Electric carbon tracking method based on electric power flow topological grid modeling

Abstract

The invention discloses an electric carbon tracking method based on electric power flow topological grid modeling, and solves the problems that the electric power transmission path is not clear and the electric carbon factor calculation is not refined enough in various voltage level electric power grids. The method comprises the following steps: based on a regional power grid architecture, establishing an electrical load flow calculation model of an electrical carbon tracking standard grid; inputting power supply output, transformer substation and line power flow, generating a unit injection matrix, a branch power flow distribution matrix and a load distribution matrix, generating a node adjacency shunt matrix, calculating the proportion of outflow power line power flow of each node, and generating a node adjacency shunt coefficient matrix; calling a depth-first search algorithm for the nodes, calculating directed line paths from the starting node to the tail end node, forming shunting coefficients contributed by all paths from power flow generated by the starting node to the tail end node, obtaining a power distribution matrix in combination with the power injected into the starting node by the corresponding unit, and obtaining a power distribution matrix according to the carbon emission intensity of each power supply; a node carbon potential matrix is obtained through calculation, and electric carbon tracking of a power supply layer, a power grid layer and a load layer is achieved.

Description

Technical Field

[0001] The invention relates to an electric carbon tracing method based on electric power flow topology grid modeling, belonging to the technical field of carbon emission detection. Background Art

[0002] The electricity carbon factor, also known as the electricity carbon dioxide emission factor or the grid carbon emission factor, refers to the carbon dioxide emissions caused by the use of unit electricity in the production process of products. It is the basic basis for calculating the indirect carbon emissions of electricity during the entire life cycle of products. Its coverage and time-varying characteristics are crucial to the accuracy of calculating the carbon footprint of products. The electricity carbon factor is not detailed and accurate enough.

[0003] The existing technology has the problems that the flow path in the existing power grid of various voltage levels is unclear and the calculation of the electric carbon factor is not detailed and accurate enough. Summary of the invention

[0004] The purpose of the present invention is to provide an electric carbon tracking method based on power flow topology grid modeling, focusing on solving the problems of unclear power flow paths and insufficiently detailed and accurate calculation of electric carbon factors in existing power grids covering various voltage levels. On the basis of a reasonable analysis of the power grid topology containing various types of power sources, lines of different voltage levels, substations of different voltage levels and loads in the region, a hybrid optimization carbon emission calculation model for the power system is established. By calculating the unit electric carbon emissions of various types of power sources, the carbon intensity of substations and lines is solved, thereby realizing real-time perception of the real-time carbon emission intensity of sources, networks and loads in the power system. The present invention provides a detailed process and precise measurement method for the real-time carbon emission intensity of sources, networks and loads at different voltage levels and locations, which is conducive to narrowing the geographical scope of electric carbon factor calculation, refining the calculation period, realizing low-carbon power generation and consumption matching at the hourly level or even shorter period, and reasonably allocating the carbon emission responsibility contained in the electric energy trading flow, so as to make the accounting results more accurate.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A method for tracking electric carbon based on regional power grid flow topology analysis includes the following steps:

[0007] 1) Based on the regional power grid architecture, the regional power grid model is simplified into a power flow topology structure, and an electrical flow calculation model of the electric carbon tracking standard grid is established;

[0008] 2) For any time t, perform the following steps:

[0009] Enter the power output and generate the unit injection matrix P G , the unit injection matrix at time t is a K×N-order matrix, K is the total number of regional generators, N is the total number of grid nodes, and Indicates that the elements in the matrix are specifically defined as follows:

[0010] If the kth generator set is connected to the regional grid node j, k = 1, 2, ... K, and the active power flow injected from the kth generator set into the grid node j is p, then otherwise

[0011] 3) Input the unit's carbon emission intensity vector, including the unit carbon emission intensity of all units, and generate the unit carbon emission intensity matrix E G , the unit electricity carbon emission intensity matrix at time t is a K×1-order matrix, using express;

[0012] 4) Generate node power flow matrix P based on the power flow of each node in the system B , the matrix at time t is an N×N matrix, using It means that the matrix contains both the topological structure information of the power network and the direction and size information of the active power flow of the power system. The elements of the node power flow matrix are specifically defined as follows:

