A Complex Power System Fault Propagation Flow Calculation Method for Transmitting Vulnerability

Abstract

The present invention discloses a method for calculating the fault propagation flow of a complex power system that transmits vulnerability, and the steps include: S1: Obtain the historical data of the power system; S2: Construct a set of fault chains; S3: Establish a fault logic diagram; S4: Calculate the weights of the fault logic diagram; S5: Extract the fault logic sub-diagram; S6: Calculate the vulnerability of all nodes in the fault logic sub-diagram; S7: Calculate the vulnerability of all lines in the fault logic sub-diagram; This method can quickly calculate the degree of interaction between different line components. When one or more lines fail, it can quickly screen out the lines with high vulnerability and protect them to timely block the further propagation of the fault.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power and energy, and particularly to a method for calculating the fault propagation flow of a complex power system that transmits vulnerability. Background Art

[0002] With the continuous expansion of the scale of the power system, the power system has become an artificial network with high non-linearity and high complexity. At the same time, there is a highly coupled relationship between components in the complex power system. The occurrence of abnormal disturbances in one or more components may cause disturbances in the system, leading to subsequent failures of other components and triggering large-scale power outages. Therefore, analyzing the coupling relationship of components in the fault propagation state, calculating the probability of interaction between different components, and constructing a new interaction calculation method have important engineering value and important practical significance for preventing large-scale power outages.

[0003] In current cascade fault research, mainly by constructing a component fault chain database through a chain fault model, and then using the fault chain database to construct a fault propagation statistical graph to analyze the fault propagation characteristics between components. And the frequency of the cascade fault relationship between components appearing in the fault chain database is used as the weight of the fault traditional statistical graph to identify the cascade propagation relationship between key components.

[0004] However, the fault statistical graph only gives the fault propagation relationship between components, and does not give a method for measuring the possibility that one or more component failures cause other one or more component failures. Although traditional Markov chains and Bayes' formula can measure the possibility of cascade faults, they mainly focus on the measurement of the possibility of one-time propagation (that is, the possibility that a component failure directly causes other component failures), and do not give a calculation method for the possibility of multiple propagations (that is, the possibility that a component failure causes other component failures after multiple cascade propagations). In addition, the Markov chain model has a high computational complexity and is not suitable for evaluating complex power systems. Therefore, proposing a method for measuring the possibility of cascade fault propagation has important theoretical significance and engineering value. Summary of the Invention

[0005] To overcome the above technical deficiencies, the present invention proposes a method for calculating the fault propagation flow of a complex power system that transmits vulnerability. This method can quantitatively calculate the coupling propagation possibility of fault propagation between transmission lines with the help of a fault logic subgraph, and is suitable for calculating complex large-scale power systems.

[0006] The object of the present invention can be achieved by the following technical solutions:

[0007] A method for calculating the fault propagation flow of a complex power system that transmits vulnerability, the method specifically includes the following steps:

[0008] S1: Obtain the historical data of the power system, including the historical operating state parameters of the power system, topological connection parameters, and line electrical parameters;

[0009] S2: Randomly disconnect the initial branches to construct a set of fault chains;

[0010] S3: According to the set of fault chains, define the nodes and edges in the set of fault chains, convert all the fault chains in the set of fault chains into directed graphs, merge the same nodes and the same edges in the directed graphs to form a fault logic graph, and define the weights of the edges in the fault logic graph;

[0011] S4: According to the fault logic graph, construct a fault logic sub-graph of the target line set, calculate the vulnerability of each node by traversing each node in the fault logic graph, and analyze the influence of the target line set on other lines during fault propagation;

[0012] S5: According to the fault logic sub-graph, construct a transfer vulnerability model, including calculating the vulnerability of all nodes in the fault logic sub-graph and the vulnerability of all lines in the fault logic sub-graph.

