:

[0039] P{X=x k}=p k k = 1, 2, 3...

[0040]

[0041] Where, P{X=x k} represents the probability of X taking each possible value, k represents the sequence number of each possible value, and E(X) represents the random number corresponding to the random variable X; x k p represents the possible values ​​of a random variable. k This indicates the probability of taking that value;

[0042] When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a continuous random variable, the probability distribution is calculated according to the following formula:

[0043]

[0044]

[0045] Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters.

[0046] Preferably, the corresponding judgment coefficient is calculated according to the following formula:

[0047] D(X)=E{[XE(X)] 2}

[0048] Where D(X) represents the judgment coefficient, and E{[XE(X)]} 2 Let} represent the variance of the random variable X, and E(X) represent the random number corresponding to the random variable X.

[0049] Preferably, the step of determining whether the coefficients need to converge specifically includes:

[0050] When the judgment coefficient is less than the preset convergence precision, it is determined that convergence is not required;

[0051] When the judgment coefficient is greater than or equal to the preset convergence precision, convergence is determined.

[0052] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0053] The carbon emission calculation method for distributed power generation random output distribution networks provided by this invention establishes the probability density functions of distributed power sources, loads, and carbon emission intensity, and calculates the mean and variance of carbon emissions for the entire system under different random input conditions. This lays a solid foundation for the comprehensive analysis and evaluation of carbon emissions under various operating conditions and meets the requirements for carbon flow calculation in distribution networks containing distributed power sources. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, obtaining other drawings based on these drawings without creative effort still falls within the scope of the present invention.

[0055] Figure 1This is a schematic diagram of the main process of a carbon emission calculation method for a distributed power generation random output distribution network according to an embodiment of the present invention.

[0056] Figure 2 This is a logical schematic diagram of a carbon emission calculation method for a distributed power generation random output distribution network according to an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings.

[0058] like Figure 1 and Figure 2 The diagram shown is a schematic representation of an embodiment of a carbon emission calculation method for a distributed power generation stochastic output distribution network provided by the present invention. In this embodiment, the method includes the following steps:

[0059] Obtain raw data of the distributed generation's random output distribution network. Raw parameters include generator voltage, phase angle, transmission line impedance per unit length, grounding capacitance, line length, transformer capacity, rated voltage on both primary and secondary sides, positive sequence leakage voltage and resistance, and tap position. All of this data can be obtained by consulting the nameplates of the corresponding equipment. The random variable types and their probability density functions or distribution laws of distributed generation output, load, and generator carbon emission intensity are obtained by fitting historical data or meteorological data from each device.

[0060] Furthermore, the raw data is input into a preset node admittance matrix to determine the distribution probability of the parameters of each node in the distribution network. Deterministic power flow calculation is then performed based on the determined distribution probability to obtain the randomly distributed generator carbon emission intensity. That is, the node admittance matrix of the system is calculated from the raw data. The node admittance matrix is ​​a fundamental concept in power systems and will not be elaborated upon here. Random numbers for distributed generation output, load, and generator carbon emission intensity are generated according to different types of random variables. Historical data and short-term forecast data for distributed generation output, load, and generator carbon emission intensity are obtained through historical data or meteorological data from each device. The historical data and short-term forecast data are fitted to obtain the type of each random variable and its probability density function or distribution probability. The randomly distributed injected power is obtained from the random numbers, and deterministic power flow calculation is performed to obtain the randomly distributed generator carbon emission intensity and power flow results.

[0061] In this embodiment, the original data is input into a preset node admittance matrix to determine the type of parameters at each node of the distribution network; a corresponding random number is calculated based on the type of each node parameter, and the distribution probability of each node parameter is determined based on the obtained random number; wherein, the type of each node parameter is at least a discrete random variable and a continuous random variable. The carbon flow algorithm based on power flow calculation does not consider system network losses; however, actual power grid line losses are significant and cannot be ignored. Therefore, it is necessary to consider network loss processing. Methods for processing network losses can transform the system from a lossy network to a lossless network.