[0013] If there is a branch connecting node i and node j (i, j = 1, 2, ..., N), and the positive active power flow from node i to node j through this branch is p, then at the same time On the contrary, if the active power flow through this branch is from node j to node i, then In other cases, In particular, for all diagonal elements, we have

[0014]

[0015] 5) Generate the node load matrix P according to the load of each node L , the node load matrix at time t is an N×N matrix, Indicates; this matrix reflects the power load mounted on the node, and the elements in the matrix are specifically defined as follows:

[0016] If node j is a node with load, and the active load is p, then otherwise Node load matrix P L is a diagonal matrix;

[0017] 6) For the node power flow matrix And the node load matrix Generate node power flow matrix The calculation formula is as follows:

[0018]

[0019] In the formula, is the active power flowing from node i to node j at time t, The load mounted on node i;

[0020] 7) The node power split coefficient is used to describe the distribution relationship between the injected power and the outflow power in a single node. In the power system, the power flow flows into node n through the i-th branch and flows out of node n through the j-th branch. The branch sets of the power flow flowing into and out of node n are N respectively. + and N － , then the active power flows from the i-th branch into node n and from the j-th branch out of node n are P i ,P j (i∈N + ,j∈N - ), the electric energy is shared among n nodes according to the split ratio, so any branch of the outflow flow has a component of each inflow flow, and the size of the component is proportional to the power of each outflow branch;

[0021] From the node power flow matrix And the node power flow matrix From the definition of The data in the i-th row represents the active power that node i flows out to other nodes j and the active power value of the load mounted by itself. Therefore, the node power split coefficient matrix is ​​calculated based on the active power data in the i-th row: The node power split coefficient matrix at time t is an N×N matrix, Represents the matrix, and the specific calculation formulas for each element in the matrix are as follows:

[0022]

[0023] In the formula, is the active power flowing from node i to node j at time t, is the power splitting coefficient of node j at node i at time t;

[0024] 8) Node adjacency splitting matrix It is regarded as the adjacency end point matrix of the directed graph, and the depth-first search algorithm is used to calculate the directed line path from the starting node to the end node to form the line path storage set F t + ;

[0025] 9) After obtaining the line path storage set F t + Later, the path set F is further calculated t +The total diversion coefficient of each path in is calculated by the following steps:

[0026] (1) Store the set F according to the line path t + , for a certain starting node i, read the node adjacency splitting coefficient matrix in turn The matrix flow splitting coefficient of each path from the starting node to the end node Multiply the above split coefficients to get the total split coefficient of the path

[0027]

[0028] In the formula, let i be the starting node, m and n be the path nodes, and s be the end node. is the power split coefficient flowing from node i to node j at time t, is the power split coefficient flowing from node j to node m at time t, is the power split coefficient from node m to node n at time t, is the power split coefficient from node n to node s at time t, is the total diversion coefficient of a single path from node i to node s at time t;

[0029] (2) By reading the unit injection matrix The power injected by the kth unit into the starting node i Get the power distribution value of the power supply from the starting node i through j, m, n to the end node s The calculation formula is as follows:

[0030]

[0031] (3) Repeat steps (1) to (2) until the split coefficients of all paths and the power distribution values ​​of all power sources are calculated, and check the total split coefficients from each starting node to ensure that the sum of the total split coefficients of all paths starting from a certain starting node is 1. After the check is correct, the distribution matrix D of the power at time t is output. t , the power distribution matrix at time t is a K×N matrix, using Indicates that the matrix form is as follows:

[0032]

[0033] The distribution matrix D of power at time t t Each column is normalized to obtain the normalized power distribution matrix

[0034]

[0035] in: It is the sum of the power allocation values ​​injected into node s by all units k in the grid;

[0036] The normalized matrix is ​​transposed and further compared with the unit carbon emission intensity matrix Multiply them together to calculate the carbon potential vector of the generator set and the node carbon potential matrix E t , realizing the electric carbon tracking at the power supply layer, grid layer and load layer;

[0037]

[0038] Right now

[0039]

[0040] Where: The node carbon potential matrix E at time t t The matrix is ​​an N×1-order matrix, which represents the electric carbon intensity of the grid node at time t.

