Topological power grid operation-based multi-level vulnerability quantitative index construction method

Abstract

The invention relates to the technical field of power grid operation, and discloses a topological power grid operation-based multi-level vulnerability quantitative index construction method, which comprises the following steps of: obtaining topological structure data of a power grid dispatching system to perform hierarchical division, analyzing node data to form a hierarchical matrix, calculating a shortest path length matrix and analyzing betweenness centrality. According to the method, a power grid is divided into a high-voltage level, a medium-voltage level and a low-voltage level according to voltage levels, so that the topological structure of each level is independently analyzed, the modularity increment change of each node moving to an adjacent node is calculated based on a Louvain algorithm, subgroup junctions are gradually optimized by maximizing the modularity increment, and therefore the topological structure of each node is optimized. The hierarchical modularity is combined with the node vulnerability index of the hierarchy to form a more comprehensive vulnerability evaluation model, and the inter-hierarchy coupling problem can be more effectively identified and solved in the power grid optimization process through the comprehensive analysis of the inter-hierarchy and intra-hierarchy vulnerability indexes.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation, and in particular to a method for constructing multi-level vulnerability quantitative indicators based on topological power grid operation. Background Art

[0002] In order to ensure the safety and stability of the power grid, traditional power grid vulnerability analysis usually focuses on global evaluation, such as the importance of nodes, network connectivity, etc. The above methods can usually accurately reflect the manifestation of different nodes and network connectivity in the vulnerability of the power grid. However, with the expansion of the scale of the power grid and the complexity of the operation, the traditional methods are not convenient for the complex correlation between different voltage levels and the potential risks within each level.

[0003] Existing vulnerability quantification methods often cannot timely reflect the dynamic changes of multi-level power grid nodes in actual operation, and ignore the impact of dynamic changes of power grids in actual operation on potential risks across levels. They fail to effectively combine power grid vulnerability indicators for comprehensive assessment, reducing the efficiency in responding to power grid disturbances. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the existing vulnerability quantification methods often cannot be based on the dynamic changes of multi-level power grid nodes in actual operation in a timely manner, and ignore the impact of the dynamic changes of the power grid in actual operation on the potential risks across levels, and fail to effectively combine power grid vulnerability indicators for comprehensive evaluation, thereby reducing the efficiency in responding to power grid disturbances.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for constructing a multi-level vulnerability quantitative index based on topological power grid operation, comprising: collecting topological structure data of the power grid, the topological structure data including node data and edge data, forming a hierarchical matrix according to the node data, and analyzing the correlation between nodes at each level;

[0007] Preliminarily divide the node into subgroups, calculate the modularity of different levels and across levels, calculate the incremental change of modularity of node movement based on Louvain algorithm, and perform repeated iterations to determine the subgroup structure, determine the hierarchical vulnerability index based on node data, and calculate the quantitative index of multi-level vulnerability of the power grid;

[0008] Draw the vulnerability index curve, set up the early warning mechanism, analyze the grid stability according to the grid fluctuation amplitude, frequency deviation and power stability, expand the topology branch with the hierarchical vulnerability index to determine the high-stability topology structure, verify the high-stability topology structure and record the data.

[0009] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the formation of a hierarchical matrix based on node data and the analysis of the correlation between nodes at each level include preprocessing the topological structure data and constructing an adjacency matrix of the power grid based on whether there are edges between nodes. ;

[0010] Count the number of nodes directly connected to node i as the degree of the node ;

[0011] The grid is divided into three levels based on the voltage level: high voltage, medium voltage and low voltage.

[0012] The high voltage level includes all substations and transmission lines of 220kV and above, the medium voltage level includes all substations and transmission lines between 35kV and 220kV, and the low voltage level includes all distribution substations and distribution lines of 35kV and below;

[0013] The voltage levels of nodes based on the adjacency matrix are allocated according to the high voltage level, medium voltage level and low voltage level to form a hierarchical matrix , the level matrix Includes high voltage level matrix , medium voltage level matrix and low voltage level matrix ;

[0014] For each level matrix , use the Floyd-Warshall algorithm to calculate the shortest path length matrix between any two nodes ;

[0015] Topological distance matrix for each level Perform analysis and calculate the betweenness centrality value of each node;

[0016] Based on whether two nodes belong to different levels and are directly connected, analyze the correlation between nodes at each level and establish a cross-level connection relationship matrix .

