Regional and county-level power industry carbon emission prediction method based on electric carbon traceability

Abstract

The invention relates to a district and county level power industry carbon emission prediction method based on power carbon traceability. The method comprises the following steps: carrying out user side load prediction in different areas and time periods; carrying out distributed power supply output prediction in different regions and different time periods; performing energy storage charging and discharging scheme prediction and source side carbon emission factor prediction; carrying out load flow calculation on each node in different regions and time periods; constructing a power matrix; measuring and calculating a single-time-point carbon emission index; and measuring and calculating the long-time scale carbon emission. According to the method, prediction data of user power utilization, clean energy power generation and energy storage charging and discharging are integrated, the carbon emission measurement and calculation requirements of different time periods and regions can be met, the applicability and flexibility of the model are enhanced, accurate prediction of the carbon emission in the whole power distribution network load storage link can be achieved, and the accuracy of carbon emission measurement and calculation is improved; the method particularly focuses on the influence of policy changes of an electric vehicle, a distributed power supply and the like on the carbon emission, so that the prediction result can reflect actual changes brought by the policy changes.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission analysis of power systems, and in particular to a method for predicting carbon emissions in the power industry at the county level based on electricity carbon tracing. Background Art

[0002] In order to solve resource and environmental problems and alleviate the pressure of energy demand, the power industry is actively promoting the development of clean energy and energy conservation and emission reduction. In this context, carbon emission calculation, as an important basis for the study of low-carbon operation of power systems, has attracted widespread attention and in-depth research from scholars at home and abroad. At present, there are many indirect carbon emission calculation methods in the power system, including but not limited to the average emission factor method of the power grid, sensitivity analysis method, cooperative game method, pricing mechanism method and carbon emission flow tracking method. Among them, the carbon emission tracking method based on the theory of electric carbon flow has attracted attention due to its advantages in the division of carbon emission responsibilities and analysis of regions and time periods. However, this method is based on historical data and judges the carbon emission level of the region in historical years. It cannot accurately measure the carbon emission level in future years and cannot provide clear guidance for future power grid carbon emissions.

[0003] With the continuous enrichment of distributed clean energy types in the power grid and the increase in the access ratio, the calculation of indirect carbon emissions at the county-level power grid has become more complicated. Therefore, it is particularly important to accurately predict carbon emissions by using the carbon emission tracking method based on the electric carbon flow theory. Accurately obtaining the carbon emission factors of regional power sources and various load flow data at different time points is the key to achieving this goal. Based on the power supply and use entities at the county level, it is particularly necessary to further explore the carbon emission prediction method for the power industry at the county level. Summary of the invention

[0004] In order to solve the problem of lack of prediction methods for carbon emissions of district and county distribution networks in future years, the purpose of the present invention is to provide a district and county power industry carbon emissions prediction method based on electricity and carbon traceability that can realize accurate prediction of carbon emissions in all links of distribution network source, network, load and storage, improve the accuracy of carbon emissions calculation, meet the development needs of high-proportion distributed power source access in new power distribution and utilization systems, and enhance the ability to control carbon emissions of new power distribution and utilization systems.

[0005] To achieve the above purpose, the present invention adopts the following technical solution: a method for predicting carbon emissions in the county-level power industry based on electric carbon tracing, the method comprising the following steps in order:

[0006] (1) Carry out user-side load forecasting by region and time period, including the load forecasting of traditional users at each time point and the load forecasting of new users at each time point, and obtain the load forecasting curve of each user at each time point in the region;

[0007] (2) Carry out distributed power output forecasts by region and time period, including forecasts for photovoltaic output, wind power output, and hydropower output, and obtain output forecast curves for each power source in the region at each time point;

[0008] (3) Predict the energy storage charging and discharging plan and the source-side carbon emission factor to obtain the energy storage status data at different time points and the dynamic carbon emission factor of the main grid source side for calculation;

[0009] (4) Based on the load forecast curves of each user at each time point in the region, the output forecast curves of each power source at each time point in the region, as well as the energy storage status data at different time points used for measurement and the dynamic carbon emission factor on the main grid source side, the power flow calculation of each node is performed in different areas and time periods to obtain the active injection power at node i in time period t;

[0010] (5) Constructing a power matrix;

[0011] (6) Calculate the carbon emission index at a single time point based on the active injected power at node i in period t and the power matrix;

[0012] (7) Calculate carbon emissions over a long period of time based on carbon emission indicators at a single point in time.