[0041] The purpose of the present invention can also be further achieved by the following technical measures:

[0042] The specific process of simplifying and modeling step 1) of the aforementioned electric carbon tracking method based on regional power grid flow topology analysis is as follows:

[0043] (1) The number of busbar nodes is determined by the number of busbars in the substation;

[0044] (2) Determine the number of nodes for transformer modeling based on the number of transformer windings;

[0045] (3) Each load line connecting the high, medium and low voltage busbars of the substation, with the end of each load line modeled as a node;

[0046] (4) For the power source connected to the substation, each power source point should be modeled separately as a node;

[0047] (5) For the branches between nodes, the branches between nodes are formed based on the interconnection current between the busbars, the current between the voltage windings of the transformer and the neutral point, the line transmission current, and the power injection current.

[0048] In the aforementioned electric carbon tracking method based on regional power grid flow topology analysis, step 8) uses a depth-first search algorithm to calculate the line path storage set F t + The specific process is as follows:

[0049] (1) Read the node power flow matrix The i-th row of The corresponding node j is included in the line path storage set F t + , mark the current node j as visited;

[0050] (2) Read the node power flow matrix The jth row of The corresponding node k is included in the line path storage set F t + ;

[0051] (3) Repeatedly select node k until the selected node In the example, the row where node k is located has only Then the node is a load node and no longer sends power outward. The node is included in the line path storage set F t + , output a complete power flow path starting from node i and mark node k as unvisited;

[0052] (4) Save the power flow path starting from node i, mark node j as unvisited, and return to node i;

[0053] (5) Repeat the above steps to complete the traversal of all nodes in (1), and then obtain the line path storage set F required by the algorithm. t + .

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. The electric carbon tracking method based on power flow topology grid modeling provided by the present invention realizes accurate calculation of main grid power carbon emissions. Relying on the power flow analysis results, the power transmission path from the power source to each substation is calculated in real time, and a power system carbon emission calculation model is further established. By calculating the unit electric carbon emissions of various types of power sources, the carbon intensity of substations and lines is solved, thereby realizing real-time perception of the real-time carbon emission intensity of sources, networks, and loads in the power system.

[0056] 2. The electricity carbon tracking method based on power flow topology grid modeling provided by the present invention is based on the electricity carbon tracking algorithm and the real-time power grid topology structure and power generation and transmission data to obtain the time-sharing and zoned electricity carbon factor, which is conducive to narrowing the geographical scope of electricity carbon factor calculation and refining the calculation period, realizing low-carbon power generation and consumption matching at the hourly level or even shorter time period, and reasonably allocating the carbon emission responsibility contained in the electricity trading flow, avoiding the repeated calculation of green rights and interests, making the accounting results more accurate, and at the same time promoting the certification of green electricity consumed by enterprises to third-party carbon footprint certification agencies for auxiliary certification, so as to meet the carbon footprint certification needs of enterprises' export products. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of the electric carbon tracking method based on regional power grid power flow topology analysis of the present invention;

[0058] Figure 2 It is a schematic diagram of power distribution. DETAILED DESCRIPTION

[0059] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0060] like Figure 1 As shown, the present invention provides an electric carbon tracking method based on regional power grid flow topology analysis, comprising the following steps:

[0061] Based on the regional power grid architecture, the regional power grid model is simplified into a power flow topology structure, and an electrical flow calculation model for the electric carbon tracking standard grid is established;

[0062] The specific process of simplification and modeling is as follows:

[0063] (1) The number of busbar nodes is determined by the number of buses in the substation. Taking a 220kV substation as an example, a 220kV substation is usually configured with double buses, which generates two nodes.

[0064] (2) Determine the number of nodes for transformer modeling based on the number of windings of the transformer. Taking the 220 kV main transformer as an example, it is usually a three-turn autotransformer with three sides of high, medium and low voltage levels. Four new nodes are added, namely the high voltage side, medium voltage side, low voltage side and the middle node. If it is a double-turn transformer, two new nodes are added, namely the high voltage side and the low voltage side.

[0065] (3) For each load line connecting the high, medium and low voltage busbars of the substation, the end of each load line should be modeled as a node separately.

[0066] (4) For the power source connected to the substation, each power source point should be modeled separately as a node.

[0067] (5) For the branches between nodes, the branches between nodes are mainly formed by the interconnection current between the busbars, the current between the voltage windings of the transformer and the neutral point, the line transmission current and the power injection current.