[0017] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the preliminary division of node subgroups, calculation of different levels and cross-level modularity, calculation of the incremental change of modularity of node movement based on Louvain algorithm, and repeated iteration to determine the subgroup structure include, based on the hierarchical matrix If two nodes belong to the same level, they are considered as subgroups. For each node , by calculating the shortest path length matrix , calculate the average path length of all nodes as the length threshold, and divide the two nodes whose shortest path length is less than the length threshold into the same subgroup;

[0018] The modularity within the level is calculated based on the adjacency matrix of each level of the power grid, expressed as:

[0019]

[0020] in, Indicates the modularity of the corresponding power grid level, Represents the total number of edges in the network at this level, Representation Node The degree of is the indicator function, Representation Node The degree of and They represent the subgroups to which the nodes belong respectively;

[0021] At the same time, the cross-level modularity is calculated and expressed as:

[0022]

[0023] in, represents cross-level modularity;

[0024] Based on the Louvain algorithm, the network structure is constructed according to the divided subsets, and the incremental change of the modularity of each node when it moves to the adjacent node is calculated. , expressed as:

[0025]

[0026]

[0027] in, represents the total number of edges across levels, Representation Node and Direct connection between represents the total number of connected edges, It indicates the change in the internal connection of the atomic group after node i is removed from the atomic group. Represents node i and node The connection status, and Respectively represent nodes and nodes The degree of

[0028] For each node , determine the maximum The largest neighboring node , and the node Move to In the same subgroup, the subgroup structure is merged and recalculated;

[0029] Repeat the iteration to merge and recalculate the subgroup structure. Based on the shortest path length matrix value between nodes, calculate the standard deviation and multiply it with the sensitivity constant as the subgroup threshold.

[0030] Real-time monitoring of modularity increments in each iteration , and in When it is less than the subgroup threshold, the iteration stops;

[0031] Based on the iterated subgroup structure, the modularity within each level and the cross-level modularity are recalculated and normalized.

[0032] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the determining of the level vulnerability index based on the node data and the calculation of the multi-level vulnerability quantitative index of the power grid include constructing a node vulnerability index based on the node and counting the betweenness centrality value and the node degree of the node;

[0033] Calculate the average distance between a node and other nodes based on the shortest path length matrix, and use the average distance as the node closeness;

[0034] The inverse of the maximum fluctuation amplitude of the node voltage is used as the voltage stability index of each node;

[0035] The sum of the comprehensive betweenness centrality value, node degree, node closeness and voltage stability index is used as the node vulnerability index of the node;

[0036] The modularity of each level is divided by the total number of subgroups at the corresponding level, and the vulnerability index of each level is calculated by considering the vulnerability index of the level node, which is expressed as:

[0037] ;

[0038] in, Represents the vulnerability index of each level, represents the modularity of each level, Represents the node set of the corresponding level, represents the node vulnerability index of node i, Indicates the number of nodes at the corresponding level;

[0039] The cross-level modularity is divided by the total number of node connections across different voltage levels to obtain the cross-level vulnerability index. The vulnerability index of each level and the cross-level vulnerability index are weighted and comprehensively calculated to form the final multi-level vulnerability quantitative index of the power grid.