[0062] Specifically, when the type of distribution network node parameters is discrete random variables, the probability distribution is calculated according to the following formula:

[0063] P{X=x k}=p k k = 1, 2, 3...

[0064]

[0065] Where, P{X=x k} represents the probability of X taking each possible value, k represents the sequence number of each possible value, and E(X) represents the random number corresponding to the random variable X; x k p represents the possible values ​​of a random variable. k This indicates the probability of taking this value.

[0066] When the type of distribution network node parameters is continuous random variable, the probability distribution is calculated according to the following formula:

[0067]

[0068]

[0069] Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters.

[0070] More specifically, when transforming a system from a lossy network to a lossless network, the relationship between the active power flow and the generator-injected power at each node of the system is first established. The active power flow vector at each node of the lossless network is then determined using the following formula:

[0071] A u ·P g =P G

[0072]

[0073] Among them, P g A represents the active power flow vector of each node after being converted into an equivalent lossless network. uRepresents an n x n matrix [A] u ] ij P G Indicates the injected power of the generator. Let c be the set of nodes that directly flow into the active power flow from node i. ij P is the power sharing factor. j-i Let P be the active power flowing from node j to node i. j Let be the active power flowing through node j.

[0074] Based on the system topology and generator input power, P can be calculated. j This allows us to calculate the active power flow of the branches and the equivalent load of the nodes in the equivalent lossless network. The equivalent active power flow of the branches and the equivalent active load of the nodes are calculated using the following formulas:

[0075]

[0076]

[0077] Among them, |P (g) i-j | represents the equivalent active power flow of branch ij, [A u -1 ] ik Let P represent an n x n matrix. i (g) P represents the active power of node i in the equivalent lossless network. GK P represents the power output of the equivalent generator node. i Let P be the active power flowing through node i. i-j Let P be the active power flowing from node i to node j, n represent the total number of nodes, k represent the equivalent generator node, and P' be the active power. Li This represents the actual load of node i per unit time. This represents the equivalent active load of node i. The equivalent active load includes the original load of the system and the portion allocated to network losses. Through the above steps, the system can be transformed from a lossy network to a lossless network. Then, using the relevant theories of carbon flow calculation in lossless networks, the nodal carbon potential, branch carbon flow density, and other related variables of the system can be calculated.

[0078] System carbon emissions are typically described using five variables: nodal carbon potential, branch carbon flow density, branch carbon flow rate, load carbon flow rate, and generator-injected carbon flow rate. Nodal carbon potential is fundamental to system carbon flow calculations. Based on the properties of carbon emission flows, when the carbon potential of a node is known, the carbon flow density on the branch from which active power flows out is the same as the node's carbon potential. Therefore, once the carbon potential of each node is determined, the carbon flow rate of each branch can be quickly calculated. Load carbon flow rate represents the carbon emissions directly generated by the load; generator-injected carbon flow rate represents the carbon emissions directly generated by the generator. The nodal carbon potential, branch carbon flow density, and branch carbon flow rate of the distribution network are calculated using the following formulas:

[0079]

[0080] R L =ρ L ·P L

[0081] ρ loadi =e Ni ·P load

[0082] ρ source =e G ·P source

[0083] Where, ρ s E represents the carbon flux density of branch s. Gi P represents the carbon emission intensity of voltage source i. BS P represents the active power injected into the branch node. Gi ρ represents the active power injected into node i by the voltage source. L P represents the branch carbon flux density. L e represents the active power flowing out of the branch node. Ni ρ represents the carbon potential at the node where the load is located. loadi P represents the carbon emissions directly generated by the load. load ρ represents the active power consumed by the load. source This indicates the amount of carbon emissions directly generated by the generator; e G Indicates the carbon emission intensity of the generator; P source This represents the active power injected by the generator. After calculating the node carbon potential, branch carbon flux density, and branch carbon flux rate, an accurate understanding of the distribution of carbon emission flows in the system can be obtained, allowing for appropriate measures to be taken in areas with severe carbon emissions.