[0068] Through the above modeling simplification, the power sources, lines, buses, main transformers and loads in the actual power grid are simplified into nodes in the power grid. In order to further clearly describe the transmission path of renewable energy, it is proposed to generate the unit injection matrix, unit carbon emission intensity matrix, node flow matrix and node load matrix, and further generate the node power flow matrix and node power diversion coefficient matrix. For any time t, the specific steps are as follows:

[0069] Enter the power output and generate the unit injection matrix P G , the unit injection matrix at time t is a K×N (K is the total number of regional generators, N is the total number of grid nodes) matrix, The elements in the matrix are defined as follows.

[0070] If the kth (k = 1, 2, ... K)th generator set is connected to the regional grid node j, and the active power flow injected from the kth generator set into the grid node j is p, then otherwise

[0071] Enter the unit's carbon emission intensity vector, which contains the unit carbon emission intensity of all units (in gCO2 / kWh), and generate the unit carbon emission intensity matrix E G , the matrix at time t is a K×1 matrix, using express;

[0072] Generate the node power flow matrix P according to the power flow of each node in the system B , the matrix at time t is an N×N matrix, using This matrix contains both the topological structure information of the power network and the direction and size information of the active power flow of the power system. The elements of the node flow matrix are specifically defined as follows:

[0073] If there is a branch connecting node i and node j (i, j = 1, 2, ..., N), and the positive active power flow from node i to node j through this branch is p, then at the same time On the contrary, if the active power flow through this branch is from node j to node i, then In other cases, In particular, for all diagonal elements, we have

[0074]

[0075] Generate the node load matrix P according to the load of each node L , the node load matrix at time t is an N×N matrix, This matrix mainly reflects the power load mounted on the node. The elements in the matrix are defined as follows:

[0076] If node j is a node with load, and the active load is p, then otherwise It can be seen that the node load matrix P L is a diagonal matrix;

[0077] For the node power flow matrix And the node load matrix Generate node power flow matrix The calculation formula is as follows:

[0078]

[0079] In the formula, is the active power flowing from node i to node j at time t, The load mounted on node i.

[0080] The node power split coefficient is used to describe the distribution relationship between the injected power and the outflow power in a single node. Assume that a node n and its adjacent branches i and j in the power system are as follows: Figure 2 As shown, the power flow flows into node n through the i-th branch and flows out of node n through the j-th branch.

[0081] If the branches of the flow inflow and outflow of node n are N + and N － , then the active power flows from the i-th branch into node n and from the j-th branch out of node n are P i ,P j (i∈N + ,j∈N - ). Considering that the electric energy is shared among n nodes in proportion to the flow distribution, any branch of the outflow flow has a component of each inflow flow, and the size of the component is proportional to the power of each outflow branch;

[0082] From the node power flow matrix And the node power flow matrix From the definition of The data in the i-th row represents the active power that node i flows out to other nodes j and the active power value of the load mounted by itself. Therefore, the node power split coefficient matrix is ​​calculated based on the active power data in the i-th row: The node power split coefficient matrix at time t is an N×N matrix, Represents the matrix. The specific calculation formulas for each element in the matrix are as follows:

[0083]

[0084] In the formula, is the active power flowing from node i to node j at time t, is the power splitting coefficient of node j at node i at time t;

[0085] The node adjacency splitting matrix It is regarded as the adjacency end point matrix of the directed graph, and the directed line path from the starting node to the end node is calculated based on the depth-first search algorithm to form a line path storage set Ft + . Based on this algorithm, the line path storage set F is formed t + The specific process is as follows:

[0086] (1) Read the node power flow matrix The i-th row of The corresponding node j is included in the line path storage set F t + , mark the current node j as visited;

[0087] (2) Read the node power flow matrix The jth row of The corresponding node k is included in the line path storage set F t + ;

[0088] (3) Repeatedly select node k until the selected node In the example, the row where node k is located has only Then the node is a load node and no longer sends power outward. The node is included in the line path storage set F t + , output a complete power flow path starting from node i and mark node k as unvisited;

[0089] (4) Save the power flow path starting from node i, mark node j as unvisited, and return to node i;

[0090] (5) Repeat the above steps to complete the traversal of all nodes in (1), and then obtain the line path storage set F required by the algorithm. t + .