[0040] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the drawing of the vulnerability index curve and the setting of the early warning mechanism include drawing the vulnerability index curve based on real-time collection of operation data during the operation of the power grid, and the operation data include voltage, power, frequency and load data;

[0041] Preprocess the collected data and use a median filter to eliminate noise in the data;

[0042] Calculate the node vulnerability index based on the real-time collected operation data , Vulnerability indicators at the corresponding level , based on MATLAB tools, draw curve graphs for visual display;

[0043] According to historical data and grid operation requirements, the sum of the mean value of the node's historical vulnerability index and twice the standard deviation of the historical vulnerability index is calculated as the node vulnerability threshold;

[0044] The sum of the mean of the historical vulnerability index at different levels and twice the standard deviation of the historical vulnerability index is calculated as the level vulnerability threshold;

[0045] If the node vulnerability index If the node vulnerability threshold is greater than the node vulnerability threshold, the node warning alarm is triggered. If the vulnerability indicators at different levels If it is greater than the layer vulnerability threshold, a layer warning alarm is triggered.

[0046] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the analysis of power grid stability based on power grid fluctuation amplitude, frequency deviation and power stability includes, based on the maximum voltage fluctuation amplitude value, comprehensively calculating the average inverse value of voltage fluctuation of all nodes as the voltage stability value;

[0047] The frequency deviation value is calculated based on the grid frequency, and the absolute value of the frequency deviation is averaged to represent the frequency stability, which is expressed as:

[0048]

[0049] in, Indicates time The frequency deviation, Represents the total number of sampling points, Indicates frequency stability;

[0050] Based on the power output and total power demand of each node, the power stability is calculated and expressed as:

[0051]

[0052] in, Indicates power stability, represents the total power output of the node, represents the total power demand of the system to which the node belongs, Indicates the amount of power imbalance;

[0053] The comprehensive voltage stability value, frequency stability value and power stability value are normalized and added together as the stability value of the power grid.

[0054] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the expansion of topological branches in cooperation with the hierarchical vulnerability index to determine a high-stability topological structure includes: based on the early warning alarm of the node vulnerability threshold and the hierarchical vulnerability threshold, confirming the nodes and corresponding subgroups that exceed the standard, visually displaying through the branch and bound algorithm and MATLAB picture tools based on the topological structure data, generating a branch topological structure by deleting the connection edges of the high vulnerability index nodes, and repeatedly expanding the branches;

[0055] Calculate the corresponding hierarchical vulnerability index for each new topology until the vulnerability index of the hierarchy in the new topology is lower than the hierarchical vulnerability threshold and the grid stability continues to increase, then continue to expand the branch, and stop expanding the branch when the grid stability stops increasing;

[0056] The boundary assessment mechanism is used in conjunction with the power grid simulation tool PSS / E to simulate and verify the scenario of each generated branch. The branches whose adjustment results cannot meet the requirements for reducing the vulnerability index are pruned, and the topology with a low vulnerability index and the highest power grid stability is selected as the high-stability structure.

[0057] As a preferred solution of the method for constructing a multi-level vulnerability quantitative index based on topological power grid operation described in the present invention, wherein: the verification of the high-stability topological structure and data recording includes applying the optimal topological structure to the power grid, starting the power grid operation and collecting key indicator data in real time, and calculating the adjusted hierarchical vulnerability index and power grid stability value in real time;

[0058] Compare the hierarchical vulnerability index and the grid stability value before and after the adjustment. If the change value of the hierarchical vulnerability index is less than or equal to 0, and the change value of the grid stability value is less than or equal to 0, it indicates an optimization error.

[0059] Summarize the changes in vulnerability indicators and stability indicators in all tests and create a comprehensive report for recording and preservation.

[0060] A computer device comprises: a memory and a processor; the memory stores a computer program, comprising: the steps of implementing any one of the methods of the present invention when the processor executes the computer program.

[0061] A computer-readable storage medium stores a computer program, comprising: when the computer program is executed by a processor, the steps of implementing any one of the methods of the present invention are implemented.

