A rapid assessment method for carbon emission factors of electricity for large-scale power systems

Abstract

The present invention relates to the technical field of electric carbon emission, and in particular to a method for rapid assessment of electric carbon emission factors applicable to large-scale power systems, the method comprising: obtaining a data set related to carbon emission factors of the power system; performing node aggregation on the data set related to carbon emission factors based on branch power flow directions; storing the data set related to carbon emission factors after node aggregation based on a sparse matrix data structure; and using the data set related to carbon emission factors stored in the sparse matrix data structure for training to extract the linear relationship between the carbon emission combination injected by the unit under different power flow directions and the electric carbon emission factors at the node. The purpose is to improve the calculation efficiency of electric carbon emission factors with high temporal and spatial resolution.

Description

Technical Field

[0001] The present invention relates to the technical field of electricity carbon emission, and in particular to a method for rapid assessment of electricity carbon emission factors applicable to large-scale power systems. Background Art

[0002] Carbon emission measurement technology is a core tool for understanding the current carbon emission status and exploring the potential for carbon emission reduction, and is also a key support for achieving carbon reduction goals. As the main source of carbon emissions, the power industry accounts for about 40% of the country's total emissions, and this proportion is expected to rise as the degree of electrification deepens. Therefore, the measurement of carbon emissions in the power industry is particularly critical, and accurate calculation of carbon emission factors can provide solid data support for carbon emission reduction strategies.

[0003] At present, the industry widely uses the average emission factor method to measure the carbon emission level of electricity. This method calculates the average emission factor of the power grid by counting the equivalent carbon emissions of local power generation fuel consumption and external power imports. Although this method is intuitive in principle, easy to implement, and has been widely adopted, the limitations of the average emission factor method are gradually emerging in the face of the growing proportion of wind and photovoltaic power generation. First, the output of wind and solar power generation is significantly intermittent and volatile, which leads to significant differences in carbon emission factors in different time periods. Secondly, due to the uneven distribution of renewable energy resources in various regions, differences in power generation structures in different regions also cause spatial heterogeneity in carbon emission factors. These problems show that the traditional average emission factor method is not sufficient in terms of temporal and spatial resolution to support the demand for accurate carbon emission measurement.

[0004] Although there have been studies devoted to developing methods for calculating carbon emission factors of electricity with higher spatiotemporal resolution, most existing technologies rely on the "matrix" central algorithm in carbon emission flow theory. Although this algorithm performs well when dealing with small-scale systems or when the computing speed is not high, it is inefficient in computing when facing large-scale power grids (such as provincial or cross-regional power grids), especially when real-time monitoring at the minute level or frequent large-scale data processing is required. Therefore, in order to overcome these challenges, there is an urgent need for a method that can improve computing efficiency while ensuring calculation accuracy to meet the high spatiotemporal resolution requirements of future large-scale power systems for carbon emission factors of electricity. Summary of the invention

[0005] In order to improve the calculation efficiency of high temporal and spatial resolution electricity carbon emission factors, the present invention provides a rapid assessment method for electricity carbon emission factors applicable to large-scale power systems. The technical solutions adopted are as follows:

[0006] The technical solution of the present invention provides a method for rapid assessment of carbon emission factors of electricity applicable to large-scale power systems, the method comprising:

[0007] Obtain data sets related to carbon emission factors of power systems;

[0008] Node aggregation of carbon emission factor related data sets based on branch flow direction;

[0009] The carbon emission factor-related data sets after node aggregation are stored based on the sparse matrix data structure;

[0010] The carbon emission factor related data set stored in the sparse matrix data structure is used for training to extract the linear relationship between the unit injected carbon emission combination and the node electricity carbon emission factor under different tidal flow directions.

[0011] Furthermore, node aggregation of the carbon emission factor related data set based on branch flow direction includes:

[0012] According to the power system topology and power flow direction information, extract the leaf nodes at the end of the power flow and with only a single injected power flow and the root nodes matching the leaf nodes;

[0013] Node aggregation is performed based on the relationship between the electricity carbon emission factors of the leaf nodes and the root nodes, and the equivalent load of the root node after aggregation is calculated;

[0014] The power flow distribution matrix and unit output matrix in the carbon emission factor related data set are updated according to the aggregated root node.