[0091] After obtaining the line path storage set F t + Later, we need to further calculate the path set F t + The total diversion coefficient of each path in is calculated by the following steps:

[0092] (1) Store the set F according to the line path t + , for a certain starting node i, read the node adjacency splitting coefficient matrix in turn The matrix flow splitting coefficient of each path from the starting node to the end node Multiply the above split coefficients to get the total split coefficient of the path

[0093]

[0094] In the formula, let i be the starting node, m and n be the path nodes, and s be the end node. is the power split coefficient flowing from node i to node j at time t, is the power split coefficient flowing from node j to node m at time t, is the power split coefficient from node m to node n at time t, is the power split coefficient from node n to node s at time t, is the total diversion coefficient of a single path from node i to node s at time t.

[0095] (2) By reading the unit injection matrix The power injected by the kth unit into the starting node i Thus, the power distribution value of the power supply from the starting node i through j, m, n to the end node s can be obtained. The calculation formula is as follows:

[0096]

[0097] (3) Repeat steps (1) to (2) until the split coefficients of all paths and the power distribution values ​​of all power sources are calculated, and check the total split coefficients from each starting node to ensure that the sum of the total split coefficients of all paths starting from a certain starting node is 1. After the check is correct, the distribution matrix D of the power at time t is output. t , the power distribution matrix at time t is a K×N matrix, using Indicates that the matrix form is as follows:

[0098]

[0099] The distribution matrix D of power at time t t Each column is normalized to obtain the normalized power distribution matrix

[0100]

[0101] in: It is the sum of the power allocation values ​​injected into node s by all units k in the grid.

[0102] The normalized matrix is ​​transposed and further compared with the unit carbon emission intensity matrix Multiply them together to calculate the carbon potential vector of the generator set and the node carbon potential matrix E t , realizing electric carbon tracking at the power supply layer, grid layer, and load layer.

[0103]

[0104] Right now

[0105]

[0106] Where: The node carbon potential matrix E at time t t The matrix is ​​an N×1-order matrix, which represents the electric carbon intensity of the grid node at time t.

[0107] In addition to the above embodiments, the present invention may also have other implementation modes. Any technical solutions formed by equivalent replacement or equivalent transformation shall fall within the protection scope required by the present invention.