[0013] The step (1) specifically comprises the following steps in order:

[0014] (1a) Load forecasting for traditional users at different time points: Traditional users are divided into three user groups, and forecasts are made for each user group:

[0015] The first category is the existing users whose electricity consumption scale has not changed: during the forecast period, the electricity consumption of this type of users has not changed significantly, and the forecast is based on historical electricity consumption data and current electricity consumption patterns;

[0016] The second category is the expansion of existing users: For users who plan to expand, the annual maximum load is obtained by analogy based on the power consumption scale after expansion, and the data at each time point is obtained by using the historical per-unit value curve;

[0017] The third category is the newly added traditional type of users: according to the user's commissioning time and reported capacity, the historical electricity consumption data of typical users in the same industry and the industry's average electricity consumption level are used to predict the annual maximum load, and the data at each time point is obtained through the historical per unit value curve;

[0018] (1b) Load forecast for new users at different time points:

[0019] New loads include data centers and electric vehicles. Based on the study of the composition of data center loads, the parameter boundary conditions are determined according to the actual operation of the cloud base data center. According to the characteristics of data center installation and commissioning in batches, the equipment load is determined by considering the number of newly added racks and the racking rate. The comprehensive resource utilization rate is used to determine the annual 8760-hour forecast of the data center.

[0020] The replacement ratio method is used to calculate the changes in the number of electric vehicles in use. Combining the actual growth of electric vehicles with the government's electric vehicle renewal and replacement ratio planning targets, we obtain an incremental forecast for each type of electric vehicle in each year.

[0021] The step (2) specifically comprises the following steps in order:

[0022] (2a) Photovoltaic output forecast, including:

[0023] (2a1) Processing historical output data, dividing each season into 16 output scenarios according to spring, summer, autumn and winter, including sunny days, cloudy days, rainy days and heavy rain days, and generating typical output curves;

[0024] (2a2) based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve of each season, combined with the distributed photovoltaic installed capacity, the distributed photovoltaic installed capacity includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the photovoltaic output forecast curve;

[0025] (2b) Wind power output forecast, including:

[0026] (2b1) The historical output data is divided into 8 typical output scenarios according to spring, summer, autumn and winter, and each season is divided into high wind days and low wind days to generate a typical output curve; the high wind day refers to a wind force reaching or exceeding level 8, and the low wind day refers to a wind force less than level 8;

[0027] (2b2) ​​based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve for each season, combined with the installed capacity of distributed wind power, which includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the long-term wind power output forecast curve;

[0028] (3) Hydropower output forecast: According to the historical output situation in the same period, the hydropower output forecast curve is obtained.

[0029] The step (3) specifically comprises the following steps in order:

[0030] (3a) If there is no new capacity for energy storage, the charging and discharging plan for the same period in history shall be followed;

[0031] (3b) If there is new energy storage capacity, the historical charge and discharge curves are standardized, and the predicted charge and discharge curves are obtained according to the long-term capacity and the standardized charge and discharge curves;

[0032] (3c) The carbon emission factor brought by the electricity obtained from the large grid or main grid directly adopts the average carbon emission factor of the regional power grid.

[0033] The step (4) specifically refers to: obtaining the distribution transformer power consumption data, the output of the distributed power source and the grid-side energy storage operation data at multiple time points of the distribution network, taking the feeder group and the power supply unit as the measurement range and the feeder group as the partition, performing the power flow calculation, and the active power and reactive power balance equation at the node i in the time period t is expressed as:

[0034]

[0035] Where: P i , Q i are the active injection power and reactive injection power of node i respectively; V i is the voltage amplitude of node i; V j is the voltage amplitude at node j; θ ij is the voltage phase difference at both ends of line ij; G ij , B ij are the real and imaginary parts of row i and column j in the node admittance matrix respectively; B i is the set of end nodes of all branches connected to node i.

[0036] The step (5) specifically means that the power matrix includes a node active flux matrix, a flow distribution matrix, a load distribution matrix and a generator output matrix, and the formula is:

[0037]

[0038] P B,t =(P B,ij,t ) N×N

[0039] P L,t =(P L,ii,t ) N×N

[0040]

[0041] Where: P N,t is the active flux matrix of the node in period t. The matrix is ​​an N-order diagonal matrix. The diagonal term P in the matrix N,ii,t is the active flux of node i; P B,t is the power flow distribution matrix of the line in period t, which is used to describe the active power flow distribution of the power system. All non-diagonal elements P of this matrix B,ij,tis the flow between branches ij, P B,ii,t =0;P L,t is the load distribution matrix, which is used to describe the load distribution of the power system. The diagonal elements P L,ii,t is the active power flow of load node i; P G,t is the generator output matrix for period t, and its matrix element P G,ki,t is defined as: when the kth generator is connected to node i and the active power flow injected into node i during period t is p, then P G,ki,t =p, otherwise P G,ki,t =0; J is the topological distribution matrix of the generator set, which is used to characterize the distribution and access to the system of the generator set. It is specifically defined as: J = (J ki ) K×N , if the kth unit is connected to the i-th node k of the system, k=1,2,...,K; i=1,2,...,N, then J ki =1, otherwise J ki =0;P B,s,t is the power flow data of branch s in the time period; is the total number of branches in the same direction, N is the total number of nodes in the district and county power grid, K is the total number of nodes including generators, P G,k,t is the power generation of power plant k in period t.