[0015] Furthermore, node aggregation is performed based on the relationship between the electric carbon emission factors of the leaf nodes and the root nodes, and the equivalent load of the root node after aggregation is calculated, including:

[0016] Based on the flow characteristics, when the leaf node and the root node have the same carbon emission factor, the leaf node and the root node are aggregated;

[0017] The equivalent load of the aggregated root node is calculated based on the sum of the loads of all aggregated leaf nodes.

[0018] Furthermore, storing the carbon emission factor-related data set after node aggregation based on the sparse matrix data structure includes:

[0019] The updated power flow distribution matrix and unit output matrix are traversed, and non-zero elements are stored according to the sparse matrix structure to obtain the sparse power flow distribution matrix and unit output matrix.

[0020] Furthermore, the carbon emission factor-related data set stored in the sparse matrix data structure is used for training to extract the linear relationship between the unit injection carbon emission combination and the node electricity carbon emission factor under different flow direction states, including:

[0021] According to the relationship between the node inflow flow and the flow distribution, the node electricity carbon emission factor vector is calculated in combination with the unit output information;

[0022] Based on the unit output data and the node electricity carbon emission factor vector training, the linear relationship between the system injected carbon emissions and the node electricity carbon emission factors is obtained.

[0023] Furthermore, the expression for calculating the node electricity carbon emission factor vector is:

[0024]

[0025] In the formula, represents the node electricity carbon emission factor vector, To include A column vector of elements, ,element Indicates Nodes in the period Carbon content of electricity represents the node flux matrix, which is Order square array, ,element Indicates time period Flow into The total amount of power flow at each node; represents transpose; Represents the carbon emission factor vector of the unit's power generation, which contains the carbon content information of each unit's kilowatt-hour; After aggregation Physical power flow distribution matrix of power system in time period; After aggregation Output matrix of units in different time periods.

[0026] Furthermore, the linear relationship between the system injected carbon emissions and the electricity carbon emission factors of each node is expressed as:

[0027]

[0028] In the formula, Indicates Nodes in the period The carbon content of electricity per kilowatt-hour is the carbon emission factor of node electricity; Representation Node In the period The nodes inject carbon emissions; Represents the linear coefficient, used to describe the node The injected carbon emissions to the node The impact of electricity carbon emission factors; Indicates The constant term of the nodes.

[0029] Furthermore, according to the unit output data and the node electricity carbon emission factor vector training, the linear relationship between the system injected carbon emissions and each node electricity carbon emission factor also includes:

[0030] The linear relationship between the system injected carbon emissions and the electric carbon emission factors of each node is converted into a matrix, and its expression is:

[0031]

[0032] In the formula, is the fitting coefficient matrix, which is a Order square array, ,element represents the linear coefficient; is the column vector of the fitting constant term, which is a , ,element Indicates The constant term of the nodes.

[0033] Furthermore, the method also includes: considering the impact of changes in flow direction on the linear relationship, defining the flow mode and calculating the electricity carbon emission factor of the node in the current time period.

[0034] Furthermore, considering the impact of the change in flow direction on the linear relationship, defining the flow mode and calculating the electricity carbon emission factor of the node in the current period include:

[0035] Different tidal current directions are defined as tidal current modes, and a tidal current mode library matrix is ​​constructed based on the fitting coefficient matrix and constant term matrix under different tidal current modes;

[0036] Based on the node power flow mode in the current period, the mode number and the corresponding fitting coefficient matrix and constant term matrix are extracted from the power flow mode library matrix;

[0037] Based on the linear relationship between the unit's injected carbon emission combination and the node's electric carbon emission factor, the value of the node's electric carbon emission factor in the current period is solved.

[0038] The present invention has the following beneficial effects:

[0039] The present invention provides a rapid assessment method for electric carbon emission factors applicable to large-scale power systems. First, from the perspective of reducing the scale of system nodes, the carbon emission factor related data sets are node aggregated based on the branch power flow direction; secondly, from the perspective of reducing the amount of data storage and accelerating the data reading and writing speed, the carbon emission factor related data sets after node aggregation are stored based on a sparse matrix data structure; finally, from the perspective of reducing computational complexity, the carbon emission factor related data sets stored in the sparse matrix data structure are used for training to extract the linear relationship between the unit injected carbon emission combination and the node electric carbon emission factor under different power flow direction states; thereby, rapid calculation of electric carbon emission factors for large-scale power systems is achieved, which can significantly reduce the calculation time when the large-scale power system uses a sampling method to perform multiple large-scale calculations of carbon emission factor information when performing high temporal and spatial resolution electricity carbon emission factor prediction and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A method flow chart of a method for rapid assessment of electric carbon emission factors applicable to large-scale power systems provided by one embodiment of the present invention;