[0084] Furthermore, the corresponding decision coefficient is calculated based on the randomly distributed generator carbon emission intensity. Depending on whether convergence is required, if convergence is required, the distribution probability of each node parameter is redefined and the deterministic power flow calculation is performed again until convergence is no longer necessary. If convergence is not required, the randomly distributed generator carbon emission intensity and power flow results are output as the final calculation results. In other words, the calculated carbon potential and branch carbon flow density are analyzed for their numerical characteristics, variance coefficients are calculated, and convergence is determined. If convergence is not achieved, the process returns to the previous step to continue the calculation.

[0085] In a specific embodiment, the types of nodal carbon potential, branch carbon flow density, and branch carbon flow rate of the calculated distribution network are determined;

[0086] When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a discrete random variable, the probability distribution is calculated according to the following formula:

[0087] P{X=x k}=p k k = 1, 2, 3...

[0088]

[0089] Where, P{X=x k} represents the probability of X taking each possible value, k represents the sequence number of each possible value, and E(X) represents the random number corresponding to the random variable X; x k p represents the possible values ​​of a random variable. k This indicates the probability of taking that value;

[0090] When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a continuous random variable, the probability distribution is calculated according to the following formula:

[0091]

[0092]

[0093] Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters.

[0094] Specifically, the corresponding judgment coefficient is calculated according to the following formula:

[0095] D(X)=E{[XE(X)] 2}

[0096] Where D(X) represents the judgment coefficient, and E{[XE(X)]} 2Let} represent the variance of random variable X, and E(X) represent the corresponding random number for random variable X. When the decision coefficient is less than the preset convergence precision, convergence is deemed unnecessary; when the decision coefficient is greater than or equal to the preset convergence precision (generally 0.05), convergence is deemed necessary. The criterion for judgment is: the maximum variance coefficient of the system is less than the convergence precision, generally 0.05. The formula for calculating the maximum variance coefficient is as follows:

[0097]

[0098]

[0099]

[0100] Where: η is the maximum variance coefficient; U is the failure probability of the system, x i Let x be a 0-1 indicator variable, obtained through Monte Carlo simulation: (Convention) i When x = 1, it indicates that the extraction was unsuccessful; by convention x i When = 0, it indicates that the extraction was successful, and V(U) is the variance of the sample mean.

[0101] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0102] The carbon emission calculation method for distributed power generation random output distribution networks provided by this invention establishes the probability density functions of distributed power sources, loads, and carbon emission intensity, and calculates the mean and variance of carbon emissions for the entire system under different random input conditions. This lays a solid foundation for the comprehensive analysis and evaluation of carbon emissions under various operating conditions and meets the requirements for carbon flow calculation in distribution networks containing distributed power sources.