[0042] The step (6) specifically comprises the following steps in order:

[0043] (6a) Load side

[0044] The load side searches for the power source one by one from the user end, and calculates the node carbon potential of the load end according to the output power proportion through the difference of carbon emission factors of each power source. The expression of the system node carbon emission factor matrix is:

[0045]

[0046] Where: U is the set of nodes within the coverage of the regional power grid n; P L,j,t is the load power of node j in time period t; e C,j,t is the carbon emission factor of the node;

[0047] (6b) Network side

[0048] The grid side traces the equivalent carbon emissions of the power plant supplying the substation, and calculates the carbon potential of each node on the power grid line by building a matrix model; since the carbon flow density of each line flowing out of the same node is equal, according to the node electricity carbon emission factor matrix, the line carbon flow density matrix is ​​calculated by the following formula:

[0049] ρ B,t =diag(E C,t )H

[0050] Where: B,t is the line carbon flow density matrix of time period t, which is used to describe the carbon flow density of each line in the system; H is the matrix representing the flow direction relationship of the branch flow, H = (H ij ) N×N , i, j = 1, 2, ..., N, if there is a branch connecting node i and node j, and the positive active power flow flowing into node j through this branch is not zero, then the element H ij =1,H ji =0; if the active power flow through the branch is reverse, that is, the current flows from the load to the substation, then H ij =0,H ji =1; N is the total number of nodes in the district and county power grid, E C,t is the node carbon flow rate;

[0051] According to the line network loss data, the line network loss distribution matrix is ​​defined as:

[0052] P BL,t =(P BL,ij,t ) N×N

[0053] Where: P BL,t is the line network loss distribution matrix in time period t, which is used to describe the network loss distribution of the power system; P BL,ij,t is the matrix P BL,t The non-diagonal elements of represent the network loss of the power flow between branches ij, i≠j; the diagonal elements P BL,ii,t is 0;

[0054] According to the line power flow distribution matrix, network loss distribution matrix and line carbon flow density matrix, the line carbon flow rate and network loss carbon flow rate are calculated:

[0055] R B,t =ρ B,t P B,t

[0056] R BL,t =ρ B,t P BL,t

[0057] In the formula, R B,t is the line carbon flow rate in period t; R BL,t is the network loss carbon flow rate in time period t, P BL,t is the line network loss distribution matrix of time period t, which is used to describe the network loss distribution of the power system; P B,t is the power flow distribution matrix of the line in period t;

[0058] (6c) Storage side: In the charging state, the energy storage element is regarded as a load, and in the discharging process, the energy storage element is regarded as a distributed unit.

[0059] The step (7) specifically comprises the following steps in order:

[0060] (7a) Divide the 365-day average load into intervals. For data segments with relatively concentrated loads, adopt a detailed grading method to ensure that the number of loads in each grading is balanced and the annual proportion of each grading is obtained; the two feeders in the same partition are required to have the same number of loads;

[0061] According to the definition of network loss carbon flow rate, the indirect carbon emissions of line network loss are calculated by the following formula:

[0062]

[0063] In the formula, is the indirect carbon emission of line network loss, R BL,t is the network loss carbon flow rate in time period t, Δt is the time period;

[0064] According to the regional carbon emission factor, the indirect carbon emissions from electricity consumption of users in different periods of time are quantified by the following formula:

[0065] F C,m,t =e′ C,n,t P L,m,t Δt

[0066] Where: F C,m,t is the indirect carbon emissions from electricity consumption of user m in period t; L,m,t is the power consumption of user m in time period t, and user m is located in regional power grid n; e′ C,n,t is the regional carbon emission factor;

[0067] (7b) Calculate the load flow of the graded load based on the feeder group to obtain the corresponding carbon emission data;

[0068] (7c) Based on the proportion of each category and the carbon emissions of each category, the probability weighted method is used to obtain the annual carbon emissions.

[0069] It can be seen from the above technical scheme that the beneficial effects of the present invention are: first, the present invention integrates the forecast data of user electricity consumption, clean energy power generation and energy storage charging and discharging, can adapt to the carbon emission measurement needs of different time periods and regions, enhances the applicability and flexibility of the model, and can realize the accurate forecast of carbon emissions in all links of distribution network source, network, load and storage, and improves the accuracy of carbon emission measurement; second, the present invention pays special attention to the impact of policy changes such as electric vehicles and distributed power sources on carbon emissions, so that the forecast results can reflect the actual changes brought about by policy changes; third, the present invention takes into account the output differences and new scales at different points in time, which is very important for accurate Predicting carbon emissions is crucial to ensure the real-time and dynamic nature of the model; fourth, the present invention calculates power flow based on the grid structure of the future distribution network, which is a key step in determining the active power of each node and provides a basis for calculating carbon emissions; fifth, the present invention can comprehensively and accurately predict the carbon emissions of the distribution network, providing a scientific basis for related decision-making, and enhancing the scientific nature and effectiveness of the decision-making; sixth, the present invention helps to understand the distribution and sources of carbon emissions in different regions, and provides a scientific basis for relevant departments to formulate carbon emission reduction measures, promote carbon emission reduction in new power distribution systems, and maximize the carbon emission reduction and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 is a flow chart of the method of the present invention;