[0042] Figure 2 A schematic diagram of the aggregation of leaf nodes and root nodes provided by an embodiment of the present invention;

[0043] Figure 3 A schematic diagram showing the difference in the matrix inversion process after the tidal flow direction has changed provided by an embodiment of the present invention;

[0044] Figure 4 A comparison chart of the acceleration effect of the linear fitting method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, characteristics and effects of a rapid assessment method of electric carbon emission factors suitable for large-scale power systems proposed in accordance with the present invention, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The following is a detailed description of a method for rapid assessment of carbon emission factors of electricity for large-scale power systems provided by the present invention in conjunction with the accompanying drawings.

[0048] See also Figure 1 , which shows a method flow chart of a method for rapid assessment of carbon emission factors of electricity for large-scale power systems provided by an embodiment of the present invention, the method comprising:

[0049] Step S100: Obtain data related to the carbon emission factor of the power system; specifically, obtain the measured output information of each unit of the power system, physical flow distribution, and carbon emission factor information of the unit power generation in the current metering period (usually 5 minutes). The specific data includes:

[0050] The unit output matrix can be expressed as:

[0051]

[0052] In the formula, express The output matrix of the unit during the period is a The matrix of order, represents the number of power system units, Represents the number of nodes in the system, each element in the unit output matrix , indicating the Units in Output during the time period;

[0053] The physical power flow distribution matrix can be expressed as:

[0054]

[0055] In the formula, express The physical power flow distribution matrix of the power system during the period is Order square array; Represents the elements in the physical power flow distribution matrix. If the node With Node When there is a branch line connecting The time period flows into the node through this branch The positive active power flow is ,but , ; If the active power flow through the branch For reverse flow, , ; If the node With Node When there is no branch connected or no current flows through the branch ; It should be noted that, in particular, for all diagonal elements ,have ;

[0056] The carbon emission factor vector of the unit power generation can be expressed as:

[0057]

[0058] In the formula, Represents the carbon emission factor vector of the unit power generation, which is a A column vector whose elements Indicates the carbon content of electricity of the unit;

[0059] Step S200: performing node aggregation on the carbon emission factor related data set based on the branch power flow direction;

[0060] Step S200 specifically includes:

[0061] Step S210: extracting leaf nodes at the end of the power flow and having only a single power flow injected therein and root nodes matching the leaf nodes according to the power system topology and power flow direction information; specifically, the power flow direction information can be obtained from the real-time physical power flow distribution matrix obtained in step S100; the leaf node specifically refers to a node at the end of the power flow and having only a single power flow injected therein, and the root node is an upper-level node matching the leaf node; for example, in a complex power network topology diagram, tracing back in the reverse direction of the power flow, find the upper-level connection node corresponding to the leaf node as the root node;

[0062] Step S220: Node aggregation is performed based on the relationship between the electric carbon emission factors of the leaf nodes and the root nodes, and the equivalent load of the root node after aggregation is calculated; specifically, in the calculation of the carbon emission flow of the power system, the carbon emission information of the flow flowing to the downstream node will not affect the carbon emission information of the flow flowing to the upstream; this means that the electric carbon emission factor information of the node is only related to the carbon emission factor of the flow flowing into the node, and has nothing to do with the carbon emission information of the flow flowing out of the node; and the electric carbon emission factor of the flow flowing out of the node is equal to the electric carbon emission factor of the node itself; therefore, based on the flow direction characteristics, when the electric carbon emission factors of the leaf node and the root node are the same, the leaf node and the root node are aggregated, and the relationship between the electric carbon emission factor of the leaf node and the root node can be expressed as:

[0063]