[0013] Furthermore, constructing the set of fault chains specifically includes the following steps:

[0014] S21: In the initial operating state, randomly disconnect one or more lines in the power system, and set the disconnected line set as

[0015] S22: Re-distribute the system power flow. If the power system does not undergo islanding, calculate the active power of the transmission lines in the network according to the power flow calculation formula shown in Equation (1):

[0016]

[0017] In Equation (1), P km represents the active power between line k-m (the line between bus node k and bus node m), V k represents the voltage of bus node k, V m represents the voltage of bus node m, g km represents the conductance of line k-m, θ km represents the phase angle difference between bus node k and bus node m;

[0018] If the power system undergoes islanding and is divided into n islands, check whether the active power of each island i (i = 1, 2,..., n) is balanced, and calculate the total power generation of island i as shown in Equation (2)

[0019]

[0020] In formula (2), represents the total power generation, represents the power generation of bus node k, M represents the number of bus nodes. Among them, if there is no generator at bus node k or this bus node does not belong to the nodes of island i, it is regarded as

[0021] Calculate the total load power of island i as shown in formula (3)

[0022]

[0023] In formula (3), represents the total load power, represents the load power of bus node k. Among them, if there is no load at bus node k or this bus node does not belong to the nodes of island i, it is regarded as

[0024] S23: According to the total power generation of island i in S2 and the total load power of island i

[0025] Check whether the power is balanced; If

[0026] If then the power is unbalanced, calculate the change amount that the generator needs to adjust as shown in formula (4), and update the output power of the bus generator:

[0027]

[0028] In formula (4), represents the change amount of the generator adjustment of bus node k, represents the droop coefficient of the generator of bus node k. Among them, if there is no generator at bus node k or this bus node does not belong to the nodes of island i, it is regarded as The output power of the generator after adjustment is

[0029] If then the power is unbalanced, calculate the change amount that the generator needs to adjust as shown in formula (5), and update the load power of the bus node according to the equal proportion principle:

[0030]

[0031] In formula (5), represents the change amount of the load adjustment of load node k;

[0032] Through formulas (4) and (5), the power balance of island i is finally achieved;

[0033] Then, calculate the active power of the transmission line of island i according to formula (1);

[0034] S24: After obtaining the power of the transmission line after the power flow redistribution, determine whether the line is overloaded according to the transmission line capacity. If the power P of the transmission line k-m km is greater than or equal to its own capacity, it is determined that the transmission line k-m is overloaded, and obtain the set of overloaded lines in the first stage of the first fault chain

[0035] S25: Remove the set of overloaded lines from the power system, and according to steps S21 to S24, obtain the set of overloaded lines in the second stage of the first fault chain

[0036] S26: According to steps S21 to S25, obtain the sets of overloaded lines in the remaining stages of the first fault chain until there are no overloaded lines in the system, and obtain the fault chain represents the set of overloaded lines in the N1th stage of the first fault chain;

[0037] S27: Select different initial broken-line branches, repeat the process of obtaining all the sets of overloaded lines in the first fault chain, obtain the second fault chain to the Xth fault chain, form a fault chain set, and represent this fault chain set as:

[0038]

[0039] Furthermore, the definition of the nodes and edges in the fault chain set specifically includes: for each fault chain in the fault chain set , regard as nodes respectively Regard "→" as the weighted edge between nodes. Then, for it can be regarded as the edge E(V i , V j ).

[0040] Furthermore, the specific method for defining the weight of the edge in the fault logic graph is shown in formula (7):

[0041]

[0042] In formula (5), W(V i , V j ) represents the weight of the edge E(V i , V j ), and n ijIndicates the number of times the edge E(V i ,V j ) appears in the set of fault chains.

[0043] Furthermore, the specific steps for constructing the fault logic subgraph of the target line set include:

[0044] S51: Define the target line set as L W ={L1, L2,..., L T}}, and define the vulnerability of each line in the target line set as TV(L1), TV(L2),..., TV(L T );

[0045] S52: Find the set of target nodes V W ={L1, L2,..., L T} corresponding to the target line set L in the fault logic graph. Traverse each node in the fault logic graph. If the set of lines W that makes up the node V i is , then add the node V i to the set of target nodes V W ;

[0046] S53: Calculate the vulnerability degree of each node in the set of target nodes V W . Calculate the vulnerability TV(V i ) of the node V i as shown in formula (8):

[0047]

[0048] In formula (8), represents the element in the set of lines that makes up the node V i , N i represents the number of lines that make up the node V i , respectively represent 's vulnerability;

[0049] S54: Starting from the nodes in the node set V W , traverse each node in the fault logic graph. If the traversed node has no connected path with any node in the node set V W in the fault logic graph, then delete the node and the edges connected to the node, and form a fault logic subgraph starting from the node set V W .