[0071] Figure 2 This is the single-line diagram of the power supply line from the Customs Road Substation to Tianzhi Road;

[0072] Figure 3 This is a schematic diagram of the carbon emission index calculation results of the medium-voltage feeder group powered by Tianzhi Road at a certain hour on a certain day. DETAILED DESCRIPTION

[0073] like Figure 1 As shown, a method for predicting carbon emissions in the power industry at the county level based on electric carbon tracing includes the following steps in order:

[0074] (1) Carry out user-side load forecasting by region and time period, including the load forecasting of traditional users at each time point and the load forecasting of new users at each time point, and obtain the load forecasting curve of each user at each time point in the region;

[0075] (2) Carry out distributed power output forecasts by region and time period, including forecasts for photovoltaic output, wind power output, and hydropower output, and obtain output forecast curves for each power source in the region at each time point;

[0076] (3) Predict the energy storage charging and discharging plan and the source-side carbon emission factor to obtain the energy storage status data at different time points and the dynamic carbon emission factor of the main grid source side for calculation;

[0077] (4) Based on the load forecast curves of each user at each time point in the region, the output forecast curves of each power source at each time point in the region, as well as the energy storage status data at different time points used for measurement and the dynamic carbon emission factor on the main grid source side, the power flow calculation of each node is performed in different areas and time periods to obtain the active injection power at node i in time period t;

[0078] (5) Constructing a power matrix;

[0079] (6) Calculate the carbon emission index at a single time point based on the active injected power at node i in period t and the power matrix;

[0080] (7) Calculate carbon emissions over a long period of time based on carbon emission indicators at a single point in time.

[0081] The step (1) specifically comprises the following steps in order:

[0082] (1a) Load forecasting for traditional users at different time points: Traditional users are divided into three user groups, and forecasts are made for each user group:

[0083] The first category is the existing users whose electricity consumption scale has not changed: during the forecast period, the electricity consumption of this type of users has not changed significantly, and the forecast is based on historical electricity consumption data and current electricity consumption patterns;

[0084] The second category is the expansion of existing users: For users who plan to expand, the annual maximum load is obtained by analogy based on the power consumption scale after expansion, and the data at each time point is obtained by using the historical per-unit value curve;

[0085] The third category is the newly added traditional type of users: according to the user's commissioning time and reported capacity, the historical electricity consumption data of typical users in the same industry and the industry's average electricity consumption level are used to predict the annual maximum load, and the data at each time point is obtained through the historical per unit value curve;

[0086] (1b) Load forecast for new users at different time points:

[0087] New loads include data centers and electric vehicles. Based on the study of the composition of data center loads, the parameter boundary conditions are determined according to the actual operation of the cloud base data center. According to the characteristics of data center installation and commissioning in batches, the equipment load is determined by considering the number of newly added racks and the racking rate. The comprehensive resource utilization rate is used to determine the annual 8760-hour forecast of the data center.

[0088] The replacement ratio method is used to calculate the changes in the number of electric vehicles in use. Combining the actual growth of electric vehicles with the government's electric vehicle renewal and replacement ratio planning targets, we obtain an incremental forecast for each type of electric vehicle in each year.

[0089] The step (2) specifically comprises the following steps in order:

[0090] (2a) Photovoltaic output forecast, including:

[0091] (2a1) Processing historical output data, dividing each season into 16 output scenarios according to spring, summer, autumn and winter, including sunny days, cloudy days, rainy days and heavy rain days, and generating typical output curves;

[0092] (2a2) based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve of each season, combined with the distributed photovoltaic installed capacity, the distributed photovoltaic installed capacity includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the photovoltaic output forecast curve;

[0093] (2b) Wind power output forecast, including:

[0094] (2b1) The historical output data is divided into 8 typical output scenarios according to spring, summer, autumn and winter, and each season is divided into high wind days and low wind days to generate a typical output curve; the high wind day refers to a wind force reaching or exceeding level 8, and the low wind day refers to a wind force less than level 8;

[0095] (2b2) ​​based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve for each season, combined with the installed capacity of distributed wind power, which includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the long-term wind power output forecast curve;

[0096] (3) Hydropower output forecast: According to the historical output situation in the same period, the hydropower output forecast curve is obtained.

[0097] The step (3) specifically comprises the following steps in order:

[0098] (3a) If there is no new capacity for energy storage, the charging and discharging plan for the same period in history shall be followed;

[0099] (3b) If there is new energy storage capacity, the historical charge and discharge curves are standardized, and the predicted charge and discharge curves are obtained according to the long-term capacity and the standardized charge and discharge curves;

[0100] (3c) The carbon emission factor brought by the electricity obtained from the large grid or main grid directly adopts the average carbon emission factor of the regional power grid.