[0064] In the formula, Represents a set of leaf nodes; represents the electricity carbon emission factor of the root node; Indicates The carbon emission factor of each leaf node is checked. If the carbon emission factor of each leaf node and its corresponding root node satisfies the formula, the leaf node and the root node are aggregated into a new root node. This is because the carbon emission factor of the leaf node is equal to the root node, so they can be aggregated into a new root node. The difference in the external characteristics of the aggregated root node is mainly reflected in the load level. The equivalent load of the aggregated root node is calculated based on the sum of the loads of all aggregated leaf nodes, which can be expressed as:

[0065]

[0066] In the formula, represents the equivalent load of the root node after aggregation, Indicates The load of each leaf node; that is, the aggregated root node load is the sum of all aggregated leaf node loads; this formula reflects the size of the power load carried by the aggregated node; for example, in a local distribution network area containing multiple leaf nodes, by calculating the equivalent load of the aggregated root node, we can clearly know the overall power consumption of this area and understand the load level that its power supply needs to meet, and then judge whether the operation status of the node or even the entire local network is in a reasonable range, and whether there is a risk of overload;

[0067] Step S200 aggregates the nodes of the data set based on the branch flow direction. First, the relationship between the flow direction and the carbon emission information is used to aggregate the leaf nodes and the root nodes, which simplifies the system structure and reduces the number of nodes. Second, the equivalent load of the aggregated root node is calculated to integrate the scattered load data, simplify the load representation at the node level from the system perspective, reduce the amount of data to be processed separately, and create a simpler system model for subsequent operations such as sparse matrix-based data storage, matrix generation, and carbon emission factor calculation, thereby effectively reducing the calculation complexity and speeding up the overall calculation speed. See Figure 2 As shown, the schematic diagram of the embodiment of node aggregation is as follows Figure 2 As shown in the figure, the power flow direction of the IEEE 33-node distribution network system is single, and 33 leaf nodes can be aggregated into one root node. After aggregation, the load on the root node is the sum of the loads on the 12 leaf nodes. Figure 2 Where L1-L12 are the loads on the 12 leaf nodes, and L' is the root node load after aggregation.

[0068] Step S230: Update the power flow distribution matrix and unit output matrix in the carbon emission factor related data set according to the aggregated root node; specifically, after aggregation, the system power flow distribution matrix is becomes , , for Order square array, is the number of system nodes after the leaf nodes and the root nodes are aggregated, and the matrix The meaning of the elements in remains unchanged, but the order of the matrix is ​​adjusted according to the number of nodes after aggregation; after aggregation, the unit output matrix becomes , , for Matrix, matrix The meaning of the elements in also remain unchanged.

[0069] Step S300: storing the carbon emission factor related data set after node aggregation based on the sparse matrix data structure; specifically, for large-scale power systems, due to the large number of system nodes, lines, and units, the scale of the power flow distribution matrix and the unit output matrix is ​​huge. If conventional matrix storage methods (such as row storage or column storage) are used, there will be a large number of storage areas that are not effectively utilized in the large-scale matrix for storing zero-value elements. At the same time, when reading and writing the matrix, it is necessary to traverse a large number of zero-value elements, resulting in a lot of time consumption; the sparse matrix data structure only stores the non-zero element values ​​and the row and column numbers corresponding to the non-zero elements. For the power flow distribution matrix and unit output matrix with a large number of zero elements in step S200, this storage method reduces the storage demand; and when performing calculations, the calculations are only for non-zero elements. In the matrix calculation process, zero elements have no effect on the results, thereby avoiding invalid calculations of zero elements. For example, when calculating the combined impact of power flow propagation and unit output, if conventional matrix calculations require traversing all elements, sparse matrix calculations can skip zero elements, which can significantly improve the calculation speed; when processing data, the program only accesses the data required for non-zero elements, reducing memory access time. This means that when reading and writing data, there is no need to search in a large amount of zero-value data, and the memory address related to the non-zero element can be directly located, which improves data processing efficiency;

[0070] Step S300 specifically includes:

[0071] Step S310: traverse the updated power flow distribution matrix and unit output matrix, store the non-zero elements according to the sparse matrix structure, and obtain a sparse power flow distribution matrix and unit output matrix; specifically, the aggregated power flow distribution matrix yes order square matrix, where only The same elements are non-zero elements, which means that in the total Among the elements, most are zero, and only the elements representing the flow of power on the line are non-zero; during the traversal process, when a non-zero element is encountered, it is stored according to the sparse matrix structure; the aggregated unit output matrix yes rank matrix, where only The same elements are non-zero elements, which shows that Of the elements, the only ones that are truly meaningful (i.e. when the unit is producing power) are non-zero elements, and most of the rest are zero;