[0050] Furthermore, the specific method for calculating the vulnerability of all nodes in the fault logic subgraph includes:

[0051] For node V i , if there is only one node V j connected to node V i by an edge, and the edge points from node V j to node V i , then the vulnerability of node V i is defined as shown in Equation (9):

[0052] TV(V i ) = W′(V j , V i ) × TV(V j ) (9)

[0053] In Equation (9), TV(V i ) and TV(V j ) respectively represent the vulnerability of node V i and node V j , and W′(V j , V i ) represents the weight of the modified edge E(V i , V j ). The weight of the modified edge W′(V j , V i ) is defined as shown in Equation (10):

[0054]

[0055] In Equation (10), represents the sum of the weights of all edges connected to node V j (these edges point from node V j to other nodes);

[0056] For node V i , if there are M i nodes connected to node V i by edges, and all of these edges point from these M i nodes to node V i , then the vulnerability of node V i is defined as shown in Equation (11). For

[0057]

[0058] Taking the nodes in the target node set V W as the starting points (note that the vulnerabilities of the nodes in the target node set V W have been calculated according to formula (8) and there is no need to calculate using formula (11)), based on formulas (9) - (11), gradually calculate the non - target node set V W in the fault logic sub - graphVulnerabilities of all nodes in

[0059] Furthermore, based on the vulnerabilities of all nodes in the fault logic subgraph, the vulnerability of any line L in the fault logic subgraph is defined as shown in Equation (12), and the vulnerabilities of all lines in the fault logic subgraph are calculated: a Vulnerability calculation of all lines in the fault logic subgraph:

[0060] TV(L a ) = max(TV(V1), TV(V2),..., TV(V Na )) (12)

[0061] In Equation (12), TV(L a ) represents the vulnerability of line L a , where V1, V2,..., V Na satisfy L a ∈ L(V1), L a ∈ L(V2), …, L a ∈ L(V Na ), and here N a represents the number of nodes containing line L a .

[0062] Compared with the prior art, the present invention has the following technical effects:

[0063] (1) This method can quickly calculate the influence degree of any branch or branch combination on other lines, with low computational complexity and is suitable for calculating complex large-scale power systems;

[0064] (2) When one or more lines fail, the lines with large vulnerabilities can be quickly screened out and protected according to the method proposed in this invention patent, and the further spread of the fault can be blocked in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is the overall flowchart of the present invention;

[0066] Figure 2 is a schematic diagram of converting a fault chain into a directed graph;

[0067] Figure 3 is a schematic diagram of converting three directed graphs into a fault logic graph;

[0068] Figure 4 is a schematic diagram of extracting a fault logic subgraph;

[0069] Figure 5 is a schematic diagram of calculating transfer vulnerability. DETAILED DESCRIPTION OF THE INVENTION

[0070] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts, any modifications, equivalent replacements, improvements, etc., shall be included in the protection scope of the present invention.

[0071] A method for calculating the fault propagation flow of a complex power system that transmits vulnerability is as Figure 1 shown. The method specifically includes the following steps:

[0072] S1: Obtain the historical data of the power system, including the historical operating state parameters, topological connection parameters, and line electrical parameters of the power system;

[0073] S2: Randomly disconnect the initial branch and construct a fault chain set;

[0074] S3: According to the fault chain set, define the nodes and edges in the fault chain set, convert all the fault chains in the fault chain set into a directed graph, merge the same nodes and the same edges in the directed graph to form a fault logic graph, and define the weights of the edges in the fault logic graph;

[0075] S4: According to the fault logic graph, construct a fault logic subgraph of the target line set, calculate the vulnerability of the nodes by traversing each node in the fault logic graph, and analyze the influence of the target line set on other lines during fault propagation;

[0076] S5: According to the fault logic subgraph, construct a vulnerability transfer model, including calculating the vulnerability of all nodes in the fault logic subgraph and the vulnerability of all lines in the fault logic subgraph.