[0101] The step (4) specifically refers to: obtaining the distribution transformer power consumption data, the output of the distributed power source and the grid-side energy storage operation data at multiple time points of the distribution network, taking the feeder group and the power supply unit as the measurement range and the feeder group as the partition, performing the power flow calculation, and the active power and reactive power balance equation at the node i in the time period t is expressed as:

[0102]

[0103] Where: P i , Q i are the active injection power and reactive injection power of node i respectively; V i is the voltage amplitude of node i; V j is the voltage amplitude at node j; θ ij is the voltage phase difference at both ends of line ij; G ij , B ij are the real and imaginary parts of row i and column j in the node admittance matrix respectively; B i is the set of end nodes of all branches connected to node i.

[0104] The step (5) specifically means that the power matrix includes a node active flux matrix, a flow distribution matrix, a load distribution matrix and a generator output matrix, and the formula is:

[0105]

[0106] P B,t =(P B,ij,t ) N×N

[0107] P L,t =(P L,ii,t ) N×N

[0108]

[0109] Where: P N,t is the active flux matrix of the node in period t. The matrix is ​​an N-order diagonal matrix. The diagonal term P in the matrix N,ii,t is the active flux of node i; P B,t is the power flow distribution matrix of the line in period t, which is used to describe the active power flow distribution of the power system. All non-diagonal elements P of this matrix B,ij,t is the flow between branches ij, P B,ii,t =0;P L,t is the load distribution matrix, which is used to describe the load distribution of the power system. The diagonal elements P L,ii,t is the active power flow of load node i; P G,t is the generator output matrix for period t, and its matrix element P G,ki,t is defined as: when the kth generator is connected to node i and the active power flow injected into node i during period t is p, then P G,ki,t =p, otherwise P G,ki,t =0; J is the topological distribution matrix of the generator set, which is used to characterize the distribution and access to the system of the generator set. It is specifically defined as: J = (J ki ) K×N, if the kth unit is connected to the i-th node k of the system, k=1,2,...,K; i=1,2,...,N, then J ki =1, otherwise J ki =0;P B,s,t is the power flow data of branch s in the time period; is the total number of branches in the same direction, N is the total number of nodes in the district and county power grid, K is the total number of nodes including generators, P G,k,t is the power generation of power plant k in period t.

[0110] The step (6) specifically comprises the following steps in order:

[0111] (6a) Load side

[0112] The load side searches for the power source one by one from the user end, and calculates the node carbon potential of the load end according to the output power proportion through the difference of carbon emission factors of each power source. The expression of the system node carbon emission factor matrix is:

[0113]

[0114] Where: U is the set of nodes within the coverage of the regional power grid n; P L,j,t is the load power of node j in time period t; e C,j,t is the carbon emission factor of the node;

[0115] (6b) Network side

[0116] The grid side traces the equivalent carbon emissions of the power plant supplying the substation, and calculates the carbon potential of each node on the power grid line by building a matrix model; since the carbon flow density of each line flowing out of the same node is equal, according to the node electricity carbon emission factor matrix, the line carbon flow density matrix is ​​calculated by the following formula:

[0117] ρ B,t =diag(E C,t )H

[0118] Where: B,t is the line carbon flow density matrix of time period t, which is used to describe the carbon flow density of each line in the system; H is the matrix representing the flow direction relationship of the branch flow, H = (H ij ) N×N , i, j = 1, 2, ..., N, if there is a branch connecting node i and node j, and the positive active power flow flowing into node j through this branch is not zero, then the element H ij =1,H ji =0; if the active power flow through the branch is reverse, that is, the current flows from the load to the substation, then H ij =0,H ji =1; N is the total number of nodes in the district and county power grid, EC,t is the node carbon flow rate;

[0119] According to the line network loss data, the line network loss distribution matrix is ​​defined as:

[0120] P BL,t =(P BL,ij,t ) N×N

[0121] Where: P BL,t is the line network loss distribution matrix in time period t, which is used to describe the network loss distribution of the power system; P BL,ij,t is the matrix P BL,t The non-diagonal elements of represent the network loss of the power flow between branches ij, i≠j; the diagonal elements P BL,ii,t is 0;

[0122] According to the line power flow distribution matrix, network loss distribution matrix and line carbon flow density matrix, the line carbon flow rate and network loss carbon flow rate are calculated:

[0123] R B,t =ρ B,t P B,t

[0124] R BL,t =ρ B,t P BL,t

[0125] In the formula, R B,t is the line carbon flow rate in period t; R BL,t is the network loss carbon flow rate in time period t, P BL,t is the line network loss distribution matrix of time period t, which is used to describe the network loss distribution of the power system; P B,t is the power flow distribution matrix of the line in period t;

[0126] (6c) Storage side: In the charging state, the energy storage element is regarded as a load, and in the discharging process, the energy storage element is regarded as a distributed unit.