[0062] Beneficial effects of the present invention: The present invention divides the power grid into high-voltage, medium-voltage and low-voltage layers according to voltage levels, so that the topological structure of each layer is analyzed independently. Based on the Louvain algorithm, the incremental change of modularity of each node when it moves to the adjacent node is calculated, and the sub-group nodes are gradually optimized by maximizing the modularity increment. The modularity of the layer is combined with the node vulnerability index of the layer to form a more comprehensive vulnerability assessment model. Through the comprehensive analysis of cross-layer and intra-layer vulnerability indicators, the coupling problem between layers can be more effectively identified and solved during the power grid optimization process. Through the branch and bound algorithm, the connection edges of nodes with high vulnerability indicators are gradually deleted to generate different branch topological structures, optimize the topological structure, and achieve the balance and improvement of multiple important performance indicators of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0064] Figure 1 An overall flow chart of a method for constructing a multi-level vulnerability quantitative index based on topological power grid operation provided by an embodiment of the present invention;

[0065] Figure 2 A schematic diagram of a vulnerability quantification index flow chart of a method for constructing a multi-level vulnerability quantification index based on topological power grid operation according to an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of a topology optimization process for a method for constructing multi-level vulnerability quantitative indicators of topological power grid operation provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0067] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0068] Example 1, reference Figure 1-Figure 3 , which is an embodiment of the present invention, provides a method for constructing a multi-level vulnerability quantitative index based on topological power grid operation, including:

[0069] S1: Obtain the topological structure data of the power grid dispatching system for hierarchical division, analyze the node data to form a hierarchical matrix, calculate the shortest path length matrix and analyze the betweenness centrality, and analyze the correlation between nodes at each level.

[0070] Preferably, the topological structure data of the power grid dispatching system is obtained for hierarchical division, the node data is analyzed to form a hierarchical matrix, the shortest path length matrix is ​​calculated and the betweenness centrality is analyzed, and the correlation between nodes at each level is analyzed, including:

[0071] The topological structure data of the power grid is collected based on the power grid dispatching system. The data includes information on nodes, such as substations and distribution stations, and edges, such as transmission lines and distribution lines.

[0072] The acquired data is preprocessed, and the adjacency matrix A of the power grid is constructed based on whether there are edges between nodes, where Indicates that there is an edge between nodes i and j. Indicates that there is no edge connection;

[0073] Count the number of nodes directly connected to node i as the degree of the node ;

[0074] The grid is divided into three levels based on the voltage level, including high voltage level, medium voltage level and low voltage level;

[0075] The high-voltage level includes all substations and transmission lines of 220kV and above, the medium-voltage level includes all substations and transmission lines between 35kV and 220kV, and the low-voltage level includes all distribution stations and distribution lines of 35kV and below.

[0076] The voltage levels of nodes based on the adjacency matrix are allocated according to the high voltage level, medium voltage level and low voltage level to form a hierarchical matrix ,include High-voltage hierarchical matrix, Medium Voltage Tier Matrix and Low voltage level matrix, , and Respectively expressed as:

[0077]

[0078]

[0079]

[0080] For each level matrix , use the Floyd-Warshall algorithm to calculate the shortest path length matrix between any two nodes , expressed as:

[0081]

[0082] in, represents the shortest path length between nodes i and j in this level, and The path length from node i to intermediate node b, and the path length from intermediate node b to node j, respectively. It means that the above formula should be applied to all pairs of nodes i and j in the network, as well as all possible intermediate nodes b.

[0083] Topological distance matrix for each level Perform analysis and calculate the betweenness centrality value of each node to evaluate the importance of the node in the hierarchical network, expressed as:

[0084]

[0085] in, represents the number of shortest paths from node s to node t, represents the number of shortest paths from node s to node t through node i, represents the betweenness centrality value of node i, It means to sum all different pairs of nodes s and t, where s and t represent any two different nodes, and s is not equal to i, and t is not equal to i.