[0072] After adopting the sparse matrix data structure, the power flow distribution matrix can be expressed as:

[0073]

[0074] In the formula, the power flow distribution matrix under the sparse matrix structure is for The first column of the matrix represents the non-zero elements in the original matrix. The corresponding row in Indicates, for example, This means that the non-zero element is in the fifth row of the original power flow distribution matrix; the second column in the matrix indicates that the non-zero element is in the original matrix The corresponding columns in Indicates, for example , which means that the non-zero element is in the 7th column of the original matrix; the third column in the matrix represents the value of the non-zero element, which is the original power flow distribution matrix Located in the middle Line The values ​​of the non-zero elements of the column represent the and In this way, the original power distribution matrix The non-zero elements and their position information in the sparse matrix are stored in a compact form. middle;

[0075] After adopting the sparse matrix structure, the unit output distribution matrix can be expressed as:

[0076]

[0077] In the formula, the unit output distribution matrix under the sparse matrix structure is for The first column of the matrix represents the non-zero elements in the original matrix. The corresponding row in express, Corresponding unit number, if , indicating that this non-zero element in the original unit output matrix The second column in the matrix indicates the non-zero elements in the original matrix. The corresponding columns in express, Corresponding node number, if , indicating that the non-zero element is related to the 10th node in the original matrix; the third column in the matrix represents the value of the non-zero element, that is , which is the original unit output matrix Located in the middle Line The values ​​of the non-zero elements in the column represent the The unit is in In this way, the non-zero elements and their position information in the original unit output matrix are also stored in a sparse matrix in a concise form. middle;

[0078] An embodiment of step S300 is provided below:

[0079] Before and after the sparse matrix data structure is adopted, the number of elements of the power flow distribution matrix of the IEEE 9-bus system, the IEEE 39-bus system and the IEEE 118-bus system is shown in Table 1; the number of lines of the IEEE 9-bus system, the IEEE 39-bus system and the IEEE 118-bus system is 9, 46 and 186 lines respectively (transformer branches are also regarded as lines), as shown in Table 1:

[0080] Table 1

[0081]

[0082] Table 1 records the number of elements in the power distribution matrix of different systems before and after the use of sparse matrix data structure.

[0083] It can be seen from Table 1 that after using sparse matrix, the number of elements in the power flow distribution matrix of IEEE 9-node system, IEEE 39-node system and IEEE118-node system decreased by 70.3%, 90.9% and 96.0% respectively; it can be seen that the more nodes there are, the more data storage can be saved by using sparse matrix; according to the carbon flow calculation comparison of our actual system in a certain province (about 6000-node system), after using sparse matrix, the data reading and writing time is reduced from about 8 minutes to about 2 minutes;

[0084] In summary, step S300 stores the data set after node aggregation based on the sparse matrix data structure. In view of the situation where there are a large number of zero elements in the power flow distribution matrix and the unit output matrix in the large-scale power system, the storage demand is greatly reduced by storing only the non-zero elements and their row and column numbers. In terms of calculation, invalid calculation of zero elements is avoided and the calculation speed is improved. For example, zero elements can be skipped when calculating the comprehensive impact of power flow propagation and unit output. In data processing, the memory access time is reduced, and the program can directly locate the memory address related to the non-zero elements. At the same time, it facilitates operations such as matrix multiplication and decomposition, and the sparsity optimization algorithm can be used to further improve the computing performance. In the data transmission process, the transmission bandwidth is saved and the communication volume is reduced, which is conducive to data interaction and remote computing between different subsystems of the large-scale power system. The more nodes there are, the greater the amount of data storage saved, which can significantly reduce the data reading and writing time. From an actual case, after a system with about 6,000 nodes adopts a sparse matrix, the data reading and writing time is greatly reduced.

[0085] Step S400: Use the carbon emission factor related data set stored in the sparse matrix data structure for training to extract the linear relationship between the unit's injected carbon emission combination and the node electricity carbon emission factor under different tidal flow states; specifically, according to the massive historical tidal flow distribution matrix, the unit output distribution matrix, the unit power generation carbon emission factor vector and the corresponding calculated node emission factor data, train to obtain the linear correlation between the unit's injected carbon emission combination and the node carbon emission factor under different tidal flow states. According to the current tidal flow situation, find the linear coefficient corresponding to the current tidal flow direction, and calculate the node carbon emission factor data for the current period. Use linear coefficient calculation to replace carbon flow nonlinear calculation to speed up the calculation process.