Claims

1. A method for tracking electric carbon based on regional power grid power flow topology analysis, characterized in that: The following steps are involved: 1) Based on the regional power grid architecture, the regional power grid model is simplified into a power flow topology structure, and an electrical flow calculation model of the electric carbon tracking standard grid is established; 2) For any time t, perform the following steps: Enter the power output and generate the unit injection matrix P G , the unit injection matrix at time t is a K×N-order matrix, K is the total number of regional generators, N is the total number of grid nodes, and Indicates that the elements in the matrix are specifically defined as follows: If the kth generator set is connected to the regional grid node j, k = 1, 2, ... K, and the active power flow injected from the kth generator set into the grid node j is p, then otherwise 3) Input the unit's carbon emission intensity vector, including the unit carbon emission intensity of all units, and generate the unit carbon emission intensity matrix E G , the unit electricity carbon emission intensity matrix at time t is a K×1-order matrix, using express; 4) Generate node power flow matrix P based on the power flow of each node in the system B , the matrix at time t is an N×N matrix, using It means that the matrix contains both the topological structure information of the power network and the direction and size information of the active power flow of the power system. The elements of the node power flow matrix are specifically defined as follows: If there is a branch connecting node i and node j (i, j = 1, 2, ..., N), and the positive active power flow from node i to node j through this branch is p, then at the same time On the contrary, if the active power flow through this branch is from node j to node i, then In other cases, In particular, for all diagonal elements, we have 5) Generate the node load matrix P according to the load of each node L , the node load matrix at time t is an N×N matrix, Indicates; this matrix reflects the power load mounted on the node, and the elements in the matrix are specifically defined as follows: If node j is a node with load, and the active load is p, then otherwise Node load matrix P L is a diagonal matrix; 6) For the node power flow matrix And the node load matrix Generate node power flow matrix The calculation formula is as follows: In the formula, is the active power flowing from node i to node j at time t, The load mounted on node i; 7) The node power split coefficient is used to describe the distribution relationship between the injected power and the outflow power in a single node. In the power system, the power flow flows into node n through the i-th branch and flows out of node n through the j-th branch. The branch sets of the power flow flowing into and out of node n are N respectively. + and N － , then the active power flows from the i-th branch into node n and from the j-th branch out of node n are P i ,P j (i∈N + ,j∈N - ), the electric energy is shared among n nodes according to the split ratio, so any branch of the outflow flow has a component of each inflow flow, and the size of the component is proportional to the power of each outflow branch; By node power flow matrix And the node power flow matrix From the definition of The data in the i-th row represents the active power that node i flows out to other nodes j and the active power value of the load mounted by itself. Therefore, the node power split coefficient matrix is ​​calculated based on the active power data in the i-th row: The node power split coefficient matrix at time t is an N×N matrix, Represents the matrix, and the specific calculation formulas for each element in the matrix are as follows: In the formula, is the active power flowing from node i to node j at time t, is the power splitting coefficient of node j at node i at time t; 8) Node adjacency splitting matrix It is regarded as the adjacency end point matrix of the directed graph, and the depth-first search algorithm is used to calculate the directed line path from the starting node to the end node to form the line path storage set F t + ; 9) After obtaining the line path storage set F t + Later, the path set F is further calculated t + The total split coefficient of each path in is calculated by the following steps: (1) Store the set F according to the line path t + , for a certain starting node i, read the node adjacency splitting coefficient matrix in turn The matrix flow splitting coefficient of each path from the starting node to the end node Multiply the above split coefficients to get the total split coefficient of the path In the formula, let i be the starting node, m and n be the path nodes, and s be the end node. is the power split coefficient flowing from node i to node j at time t, is the power split coefficient flowing from node j to node m at time t, is the power split coefficient from node m to node n at time t, is the power split coefficient from node n to node s at time t, is the total diversion coefficient of a single path from node i to node s at time t; (2) By reading the unit injection matrix The power injected by the kth unit into the starting node i Get the power distribution value of the power supply from the starting node i through j, m, n to the end node s The calculation formula is as follows: (3) Repeat steps (1) to (2) until the split coefficients of all paths and the power distribution values ​​of all power sources are calculated, and check the total split coefficients from each starting node to ensure that the sum of the total split coefficients of all paths starting from a certain starting node is 1. After the check is correct, the distribution matrix D of the power at time t is output. t , the power distribution matrix at time t is a K×N matrix, using Indicates that the matrix form is as follows: The distribution matrix D of power at time t t Each column is normalized to obtain the normalized power distribution matrix in: It is the sum of the power allocation values ​​injected into node s by all units k in the grid; The normalized matrix is ​​transposed and further compared with the unit carbon emission intensity matrix Multiply them together to calculate the carbon potential vector of the generator set and the node carbon potential matrix E t , realizing the electric carbon tracking at the power supply layer, grid layer and load layer; Right now Where: The node carbon potential matrix E at time t t The matrix is ​​an N×1-order matrix, which represents the electric carbon intensity of the grid node at time t.

2. The method for tracking electric carbon based on regional power grid power flow topology analysis according to claim 1, characterized in that: The specific process of simplification and modeling of step 1) is as follows: (1) The number of busbar nodes is determined by the number of busbars in the substation; (2) Determine the number of nodes for transformer modeling based on the number of transformer windings; (3) Each load line connecting the high, medium and low voltage busbars of the substation, with the end of each load line modeled as a node; (4) For the power source connected to the substation, each power source point should be modeled separately as a node; (5) For the branches between nodes, the branches between nodes are formed based on the interconnection current between the busbars, the current between the voltage windings of the transformer and the neutral point, the line transmission current, and the power injection current.

3. The electric carbon tracking method based on regional power grid power flow topology analysis according to claim 1 is characterized in that: Step 8) Use the depth-first search algorithm to calculate the line path storage set F t + The specific process is as follows: (1) Read the node power flow matrix The i-th row of The corresponding node j is included in the line path storage set F t + , mark the current node j as visited; (2) Read the node power flow matrix The jth row of The corresponding node k is included in the line path storage set F t + ; (3) Repeatedly select node k until the selected node In the example, the row where node k is located has only Then the node is a load node and no longer sends power outward. The node is included in the line path storage set F t + , output a complete power flow path starting from node i and mark node k as unvisited; (4) Save the power flow path starting from node i, mark node j as unvisited, and return to node i; (5) Repeat the above steps to complete the traversal of all nodes in (1), and then obtain the line path storage set F required by the algorithm. t + .

Images