[0077] The construction of the fault chain set specifically includes the following steps:

[0078] S21: In the initial operating state, randomly disconnect one or more lines in the power system, and the set of disconnected lines is set as

[0079] S22: Re-distribute the system power flow. If the power system does not undergo disconnection, calculate the active power of the transmission lines in the network according to the power flow calculation formula shown in Equation (1):

[0080]

[0081] In Equation (1), P km represents the active power between line k-m (the line between bus node k and bus node m), V kDenote the voltage of bus node k, V m Denote the voltage of bus node m, g km Denote the conductance of line k-m, θ km Denote the phase angle difference between bus node k and bus node m;

[0082] If the power system is islanded and divided into n islands, check whether the active power of each island i (i = 1, 2,..., n) is balanced, and calculate the total generation power of island i as shown in Equation (2)

[0083]

[0084] In Equation (2), Denote the total generation power, Denote the generation power of bus node k, M denotes the number of bus nodes, where if bus node k has no generator or this bus node does not belong to the nodes of island i, then it is regarded as

[0085] Calculate the total load power of island i as shown in Equation (3)

[0086]

[0087] In Equation (3), Denote the total load power, Denote the load power of bus node k, where if bus node k has no load or this bus node does not belong to the nodes of island i, then it is regarded as

[0088] S23: According to the total generation power of island i in S2 And the total load power of island i Check whether the power is balanced;

[0089] If Then the power is balanced, and calculate the active power of the transmission line of island i as shown in Formula (1);

[0090] If Then the power is unbalanced, and calculate the change amount that the generator needs to adjust as shown in Formula (4), and update the output power of the bus generator:

[0091]

[0092] In Equation (4), Denote the change amount of the generator adjustment of bus node k, Denote the droop coefficient of the generator at bus node k. Among them, if there is no generator at bus node k or this bus node does not belong to the nodes of island i, it is regarded as The adjusted output power of the generator is

[0093] If Then there is a power imbalance. Calculate the change amount that the generator needs to adjust as shown in formula (5), and update the load power of the bus node according to the equal proportion principle:

[0094]

[0095] In formula (5), Denote the change amount of load regulation at load node k;

[0096] Through formulas (4) and (5), finally achieve the power balance of island i;

[0097] Then calculate the active power of the transmission line of island i according to formula (1);

[0098] S24: After obtaining the power of the transmission line after the power flow reallocation, judge whether the line is overloaded according to the transmission line capacity. If the power P of the transmission line k-m km is greater than or equal to its own capacity, judge that the transmission line k-m is overloaded, and obtain the set of overloaded lines in the first stage of the first fault chain

[0099] S25: Remove the set of overloaded lines from the power system, and obtain the set of overloaded lines in the second stage of the first fault chain according to steps S21~S24

[0100] S26: According to steps S21~S25, obtain the sets of overloaded lines in the remaining stages of the first fault chain until there are no overloaded lines in the system, and obtain the fault chain Denote the set of overloaded lines in the N1th stage of the first fault chain;

[0101] S27: Select different initial broken branches, repeat the process of obtaining the sets of all overloaded lines in the first fault chain, obtain the second fault chain~the Xth fault chain, form a fault chain set, and represent this fault chain set as:

[0102]

[0103] Define the nodes and edges in the fault chain set, specifically including: for each fault chain in the fault chain set in, Respectively regarded as nodes Regarding "→" as a weighted edge between nodes, then for It can be regarded as edge E(V i ,V j ).

[0104] Such as Figure 2 shown, all the fault chains in the fault chain set are converted into directed graphs.

[0105] The specific method for defining the weight of the edge in the fault logic graph is shown in Equation (7):

[0106]

[0107] In Equation (5), W(V i ,V j ) represents the weight of edge E(V i ,V j ), and n ij represents the number of times edge E(V i ,V j ) appears in the fault chain set.

[0108] The specific steps for constructing the fault logic sub-graph of the target line set include:[[]]

[0109] S51: Define the target line set as L W ={L1, L2,..., L T}, and define the vulnerability of each line in the target line set as TV(L1), TV(L2),..., TV(L T );

[0110] S52: Find the target node set V W ={L1, L2,..., L T} corresponding to the target line set in the fault logic graph. Traverse each node in the fault logic graph. If the line set W that makes up node V i then add node V to the target node set V i ; W in;

[0111] S53: Calculate the vulnerability degree of each node in the target node set V W . Calculate the vulnerability TV(V i ) of node V i as shown in formula (8):

[0112]

[0113] In formula (8), Represents an element in the set of lines that make up node V i , and N i Represents the number of lines that make up node V i , Respectively represent the vulnerability;

[0114] S54: Starting from the nodes in the node set V W , traverse each node in the fault logic diagram. If there is no connected path in the fault logic diagram between the traversed node and any one of the nodes in the node set V W , then delete the node and the edges connected to this node, and form a fault logic sub-diagram starting from the node set V W .