[0127] The step (7) specifically comprises the following steps in order:

[0128] (7a) Divide the 365-day average load into intervals. For data segments with relatively concentrated loads, adopt a detailed grading method to ensure that the number of loads in each grading is balanced and the annual proportion of each grading is obtained; the two feeders in the same partition are required to have the same number of loads;

[0129] According to the definition of network loss carbon flow rate, the indirect carbon emissions of line network loss are calculated by the following formula:

[0130]

[0131] In the formula, is the indirect carbon emission of line network loss, R BL,t is the network loss carbon flow rate in time period t, Δt is the time period;

[0132] According to the regional carbon emission factor, the indirect carbon emissions from electricity consumption of users in different periods of time are quantified by the following formula:

[0133] F C,m,t =e′ C,n,t P L,m,t Δt

[0134] Where: F C,m,t is the indirect carbon emissions from electricity consumption of user m in period t; L,m,t is the power consumption of user m in time period t, and user m is located in regional power grid n; e′ C,n,t is the regional carbon emission factor;

[0135] (7b) Calculate the load flow of the graded load based on the feeder group to obtain the corresponding carbon emission data;

[0136] (7c) Based on the proportion of each category and the carbon emissions of each category, the probability weighted method is used to obtain the annual carbon emissions.

[0137] Embodiment 1

[0138] like Figure 2 As shown, the present invention calculates the annual carbon emissions of Tianzhi Road Station 204-Tianzhi Road Station 21, and the specific implementation process is as follows:

[0139] The first step is to forecast the user side load by region and time period.

[0140] The load of this line is for traditional users, all of whom are commercial users. After communication and confirmation, there will be no changes in users in the region in the coming years, so only the load change patterns of traditional user types need to be considered.

[0141] Step 2 and Step 3: Distributed power output forecast, energy storage charging and discharging plan forecast, and source-side carbon emission factor forecast by region and time period:

[0142] There is no distributed power generation and no energy storage. Since there is no new energy construction, the source-side carbon emission factor remains unchanged.

[0143] Step 4 and Step 5: Calculate the power flow of each node in different time periods and construct the power matrix:

[0144] The distribution transformer power consumption data, distributed generation output and grid-side energy storage operation data at multiple time points of the distribution network are obtained, and the feeder group and power supply unit are used as the measurement range, and the feeder group is used as the partition to perform power flow calculation. The node active flux matrix, branch carbon flow density matrix and load distribution matrix are shown in Table 1, Table 2 and Table 3 respectively.

[0145] Table 1

[0146]

[0147] Table 2

[0148]

[0149] Table 3

[0150]

[0151] Steps 6 and 7: Calculate carbon emissions over a long period of time:

[0152] A single point in time diagram Figure 3 As shown in Table 4, the carbon emission results of Tianzhi Road Station 204-Tianzhi Road Station 21 in 2030 are shown in Table 4:

[0153] Table 4

[0154]

[0155]

[0156] In summary, the present invention integrates the forecast data of user electricity consumption, clean energy power generation and energy storage charging and discharging, can adapt to the carbon emission measurement needs of different time periods and regions, enhances the applicability and flexibility of the model, can realize the accurate prediction of carbon emissions in all links of distribution network source, network, load and storage, and improves the accuracy of carbon emission measurement; the present invention pays special attention to the impact of policy changes such as electric vehicles and distributed power sources on carbon emissions, so that the prediction results can reflect the actual changes brought about by policy changes; the present invention takes into account the output differences and new scales at different time points, which are crucial for accurately predicting carbon emissions and ensure the real-time and dynamic nature of the model; the present invention performs power flow calculation based on the grid structure of the future distribution network, which is a key step in determining the active power of each node and provides a basis for the calculation of carbon emissions; the present invention can comprehensively and accurately predict the carbon emissions of the distribution network, provide a scientific basis for relevant decision-making, and enhance the scientific nature and effectiveness of decision-making; the present invention helps to understand the distribution and source of carbon emissions in different regions, provide a scientific basis for relevant departments to formulate carbon emission reduction measures, promote carbon emission reduction of new distribution and power systems, and maximize the realization of carbon emission reduction benefits and economic benefits.

Claims

1. A method for predicting carbon emissions in the power industry at the county level based on electric carbon traceability, characterized by: The method comprises the following steps in order: (1) Carry out user-side load forecasting by region and time period, including the load forecasting of traditional users at each time point and the load forecasting of new users at each time point, and obtain the load forecasting curve of each user at each time point in the region; (2) Carry out distributed power output forecasts by region and time period, including forecasts for photovoltaic output, wind power output, and hydropower output, and obtain output forecast curves for each power source in the region at each time point; (3) Predict the energy storage charging and discharging plan and the source-side carbon emission factor to obtain the energy storage status data at different time points and the dynamic carbon emission factor of the main grid source side for calculation; (4) Based on the load forecast curves of each user at each time point in the region, the output forecast curves of each power source at each time point in the region, as well as the energy storage status data at different time points used for measurement and the dynamic carbon emission factor on the main grid source side, the power flow calculation of each node is performed in different areas and time periods to obtain the active injection power at node i in time period t; (5) Constructing a power matrix; (6) Calculate the carbon emission index at a single time point based on the active injected power at node i in period t and the power matrix; (7) Calculate carbon emissions over a long period of time based on carbon emission indicators at a single point in time.