[0086] Based on whether two nodes belong to different levels and are directly connected, analyze the correlation between nodes at each level and establish a cross-level connection relationship matrix , expressed as:

[0087]

[0088] The construction of the adjacency matrix provides a basis for subsequent calculations, defines the topological structure of the power grid, and provides necessary input data for calculating the shortest path length and betweenness centrality between nodes. The shortest path length matrix is ​​calculated to store the shortest path length between any two nodes in the network, which directly affects the accuracy of the betweenness centrality. Without this matrix, the connectivity between nodes cannot be effectively measured, and the importance of nodes cannot be accurately evaluated. The calculation of the betweenness centrality value indicates the proportion of the number of times node i acts as an intermediary in the shortest path between all node pairs. The importance of a node in the network is evaluated by counting the frequency of the node as an intermediary node, which provides a basis for identifying key nodes in the power grid. The betweenness centrality is a key indicator for quantifying vulnerability, which can identify nodes with high intermediary effects in the power grid. Once these nodes fail, they will have a significant impact on the connectivity of the network. The analysis and judgment of the connection relationship matrix is ​​helpful to analyze the overall connectivity of the power grid, especially in the case of cross-level, to identify the dependencies between different levels.

[0089] By dividing the power grid into high-voltage, medium-voltage and low-voltage layers according to voltage levels, and constructing independent hierarchical matrices (high-voltage, medium-voltage and low-voltage matrices), the topological structure of each layer is analyzed independently. Unlike traditional power grid topology analysis methods, this hierarchical modeling method can analyze the relationship between layers more carefully, especially the cross-layer correlation. The Floyd-Warshall algorithm is usually used in the global shortest path calculation of graphs, but here it is applied to the analysis of hierarchical matrices, which can refine the network structure of each layer and identify potential weaknesses within and across layers.

[0090] S2: Preliminarily divide the node into subgroups, calculate the modularity of different levels and across levels, calculate the incremental change of modularity of node movement based on the Louvain algorithm, and perform repeated iterations to determine the subgroup structure, determine the hierarchical vulnerability index based on the node data, analyze the cross-level vulnerability index, and calculate the quantitative index of multi-level vulnerability of the power grid.

[0091] Preferably, the node subgroups are initially divided, the modularity of different levels and across levels is calculated, the incremental change of modularity of node movement is calculated based on the Louvain algorithm, and repeated iterations are performed to determine the subgroup structure, including:

[0092] Based on the hierarchy matrix If two nodes belong to the same level, they are considered as subgroups. For each node i, the shortest path length matrix is ​​calculated , calculate the average path length of all nodes as the length threshold, and divide the two nodes whose shortest path length is less than the length threshold into the same subgroup;

[0093] The modularity within each level (high voltage, medium voltage, low voltage) is calculated based on the adjacency matrix of the power grid, expressed as:

[0094]

[0095]

[0096]

[0097] in, Indicates the modularity of the corresponding power grid level, Represents the total number of edges in the network at this level, represents the degree of node i, is an indicator function. If nodes i and j belong to the same subgroup, it means 1, otherwise it means 0. It means to sum all pairs of nodes in the level. represents the degree of node j, and They represent the subgroups to which the nodes belong.

[0098] At the same time, the cross-level modularity is calculated and expressed as:

[0099]

[0100] in, Represents the total number of cross-level edges ( ), Represents cross-level modularity.

[0101] Based on the Louvain algorithm, the network structure is constructed according to the divided subsets, and the incremental change of the modularity of each node when it moves to the adjacent node is calculated. , expressed as:

[0102]

[0103]

[0104] in, represents the total number of edges across levels, Representation Node and Direct connection between represents the total number of connected edges, It indicates the change in the internal connection of the atomic group after node i is removed from the atomic group. Represents node i and node The connection status, and Respectively represent nodes and nodes The degree.

[0105] For each node i, determine the maximum The largest adjacent node j is found, and node i is moved to the same subgroup as j to complete the merging and recalculation of the subgroup structure;

[0106] Repeat the iteration to merge and recalculate the subgroup structure. Based on the shortest path length matrix value between nodes, calculate the standard deviation and multiply it with the sensitivity constant (set to a small value based on historical experience, such as 0.01) as the subgroup threshold.