[0103] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for carbon emission calculation of a distribution network with distributed generation stochastic output, characterized in that, include: Obtain raw data of the distributed generation random output distribution network; The raw data is input into a preset node admittance matrix to determine the distribution probability of the parameters of each node in the distribution network, and deterministic power flow calculation is performed based on the determined distribution probability to obtain the generator carbon emission intensity distributed randomly. The step of performing deterministic power flow calculations based on determined distribution probabilities to obtain the generator carbon emission intensity according to a random distribution specifically includes: The active power flow vector of each node in the lossless network is determined using the following formula: A u • P g = P G Among them, P g A represents the active power flow vector of each node after being converted into an equivalent lossless network. u Represents an n x n matrix [A] u ] ij P G Indicates the injected power of the generator. Let c be the set of nodes that directly flow into the active power flow from node i. ij P is the power sharing factor. j-i Let P be the active power flowing from node j to node i. j Let be the active power flowing through node j; The equivalent active power flow of the branch and the equivalent active power load of the node are calculated using the following formulas: Among them, |P (g) i-j | represents the equivalent active power flow of branch ij, [A u -1 ] ik Let P represent an n x n matrix. i (g) P represents the active power of node i in the equivalent lossless network. GK P represents the power output of the equivalent generator node. i Let P be the active power flowing through node i. i-j Let P be the active power flowing from node i to node j, n represent the total number of nodes, k represent the equivalent generator node, and P' be the active power. Ki This represents the actual load of node i per unit time. This represents the equivalent active load of node i; Calculate the nodal carbon potential, branch carbon flux density, and branch carbon flux rate of the distribution network using the following formulas: R L = p L · P L p loadi = e Ni · P load p source = e G · P source Where, ρ s E represents the carbon flux density of branch s. Gi P represents the carbon emission intensity of voltage source i. BS P represents the active power injected into the branch node. Gi ρ represents the active power injected into node i by the voltage source. L P represents the branch carbon flux density. L e represents the active power flowing out of the branch node. Ni ρ represents the carbon potential at the node where the load is located. loadi P represents the carbon emissions directly generated by the load. load ρ represents the active power consumed by the load. source This indicates the amount of carbon emissions directly generated by the generator; e G Indicates the carbon emission intensity of the generator; P source This indicates the active power injected into the generator; The judgment coefficient is calculated based on the randomly distributed carbon emission intensity of the generators; the calculation of the judgment coefficient specifically includes: Determine the types of nodal carbon potential, branch carbon flux density, and branch carbon flux rate of the calculated distribution network. When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a discrete random variable, the probability distribution is calculated according to the following formula: P{X=x k} = p k , k = 1, 2, 3… where P{X=x k} denotes the probability of X taking each possible value, k denotes the sequence number of each possible value, and E(X) denotes the corresponding random number of the random variable X; x k denotes the value of the random variable, and p k denotes the probability of the value. When the node carbon potential, branch carbon flow density, or branch carbon flow rate of the distribution network is a continuous random variable, the probability distribution is calculated according to the following formula: Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters; Calculate the corresponding judgment coefficient using the following formula: D(X) = E{[X - E(X)] 2} where D(X) represents a judgment coefficient, E{X-E(X)} 2 represents a variance of the random variable X, and E(X) represents a corresponding random number of the random variable X. Based on whether the coefficients need to converge, if convergence is required, the distribution probability of each node parameter is redefined and the deterministic power flow calculation is performed again until convergence is no longer required; if convergence is not required, the generator carbon emission intensity and power flow results obtained by random distribution are output as the final calculation results.

2. The method of claim 1, wherein, The determination of the distribution probability of parameters at each node of the distribution network specifically includes: The raw data is input into a preset node admittance matrix to determine the type of parameters of each node in the distribution network; Calculate the corresponding random number based on the type of parameters of each node in the distribution network, and determine the distribution probability of each node parameter based on the obtained random number; The types of the parameters of each node are at least discrete random variables and continuous random variables.

3. The method of claim 2, wherein, When the type of distribution network node parameters is discrete random variables, the probability distribution is calculated according to the following formula: P{X=x k} = p k , k = 1, 2, 3... Where, P{X=x k } represents the probability of X taking each possible value, k represents the sequence number of each possible value, and E(X) represents the random number corresponding to the random variable X; x k p represents the possible values ​​of a random variable. k This indicates the probability of taking this value.

4. The method of claim 2, wherein, When the type of distribution network node parameters is continuous random variable, the probability distribution is calculated according to the following formula: Where E(X) represents the random number corresponding to the random variable X, f(x) represents the probability density function of the random variable, and μ and σ represent the normal distribution parameters.

5. The method of claim 4, wherein, The step of determining whether the coefficients need to converge specifically includes: When the judgment coefficient is less than the preset convergence precision, it is determined that convergence is not required; When the judgment coefficient is greater than or equal to the preset convergence precision, convergence is determined.

Images