2. The district and county-level electric power industry carbon emission prediction method based on electric carbon traceability according to claim 1 is characterized by: The step (1) specifically comprises the following steps in order: (1a) Load forecasting for traditional users at different time points: Traditional users are divided into three user groups, and forecasts are made for each user group: The first category is the existing users whose electricity consumption scale has not changed: during the forecast period, the electricity consumption of this type of users has not changed significantly, and the forecast is based on historical electricity consumption data and current electricity consumption patterns; The second category is the expansion of existing users: For users who plan to expand, the annual maximum load is obtained by analogy based on the power consumption scale after expansion, and the data at each time point is obtained by using the historical per-unit value curve; The third category is the newly added traditional type of users: according to the user's commissioning time and reported capacity, the historical electricity consumption data of typical users in the same industry and the industry's average electricity consumption level are used to predict the annual maximum load, and the data at each time point is obtained through the historical per unit value curve; (1b) Load forecast for new users at different time points: New loads include data centers and electric vehicles. Based on the study of the composition of data center loads, the parameter boundary conditions are determined according to the actual operation of the cloud base data center. According to the characteristics of data center installation and commissioning in batches, the equipment load is determined by considering the number of newly added racks and the racking rate. The comprehensive resource utilization rate is used to determine the annual 8760-hour forecast of the data center. The replacement ratio method is used to calculate the changes in the number of electric vehicles in use. Combining the actual growth of electric vehicles with the government's electric vehicle renewal and replacement ratio planning targets, we obtain an incremental forecast for each type of electric vehicle in each year.

3. The method for predicting carbon emissions in the district and county-level power industry based on electric carbon traceability according to claim 1 is characterized by: The step (2) specifically comprises the following steps in order: (2a) Photovoltaic output forecast, including: (2a1) Processing historical output data, dividing each season into 16 output scenarios according to spring, summer, autumn and winter, including sunny days, cloudy days, rainy days and heavy rain days, and generating typical output curves; (2a2) based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve of each season, combined with the distributed photovoltaic installed capacity, the distributed photovoltaic installed capacity includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the photovoltaic output forecast curve; (2b) Wind power output forecast, including: (2b1) The historical output data is divided into 8 typical output scenarios according to spring, summer, autumn and winter, and each season is divided into high wind days and low wind days to generate a typical output curve; the high wind day refers to a wind force reaching or exceeding level 8, and the low wind day refers to a wind force less than level 8; (2b2) ​​based on the probability of occurrence of various weather conditions in each season in history, using the weighted average method to obtain the typical output curve for each season, combined with the installed capacity of distributed wind power, which includes the capacity already put into operation and the capacity planned to be constructed in that year, to determine the long-term wind power output forecast curve; (3) Hydropower output forecast: According to the historical output situation in the same period, the hydropower output forecast curve is obtained.

4. The district and county-level power industry carbon emission prediction method based on electric carbon traceability according to claim 1 is characterized by: The step (3) specifically comprises the following steps in order: (3a) If there is no new capacity for energy storage, the charging and discharging plan for the same period in history shall be followed; (3b) If there is new energy storage capacity, the historical charge and discharge curves are standardized, and the predicted charge and discharge curves are obtained according to the long-term capacity and the standardized charge and discharge curves; (3c) The carbon emission factor brought by the electricity obtained from the large grid or main grid directly adopts the average carbon emission factor of the regional power grid.

5. The method for predicting carbon emissions in the county-level power industry based on electric carbon traceability according to claim 1 is characterized by: The step (4) specifically refers to: obtaining the distribution transformer power consumption data, the output of the distributed power source and the grid-side energy storage operation data at multiple time points of the distribution network, taking the feeder group and the power supply unit as the measurement range and the feeder group as the partition, performing the power flow calculation, and the active power and reactive power balance equation at the node i in the time period t is expressed as: Where: P i , Q i are the active injection power and reactive injection power of node i respectively; V i is the voltage amplitude of node i; V j is the voltage amplitude at node j; θ ij is the voltage phase difference at both ends of line ij; G ij , B ij are the real and imaginary parts of row i and column j in the node admittance matrix respectively; B i is the set of end nodes of all branches connected to node i.