[0107] Real-time monitoring of modularity increments in each iteration , and in When it is less than the subgroup threshold, the iteration stops;

[0108] Based on the iterated subgroup structure, the modularity within each level and the cross-level modularity are recalculated and normalized.

[0109] By calculating the average path length of all nodes as the length threshold, two nodes whose shortest path length is less than the threshold are divided into the same subgroup. The global path information and local node connection are innovatively combined to avoid the problem of detail loss that may be caused by the traditional subgroup division based on direct adjacency. Based on the Louvain algorithm, the incremental change of modularity of each node when it moves to the adjacent node is calculated, and the subgroup structure is gradually optimized by maximizing the modularity increment. Not only the initial subgroup affiliation of the node is considered, but also the subgroup position of each node is iteratively optimized to achieve the optimal overall modularity. By introducing the statistical characteristics of the path length distribution, the merging stop condition of the subgroup structure is more reasonably determined to avoid the risk of over-merging or premature stopping. By calculating the modularity within each level, the modularity across levels is also independently analyzed, and the results are finally normalized. This method can more comprehensively reflect the multi-level coupling characteristics of the power grid, help identify potential risks and dependencies between different voltage levels, and ensure that while optimizing the modularity, reasonable structural complexity and hierarchical coordination are maintained.

[0110] Furthermore, based on the node data, the hierarchical vulnerability index is determined, and the cross-level vulnerability index is analyzed to calculate the multi-level vulnerability quantitative index of the power grid, including:

[0111] Construct node vulnerability index based on nodes, and count the betweenness centrality value and node degree of nodes;

[0112] The average distance between a node and other nodes is calculated based on the shortest path length matrix as the node closeness;

[0113] The inverse of the maximum fluctuation amplitude of the node voltage is used as the voltage stability index of each node;

[0114] The sum of the betweenness centrality value, node degree, node closeness and voltage stability index is taken as the node vulnerability index of the node.

[0115] Based on the modularity of each level divided by the total number of subgroups at the corresponding level, and considering the vulnerability index of the level node, the vulnerability index of each level is calculated, which is expressed as:

[0116] ;

[0117] in, Represents the vulnerability index of each level, represents the modularity of each level, Represents the node set of the corresponding level, represents the node vulnerability index of node i, Indicates the number of nodes at the corresponding level;

[0118] The cross-level modularity is divided by the sum of the number of node connections across different voltage levels to obtain the cross-level vulnerability index. The vulnerability index of each level and the cross-level vulnerability index are weighted and comprehensively calculated to form the final multi-level vulnerability quantitative index of the power grid, which is expressed as:

[0119]

[0120] in, represents the comprehensive vulnerability index, Indicates the vulnerability index of the high pressure level, represents the vulnerability index of the medium voltage level, Indicates the vulnerability index of the low-voltage level, Represents vulnerability indicators across levels.

[0121] By comprehensively considering the node's betweenness centrality value, node degree, node compactness and voltage stability index, a multi-dimensional node vulnerability index is constructed to form a more comprehensive vulnerability assessment model. By combining the modularity of the level with the node vulnerability index of the level, the vulnerability index of each level is calculated. The subgroups are optimized through iterative calculation, and the content of the calculated modularity is put into practical use. It can not only reflect the structural compactness within the level, but also reflect the potential weak points in the structure through the weighting of the node vulnerability index, so as to achieve a more comprehensive level vulnerability assessment. The cross-level vulnerability index is obtained by summing the cross-level modularity and the number of node connections across different voltage levels, and is compared with the cross-level vulnerability index of each level. The vulnerability indicators are weighted and integrated to form the final quantitative indicators of multi-level vulnerability of the power grid. Through the hierarchical vulnerability calculation combining multi-dimensional node vulnerability indicators and modularity, the evaluation process is more comprehensive and accurate, which can better identify the weak links and key nodes in the power grid and improve the reliability and stability of the power grid. Through the comprehensive analysis of cross-level and intra-level vulnerability indicators, the coupling problems between levels can be more effectively identified and solved during the power grid optimization process, ensuring that the power grid maintains balance and stability between levels and reducing potential risks. Through the formation of comprehensive vulnerability indicators, a systematic reference basis is provided for the dispatching and protection measures of the power grid, which can timely adjust the power grid structure in actual operation to prevent the spread of faults and the occurrence of systemic risks.