[0115] As Figure 3 shown, it is a schematic diagram of converting a three-directed graph into a fault logic diagram.

[0116] The specific method for calculating the vulnerability of all nodes in the fault logic sub-diagram includes:

[0117] For node V i , if there is only one node V j connected to node V i by an edge, and the edge points from node V j to node V i , then the vulnerability of node V i is defined as shown in Equation (9):

[0118] TV(V i ) = W′(V j , V i ) × TV(V j ) (9)

[0119] In Equation (9), TV(V i ) and TV(V j ) respectively represent the vulnerability of node V i and node V j , and W′(V j , V i ) represents the weight of the corrected edge E(V i , V j ). The corrected edge weight W′(V j , V i ) is defined as shown in Equation (10):

[0120]

[0121] In Equation (10), represents the node connected to node V jThe sum of the weights of all the connected edges (these edges are pointed from node V j to other nodes);

[0122] For node V i , if there are M i nodes connected to node V i by edges, and all these M i nodes point to node V i , then the vulnerability of node V i is defined as shown in Equation (11). For:

[0123]

[0124] Taking the nodes in the target node set V W as the starting points (it should be noted that the vulnerabilities of the nodes in the target node set V W have been calculated according to Equation (8) and there is no need to calculate using Equation (11)), according to Equations (9) - (11), gradually calculate the vulnerabilities of all the nodes in the non-target node set V W in the fault logic subgraph.

[0125] According to the vulnerabilities of all the nodes in the fault logic subgraph, define the vulnerability of any line L a in the fault logic subgraph as shown in Equation (12), and calculate the vulnerabilities of all the lines in the fault logic subgraph:

[0126] TV(L a ) = max(TV(V1), TV(V2),..., TV(V Na )) (12)

[0127] In Equation (12), TV(L a ) represents the vulnerability of line L a , where V1, V2,..., V Na satisfy L a ∈L(V1), L a ∈L(V2), …, L a ∈L(V Na ), and here N a represents the number of nodes containing line L a .

[0128] Such as Figure 4As shown in the figure, it is a schematic diagram for extracting the fault logic subgraph. Taking the node where Line 1 is located as the starting point as an example, the fault logic subgraph is extracted. Traverse all the nodes in the fault logic graph to check if there is a connected path to the node where Line 1 is located. For example, there are connected paths 1→2, 7→3, 9 between the node where Line 1 is located and the nodes where Line Combinations 3 and 9 are located. Therefore, the nodes where Line Combinations 3 and 9 are located are retained. There is no connected path between the node where Line 1 is located and the node where Line 5 is located, so the node where Line 5 is located is deleted. And all the edges connected to the node where Line 5 is located are deleted, including the edge between the node where Line 1 is located and the node where Line 5 is located, the edge between the node where Line 7 is located and the node where Line 5 is located, the edge between the node where Line 7 is located and the node where Line 5 is located, and the edge between the nodes where Line Combinations 2 and 7 are located and the node where Line 5 is located. Through the above operations, the fault logic subgraph of the node where the target Line 1 is located is finally obtained.

[0129] As Figure 5 shown, analyze the influence degree of the node where Line 1 is located on other components during the fault propagation process. In the figure, the weight of each edge is given, and in practice, it can be calculated according to formula (7). Assume the vulnerability of the node where Line 1 is located is TV(1) = 1.

[0130] First, correct the weights of the fault logic subgraph according to formula (10). Since there are two edges from the nodes where Line Combinations 3 and 9 are located to other nodes, namely Edge 3 and Edge 4. The weights of Edge 3 and Edge 4 are corrected to 0.54 and 0.46. Similarly, the weights of Edge 5 and Edge 6 are corrected to 0.43 and 0.57.

[0131] Furthermore, solve the vulnerability of each node according to the corrected weights.

[0132] Since there is only one edge pointing to the nodes where Line Combinations 2 and 7 are located, therefore, according to formula (9), the vulnerability of the nodes where Line Combinations 2 and 7 are located can be calculated as TV(2,7) = 0.9×TV(1) = 0.9×1 = 0.9.