6. The district and county-level power industry carbon emission prediction method based on electric carbon traceability according to claim 1 is characterized by: The step (5) specifically means that the power matrix includes a node active flux matrix, a flow distribution matrix, a load distribution matrix and a generator output matrix, and the formula is: P B,t =(P B,ij,t ) N×N P L,t =(P L,ii,t ) N×N Where: P N,t is the active flux matrix of the node in period t. The matrix is ​​an N-order diagonal matrix. The diagonal term P in the matrix N,ii,t is the active flux of node i; P B,t is the power flow distribution matrix of the line in period t, which is used to describe the active power flow distribution of the power system. All non-diagonal elements P of this matrix B,ij,t is the flow between branches ij, P B,ii,t =0;P L,t is the load distribution matrix, which is used to describe the load distribution of the power system. The diagonal elements P L,ii,t is the active power flow of load node i; P G,t is the generator output matrix for period t, and its matrix element P G,ki,t is defined as: when the kth generator is connected to node i and the active power flow injected into node i during period t is p, then P G,ki,t =p, otherwise P G,ki,t =0; J is the topological distribution matrix of the generator set, which is used to characterize the distribution and access to the system of the generator set. It is specifically defined as: J = (J ki ) K×N , if the kth unit is connected to the i-th node k of the system, k=1,2,...,K; i=1,2,...,N, then J ki =1, otherwise J ki =0;P B,s,t is the power flow data of branch s in the time period; is the total number of branches in the same direction, N is the total number of nodes in the district and county power grid, K is the total number of nodes including generators, P G,k,t is the power generation of power plant k in period t.

7. The method for predicting carbon emissions in the county-level power industry based on electric carbon traceability according to claim 1 is characterized by: The step (6) specifically comprises the following steps in order: (6a) Load side The load side searches for the power source one by one from the user end, and calculates the node carbon potential of the load end according to the output power proportion through the difference of carbon emission factors of each power source. The expression of the system node carbon emission factor matrix is: Where: U is the set of nodes within the coverage of the regional power grid n; P L,j,t is the load power of node j in time period t; e C,j,t is the carbon emission factor of the node; (6b) Network side The grid side traces the equivalent carbon emissions of the power plant supplying the substation, and calculates the carbon potential of each node on the power grid line by building a matrix model; since the carbon flow density of each line flowing out of the same node is equal, according to the node electricity carbon emission factor matrix, the line carbon flow density matrix is ​​calculated by the following formula: ρ B,t =diag(E C,t )H Where: B,t is the line carbon flow density matrix of time period t, which is used to describe the carbon flow density of each line in the system; H is the matrix representing the flow direction relationship of the branch flow, H = (H ij ) N×N , i, j = 1, 2, ..., N, if there is a branch connecting node i and node j, and the positive active power flow flowing into node j through this branch is not zero, then the element H ij =1,H ji =0; if the active power flow through the branch is reverse, that is, the current flows from the load to the substation, then H ij =0,H ji =1; N is the total number of nodes in the district and county power grid, E C,t is the node carbon flow rate; According to the line network loss data, the line network loss distribution matrix is ​​defined as: P BL,t =(P BL,ij,t ) N×N Where: P BL,t is the line network loss distribution matrix in time period t, which is used to describe the network loss distribution of the power system; P BL,ij,t is the matrix P BL,t The non-diagonal elements of represent the network loss of the power flow between branches ij, i≠j; the diagonal elements P BL,ii,t is 0; According to the line power flow distribution matrix, network loss distribution matrix and line carbon flow density matrix, the line carbon flow rate and network loss carbon flow rate are calculated: R B,t =ρ B,t P B,t R BL,t =ρ B,t P BL,t In the formula, R B,t is the line carbon flow rate in period t; R BL,t is the network loss carbon flow rate in time period t, P BL,t is the line network loss distribution matrix of time period t, which is used to describe the network loss distribution of the power system; P B,t is the power flow distribution matrix of the line in period t; (6c) Storage side: In the charging state, the energy storage element is regarded as a load, and in the discharging process, the energy storage element is regarded as a distributed unit.

8. The method for predicting carbon emissions in the district and county-level power industry based on electric carbon traceability according to claim 1 is characterized by: The step (7) specifically comprises the following steps in order: (7a) Divide the 365-day average load into intervals. For data segments with relatively concentrated loads, adopt a detailed grading method to ensure that the number of loads in each grading is balanced and the annual proportion of each grading is obtained; the two feeders in the same partition are required to have the same number of loads; According to the definition of network loss carbon flow rate, the indirect carbon emissions of line network loss are calculated by the following formula: In the formula, is the indirect carbon emission of line network loss, R BL,t is the network loss carbon flow rate in time period t, Δt is the time period; According to the regional carbon emission factor, the indirect carbon emissions from electricity consumption of users in different periods of time are quantified by the following formula: F C,m,t =e′ C,n,t P L,m,t Δt Where: F C,m,t is the indirect carbon emissions from electricity consumption of user m in period t; L,m,t is the power consumption of user m in time period t, and user m is located in regional power grid n; e′ C,n,t is the regional carbon emission factor; (7b) Calculate the load flow of the graded load based on the feeder group to obtain the corresponding carbon emission data; (7c) Based on the proportion of each category and the carbon emissions of each category, the probability weighted method is used to superimpose and obtain the annual carbon emissions.

Images