[0122] S3: Draw a vulnerability index curve based on the operation of the power grid, set up an early warning alarm mechanism, analyze the stability of the power grid according to the power grid fluctuation amplitude, frequency deviation and power stability, expand the topology branch with the hierarchical vulnerability index to determine the high-stability topology structure, verify the high-stability topology structure and record the data.

[0123] Preferably, a vulnerability index curve is drawn based on the operation of the power grid, and an early warning alarm mechanism is set, including:

[0124] Draw a vulnerability index curve based on real-time collection of operating data during the operation of the power grid, wherein the operating data includes voltage, power, frequency and load data;

[0125] Preprocess the collected data and use a median filter to eliminate noise in the data;

[0126] Calculate the node vulnerability index based on the real-time collected operation data , Vulnerability indicators at the corresponding level , based on MATLAB tools, draw curve graphs for visual display;

[0127] According to historical data and grid operation requirements, the sum of the mean value of the node's historical vulnerability index and twice the standard deviation of the historical vulnerability index is calculated as the node vulnerability threshold;

[0128] The sum of the mean of the historical vulnerability index at different levels and twice the standard deviation of the historical vulnerability index is calculated as the level vulnerability threshold;

[0129] If the node vulnerability index If the node vulnerability threshold is greater than the node vulnerability threshold, the node warning alarm is triggered. If the vulnerability indicators at different levels If it is greater than the layer vulnerability threshold, a layer warning alarm is triggered.

[0130] By combining real-time collected data with historical data, dynamically setting the vulnerability thresholds of nodes and layers, and using median filters to eliminate noise in data preprocessing, the subsequent vulnerability index calculation is more accurate and reliable. The vulnerability indicators of nodes and layers are visualized through MATLAB tools, so that complex power grid vulnerability information can be presented in the form of intuitive charts. The visualization not only facilitates real-time monitoring by power grid managers, but also provides clear early warning information when anomalies occur. The hierarchical early warning mechanism can more accurately locate risk points in the power grid and avoid misoperation or neglect of problems caused by excessive or insufficient early warnings. The hierarchical early warning mechanism of nodes and layers can more accurately locate and respond to risk points in the power grid, prevent local problems from evolving into global failures, and provide more powerful guarantees for the stable operation of the power grid.

[0131] Furthermore, the grid stability is analyzed based on the grid fluctuation amplitude, frequency deviation and power stability, including:

[0132] Based on the maximum voltage fluctuation amplitude value, the average inverse value of the voltage fluctuation is calculated for all nodes as the voltage stability value;

[0133] The frequency deviation value is calculated based on the grid frequency, and the absolute value of the frequency deviation is averaged to represent the frequency stability, which is expressed as:

[0134]

[0135] in, Indicates time The frequency deviation, Represents the total number of sampling points, Indicates frequency stability.

[0136] Based on the power output and total power demand of each node, the power stability is calculated and expressed as:

[0137]

[0138] in, Indicates power stability, represents the total power output of the node, represents the total power demand of the system to which the node belongs, Represents the amount of power imbalance, which is calculated by subtracting the total power demand of the system to which the node belongs from the total power output of the node.

[0139] The comprehensive voltage stability value, frequency stability value and power stability value are normalized and added together as the stability value of the power grid.