[0086] Step S400 specifically includes:

[0087] Step S410: Calculate the node electricity carbon emission factor vector based on the relationship between the node inflow flow and the flow distribution and the unit output information. The expression is:

[0088]

[0089] In the formula, represents the node electricity carbon emission factor vector, To include A column vector of elements, ,element Indicates Nodes in the period Carbon content of electricity represents the node flux matrix, which is Order square array, ,element Indicates time period Flow into The total amount of power flow at each node; represents transpose; Represents the carbon emission factor vector of the unit's power generation, which contains the carbon content information of each unit's kilowatt-hour; After aggregation Physical power flow distribution matrix of power system in time period; After aggregation Output matrix of units in each period;

[0090] Step S420: According to the unit output data and the node electric carbon emission factor vector training, the linear relationship between the system injected carbon emission and the electric carbon emission factor of each node is obtained; specifically, a large amount of historical data is collected, including the unit output data of each node in each time period (for example, the actual output situation record of each unit at each node at different times) and the node electric carbon emission factor vector data of the corresponding time period calculated by step S410. The time span of these historical data should be long enough and cover a variety of different system operating conditions to ensure that the system characteristics can be fully reflected; for each node, the linear relationship between its electric carbon emission factor and the injected carbon emission of all nodes is found through the analysis and calculation of historical data, which can be expressed as:

[0091]

[0092] In the formula, Indicates Nodes in the period The carbon content of electricity per kilowatt-hour is the carbon emission factor of node electricity; Representation Node In the period The nodes inject carbon emissions; Represents the linear coefficient, used to describe the node The injected carbon emissions to the node The impact of electricity carbon emission factors; Indicates The constant term of the nodes, As a fixed offset in the linear relationship; using rich historical data to mine the linear relationship between the electricity carbon emission factor of each node and the carbon emissions injected by all nodes in the system, revealing the inherent correlation between the nodes in carbon emissions, so that the node electricity carbon emission factor can be easily calculated based on this quantitative linear relationship through the known node injection carbon emissions, simplifying the complex carbon emission factor analysis process, and providing key theoretical and data basis for more efficient and accurate evaluation of electricity carbon emission factors;

[0093] Step S430: Convert the linear relationship between the system injected carbon emissions and the electric carbon emission factors of each node into a matrix, which is expressed as follows:

[0094]

[0095] In the formula, is the fitting coefficient matrix, which is a Order square array, ,element represents the linear coefficient; is the column vector of the fitting constant term, which is a , ,element Indicates The constant term of each node; specifically, this formula presents the originally scattered linear relationship between the system's injected carbon emissions and the electric carbon emission factors of each node in a unified and concise matrix form, making this complex correlation clearer and more intuitive, easier to understand, and convenient for efficient calling and calculation in the overall calculation framework of large-scale power systems; it is helpful to use this matrix form to quickly complete the calculation of the node electric carbon emission factors when processing complex situations such as different operating conditions and different power flow directions in the subsequent process, thereby improving the systematicness and computational efficiency of the entire electric carbon emission factor evaluation process.

[0096] The rapid assessment method for carbon emission factors of electricity for large-scale power systems also includes:

[0097] Step S500: Considering the influence of the change of tidal flow direction on the linear relationship, the tidal flow mode is defined and the corresponding matrix is ​​used to store relevant data;

[0098] Step S500 specifically includes:

[0099] Step S510: define different tidal flow directions as tidal modes, and construct a tidal mode library matrix according to the fitting coefficient matrix and constant term matrix under different tidal modes; specifically, the tidal flow direction is crucial for carbon flow calculation. When the tidal flow is determined, the matrix There is a linear relationship between the flow injected into the system, but if the flow direction changes, the flow distribution matrix The position of the non-zero elements in the matrix will change, and this position change has a significant and discontinuous effect on the matrix inversion result, which is further reflected in the linear relationship between the system injection flow (unit output injection) and the carbon emission factor of each node, which is manifested as a jump in the value of the fitted linear coefficient and the value of the constant term; the value change of the non-zero elements of the sparse matrix will have a small and continuous effect on the matrix inversion result, but the position change of the non-zero elements of the sparse matrix will have a significant and discontinuous effect on the matrix inversion result; the difference in the matrix inversion process after the flow direction changes can be seen in Figure 3As shown;