[0133] Since there is only one edge pointing to the nodes where Line Combinations 3 and 9 are located, therefore, according to formula (9), the vulnerability of the nodes where Line Combinations 3 and 9 are located can be calculated as TV(3,9) = 0.9×TV(2,7) = 0.8×9 = 0.72.

[0134] Since there is only one edge pointing to the node where Line 8 is located, therefore, according to formula (9), the vulnerability of the node where Line 8 is located can be calculated as TV(8) = 0.7×TV(3,9) = 0.54×0.72 = 0.39.

[0135] Since there is only one edge pointing to the node where line 7 is located, according to formula (9), the vulnerability of the node where line 7 is located can be calculated as TV(7) = 0.46×TV(3,9) = 0.46×0.72 = 0.33.

[0136] Since there is only one edge pointing to the node where line 6 is located, according to formula (9), the vulnerability of the node where line 6 is located can be calculated as TV(6) = 0.43×TV(7) = 0.43×0.33 = 0.14.

[0137] Since there are two edges pointing to the node where line 9 is located, according to formula (9) and (11), the vulnerability of the node where line 9 is located can be calculated as:

[0138] TV(9) = 1 - (1 - 0.43×TV(6))(1 - 0.57×TV(7))

[0139] = 1 - (1 - 0.43×0.14)(1 - 0.57×0.33) = 0.68

[0140] Finally, according to formula (12), the vulnerabilities of lines 2, 3, 6, 7, 8, and 9 are calculated to be 0.9, 0.72, 0.14, 0.9, 0.39, and 0.68 respectively. Judging from the results, the vulnerability values of lines 2 and 7 are the highest, so they need to be protected with priority. From the above reasoning process, the transfer vulnerability model constructed in this paper has simple reasoning and does not require complex calculation processes, and is very suitable for application to the calculation of fault propagation in large-scale complex power grids.

Claims

1. A method for calculating the fault propagation flow of a complex power system that transmits vulnerability, characterized in that The method specifically includes the following steps: S1: Obtain the historical data of the power system, including the historical operating state parameters, topological connection parameters, and line electrical parameters of the power system; S2: Randomly disconnect the initial branch and construct a set of fault chains; S3: According to the set of fault chains, define the nodes and edges in the set of fault chains, convert all the fault chains in the set of fault chains into directed graphs, merge the same nodes and the same edges in the directed graphs to form a fault logic graph, and define the weights of the edges in the fault logic graph; S4: According to the fault logic graph, construct a fault logic sub-graph of the target line set, calculate the vulnerability of each node by traversing each node in the fault logic graph, and analyze the influence of the target line set on other lines during fault propagation; The specific steps for constructing the fault logic sub-graph of the target line set include: S41: Define the target line set as , and define the vulnerability of each line in the target line set as ; S42: Search for the set of target lines The set of target nodes corresponding to in the fault logic diagram , traverse each node in the fault logic diagram. If the set of lines that make up the node , then add the node to the set of target nodes ; S43: Calculate the vulnerability degree of each node in the target node set as shown in formula (1), calculate the node vulnerability degree : ， （1） In formula (1), represents an element in the set of lines that make up the node represents the number of lines that make up the node respectively represent the vulnerability of S44: Starting from the nodes in the node set , traverse each node in the fault logic diagram. If there is no connection path in the fault logic diagram between the traversed node and any node in the node set , then delete the node and the edges connected to this node, and form a fault logic sub-diagram starting from the node set ; S5: According to the fault logic sub-graph, construct a transfer vulnerability model, including calculating the vulnerability of all nodes in the fault logic sub-graph and the vulnerability of all lines in the fault logic sub-graph.