[0100] according to Figure 3 It can be seen that the above characteristics are reflected in the linear relationship between the system injection flow (unit output injection) and the carbon emission factor of each node: when the flow direction changes, the fitted linear coefficient value and the constant term value will jump; therefore, for a system with M flow directions, the fitting coefficient matrix and the constant term matrix It is expressed as:

[0101]

[0102]

[0103] The M kinds of tidal current directions are defined as tidal current modes, and the tidal current mode library matrix is ​​used Stores power flow mode data. The power flow mode is represented by 0,1 variables. , , whose elements Indicates that when there is a positive flow between nodes i and j, the element ,otherwise ;

[0104] Step S520: Based on the node power flow mode in the current period, the mode number and the corresponding fitting coefficient matrix and constant term matrix are extracted from the power flow mode library matrix; in the calculation process, firstly, according to the power flow mode in the current situation, the mode number in the mode library is searched, the current power flow mode number m is determined, and the fitting coefficient vector under the corresponding mode is selected. and constant term vector ; These matrices and vectors carry the key parameter information of the linear relationship between the system's injected carbon emissions and the electricity carbon emission factors of each node under the current power flow mode, and are an important basis for the subsequent calculation of the value of the node electricity carbon emission factor in the current period;

[0105] Step S530: Solve the electric carbon emission factor value of the node in the current period based on the linear relationship between the unit's injected carbon emission combination and the node's electric carbon emission factor; use the fitting coefficient matrix and constant term vector extracted in step S520, combined with the linear relationship between the unit's injected carbon emission combination and the node's electric carbon emission factor to perform calculation operations; substitute the known unit's injected carbon emission related data in the current period into the linear relationship expression, and accurately calculate the electric carbon emission factor value of the node in the current period according to the operation rules such as matrix multiplication and addition, so as to obtain the carbon content of electricity per kilowatt-hour of each node under the current system operation state;

[0106] Step S500 defines the power flow mode and constructs the power flow mode library matrix by considering the impact of power flow direction changes on the linear relationship, which can cope with the problem of linear relationship parameter jumps caused by power flow direction changes; in actual calculations, the corresponding parameters can be quickly extracted and calculated based on the current power flow mode, which not only enables the entire electric carbon emission factor calculation process to flexibly adapt to different power flow directions and ensure the accuracy of the calculation, but also greatly improves the efficiency of electric carbon emission factor calculation in complex and changeable power system operating environments (there are various power flow direction change scenarios), providing a strong guarantee for large-scale power systems to achieve accurate and efficient electric carbon emission assessment.

[0107] For specific embodiments, see Figure 4 As shown, Figure 4 The IEEE 9-node, IEEE 39-node, and IEEE 118-node systems were used to test the effect of the method of the present application on improving the computing speed. The test results are as follows: Figure 4 As shown; Figure 4 The left axis in the figure represents the logarithm of the computing time, and the right axis represents the improvement in computing efficiency. The IEEE 9-node, IEEE 39-node, and IEEE 118-node systems were sampled 80,000 times, 100,000 times, and 1 million times, respectively. Figure 4 It can be seen that for the 9-node system, the calculation efficiency of the algorithm is not greatly improved. The proposed calculation framework can reduce the calculation time from 79.8s to 65.5s, reducing the calculation time by 17.9%; for the 39-node system, the calculation efficiency is improved by 34.2% (from 150.1s to 98.7s) through the adaptive calculation framework, and for the 118-node system, the calculation efficiency is improved by 42.8% (from 1460.27s to 834.49s). It can be seen that the high spatiotemporal resolution emission factor fast calculation method proposed in the present invention has better effect for larger systems, and can effectively improve the calculation efficiency of electric carbon emission factors.

[0108] It should be noted that the reduction of computing scale through node aggregation, the reduction of storage scale and improvement of reading and writing speed through sparse matrices, and the reduction of computing complexity through adaptive linearization proposed in the present invention can be used alone or in combination.

[0109] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A rapid assessment method for carbon emission factors of electricity for large-scale power systems, characterized in that: The method comprises: Obtain data sets related to carbon emission factors of power systems; Node aggregation of carbon emission factor related data sets based on branch flow direction; The carbon emission factor-related data sets after node aggregation are stored based on the sparse matrix data structure; The carbon emission factor-related data set stored in the sparse matrix data structure is used for training to extract the linear relationship between the unit injection carbon emission combination and the node electricity carbon emission factor under different flow direction states, including: According to the relationship between the node inflow flow and the flow distribution, the node electricity carbon emission factor vector is calculated in combination with the unit output information; According to the unit output data and the node electricity carbon emission factor vector training, the linear relationship between the system injected carbon emissions and the node electricity carbon emission factors is obtained, and its expression is: In the formula, Indicates Nodes in the period The carbon content of electricity per kilowatt-hour, i.e., the carbon emission factor of node electricity; Representation Node In the period The nodes inject carbon emissions; Represents the linear coefficient, used to describe the node The injected carbon emissions to the node The impact of electricity carbon emission factors; Indicates The constant term of the nodes; Considering the impact of the change in flow direction on the linear relationship, the flow mode is defined and the electricity carbon emission factor of the node in the current period is calculated, including: Different tidal current directions are defined as tidal current modes, and a tidal current mode library matrix is ​​constructed based on the fitting coefficient matrix and constant term matrix under different tidal current modes; Based on the node power flow mode in the current period, the mode number and the corresponding fitting coefficient matrix and constant term matrix are extracted from the power flow mode library matrix; Based on the linear relationship between the unit's injected carbon emission combination and the node's electric carbon emission factor, the value of the node's electric carbon emission factor in the current period is solved.

2. The rapid assessment method for carbon emission factors of electricity according to claim 1, characterized in that: Node aggregation of carbon emission factor related data sets based on branch flow direction includes: According to the power system topology and power flow direction information, extract the leaf nodes at the end of the power flow and with only a single injected power flow and the root nodes matching the leaf nodes; Node aggregation is performed based on the relationship between the electricity carbon emission factors of the leaf nodes and the root nodes, and the equivalent load of the root node after aggregation is calculated; The power flow distribution matrix and unit output matrix in the carbon emission factor related data set are updated according to the aggregated root node.

3. The rapid assessment method for carbon emission factors of electricity according to claim 2, characterized in that: Node aggregation is performed based on the relationship between the electric carbon emission factors of the leaf nodes and the root nodes, and the equivalent load of the root node after aggregation is calculated, including: Based on the flow characteristics, when the leaf node and the root node have the same carbon emission factor, the leaf node and the root node are aggregated; The equivalent load of the aggregated root node is calculated based on the sum of the loads of all aggregated leaf nodes.

4. The rapid assessment method for carbon emission factors of electricity according to claim 2, characterized in that: The storage of the carbon emission factor related data set after node aggregation based on the sparse matrix data structure includes: The updated power flow distribution matrix and unit output matrix are traversed, and non-zero elements are stored according to the sparse matrix structure to obtain the sparse power flow distribution matrix and unit output matrix.

5. The rapid assessment method for carbon emission factors of electricity according to claim 1, characterized in that: The expression for calculating the node electricity carbon emission factor vector is: In the formula, represents the node electricity carbon emission factor vector, To include A column vector of elements, ,element Indicates Nodes in the period Carbon content of electricity per kilowatt-hour; represents the node flux matrix, which is Order square array, ,element Indicates time period Flow into The total amount of power flow at each node; represents transpose; Represents the carbon emission factor vector of the unit's power generation, which contains the carbon content information of each unit's kilowatt-hour; After aggregation Physical power flow distribution matrix of power system during the period; After aggregation Output matrix of units in different time periods.

6. The rapid assessment method for carbon emission factors of electricity according to claim 5, characterized in that: According to the unit output data and node electricity carbon emission factor vector training, the linear relationship between the system injected carbon emissions and each node electricity carbon emission factor also includes: The linear relationship between the system injected carbon emissions and the electric carbon emission factors of each node is converted into a matrix, and its expression is: In the formula, is the fitting coefficient matrix, which is a Order square array, ,element represents the linear coefficient; is the column vector of the fitting constant term, which is a , ,element Indicates The constant term of the nodes.

Images