2. The method for calculating the fault propagation flow of a complex power system for transmitting vulnerability according to claim 1, characterized in that The specific steps for constructing the set of fault chains include: S21: In the initial operating state, randomly disconnect one or more lines in the power system, and the set of disconnected lines is set to ; S22: Re-distribute the system power flow. If the power system does not undergo islanding, calculate the active power of the transmission lines in the network according to the power flow calculation formula shown in Equation (2): （2） In formula (2), represents the active power between lines ; represents the voltage of bus node ; represents the voltage of bus node ; represents the conductance of line ; represents the phase angle difference between bus node and bus node ; If the power system is islanded and divided into isolated islands, check whether the active power of each isolated island is balanced. Among them, , calculate the total power generation of the isolated island as shown in Equation (3): ​ （3） In formula (3), represents the total power generation, represents the power generation of bus node , represents the number of bus nodes. Among them, if bus node has no generator or this bus node does not belong to the nodes of the island , it is regarded as ; Calculate the total load power of the island as shown in Equation (4). :​ （4） In formula (4), represents the total load power, represents the load power of bus node , where if the bus node has no load or the bus node does not belong to the nodes of the island , it is regarded as ; S23: According to the total power generation of the island in S2 and the total load power of the island to check whether the power is balanced; ​​ If , calculate the active power of the island transmission line as shown in formula (2) ; If , calculate the change amount that the generator needs to adjust as shown in formula (5), and update the output power of the bus generator: （5） In formula (5), represents the change in generator regulation at the bus node , and represents the droop coefficient of the generator at the bus node . If there is no generator at the bus node or the bus node does not belong to the island node, it is regarded as , and the output power of the generator after adjustment is ; ​ If , calculate the change amount that the generator needs to adjust as shown in formula (6), and update the load power of the bus node according to the equal proportion principle: （6） In formula (6), represents the load node the change in load regulation; Through formulas (5) and (6), islanding is finally achieved to achieve power balance; Then calculate the island according to formula (2). The active power of the transmission line S24: After obtaining the power of the transmission line after the power flow redistribution, determine whether the line is overloaded according to the transmission line capacity. If the power of the transmission line is greater than or equal to its own capacity, determine that the transmission line is overloaded, and obtain the set of overloaded lines in the first stage of the first fault chain ; S25: Remove the overloaded line set from the power system, and according to steps S21 - S24, obtain the overloaded line set in the second stage of the first fault chain ; S26: According to steps S21 - S25, obtain the set of overloaded lines in the remaining stages of the first fault chain until there are no overloaded lines in the system, and obtain the fault chain … , represents the set of overloaded lines in the th stage of the first fault chain; S27: Select different initial disconnected branches and repeat the process of obtaining all the overloaded line sets in the first fault chain to obtain the second fault chain to the th fault chain, form a fault chain set, and represent this fault chain set as: （7）。 3. The method for calculating the fault propagation flow of a complex power system for transmitting vulnerability according to claim 2, characterized in that The nodes and edges in the defined fault chain set specifically include: For each fault chain in the fault chain set , , ,…, are respectively regarded as nodes , ,…, , and the " " is regarded as the weighted edge between nodes. Then for it can be regarded as the edge .

4. The method for calculating the fault propagation flow of a complex power system for transmitting vulnerability according to claim 3, characterized in that The specific method for defining the weights of the edges in the fault logic graph is shown in Equation (8): （8） In formula (8), represents the weight of edge , represents the number of times edge appears in the failure chain set.

5. The method for calculating the fault propagation flow of a complex power system that transmits vulnerability according to claim 4, characterized in that, The specific method for calculating the vulnerability of all nodes in the fault logic sub-graph includes: For a node , if there is only one node connected to node by an edge, and the edge points from node to node , then the vulnerability of node is defined as shown in Equation (9): （9） In Equation (9), and respectively represent the vulnerability of node and node . represents the weight of the corrected edge . The weight of the corrected edge is defined as shown in Equation (10) : （10） In formula (10), represents the sum of the weights of all edges connected to node , and the edges connected to node are the edges pointing from node to other nodes; For a node , if there are nodes connected to the node by edges, and all of them point to the node from these nodes, then the vulnerability of the node is defined as shown in Equation (11). For: （11） Starting from the nodes in the target node set and according to formulas (9) to (11), gradually calculate the vulnerability of all nodes in the non-target node set in the fault logic subgraph.

6. The method for calculating the fault propagation flow of a complex power system for transmitting vulnerability according to claim 5, characterized in that Define the vulnerability of any line in the fault logic subgraph according to the vulnerability of all nodes in the fault logic subgraph, as shown in Equation (12). Calculate the vulnerability of all lines in the fault logic subgraph: （12） In formula (12), represents the vulnerability of the line , where satisfies , , …, , where represents the number of nodes including the line .

Images