[0140] By comprehensively considering voltage stability, frequency stability and power stability, a multi-dimensional grid stability index is formed. Among them, by measuring the voltage fluctuation inversely, the impact of voltage fluctuation on grid stability can be more intuitively reflected. The smaller the fluctuation, the greater the voltage stability value. The frequency stability is measured by calculating the absolute value average of the frequency deviation. The power stability of the grid is measured by calculating the difference between power output and power demand, that is, the power imbalance. Based on the inverse value of the voltage fluctuation amplitude, the absolute value average of the frequency deviation, and the quantification of the power imbalance, the real-time monitoring and dispatching decisions of the grid can be based on more sensitive and accurate indicators, which improves the operational safety and efficiency of the grid. The final grid stability value formed by normalizing and comprehensively calculating various stability indicators can be used as the basis for grid stability early warning, so that the early warning system can identify potential stability problems earlier and more accurately to prevent the expansion of faults.

[0141] In the calculation of voltage stability, the inverse value of voltage fluctuation is used to measure stability. When the voltage fluctuation is large, the stability value will decrease rapidly, and vice versa, the stability value will increase rapidly. It can sensitively reflect the voltage fluctuation and can timely warn the impact of voltage fluctuation on grid stability. For the calculation of frequency stability, the absolute value average of frequency deviation is calculated to eliminate the sign influence of deviation and ensure that both positive and negative deviations are taken into account. Unlike the maximum deviation value, the average absolute deviation can avoid the excessive impact of extreme values ​​on stability assessment, making the assessment result more robust, especially when the frequency fluctuation is not large, it can more truly reflect the actual operation status of the grid. The calculation of the power imbalance directly measures the balance between power supply and demand, and can accurately reflect the power stability of the power grid. By using the exponential decay method for the power imbalance, the impact of the imbalance on the power stability can be highlighted, so that slight imbalances can be identified in time, and the impact of larger imbalances will be significantly amplified, thereby enhancing the early warning effect. By integrating the stability indicators of voltage, frequency and power, the key stability factors in the operation of the power grid are fully covered, and the normalization process ensures that the various stability indicators are summed up at the same level, so that the stability indicators of different dimensions can fairly occupy the corresponding weights in the overall stability, thereby avoiding the excessive influence of a certain indicator on the overall result.

[0142] Furthermore, the topology branch is expanded with the hierarchical vulnerability index to determine the high stability topology structure, including:

[0143] Based on the early warning alarm of node vulnerability threshold and layer vulnerability threshold, the nodes and corresponding subgroups that exceed the standard are confirmed, and the branch and bound algorithm is used based on the initial topological structure data and visualized through MATLAB picture tools. The branch topology structure is generated by deleting the connecting edges of the nodes with high vulnerability indicators, and the branches are repeatedly expanded.

[0144] The corresponding hierarchical vulnerability index is calculated for each new topology structure until the vulnerability index of the hierarchy in the new topology structure is lower than the hierarchical vulnerability threshold and the grid stability continues to increase, then the branch continues to be expanded. If the change in the vulnerability index decreases significantly and the grid stability stops increasing or continues to decrease, then the branch expansion is stopped.

[0145] The boundary assessment mechanism is used in conjunction with the power grid simulation tool PSS / E to simulate and verify the scenario of each generated branch. The branches whose adjustment results cannot meet the requirements for reducing the vulnerability index are pruned, and the topology with a low vulnerability index and the highest power grid stability is selected as the high-stability structure.

[0146] Through the branch and bound algorithm, the connection edges of the nodes with high vulnerability index are gradually deleted to generate different branch topological structures, and these structures are expanded and evaluated to achieve the purpose of reducing the hierarchical vulnerability index. On the basis of the branch and bound algorithm, a limit evaluation mechanism is introduced to evaluate each generated branch and prune it according to the changes in the vulnerability index and the changes in the power grid stability, which effectively avoids the expansion of invalid branches, improves the efficiency of the algorithm, and ensures that the final selected topological structure can significantly reduce the vulnerability index and improve the power grid stability. By considering the reduction of vulnerability index and the increase of power grid stability at the same time, the criteria for branch expansion and pruning are formulated, so that the final optimized structure can not only reduce the vulnerability of the power grid, but also improve the overall stability of the power grid, achieve the balance and improvement of multiple important performance indicators of the power grid, avoid the negative impact of single optimization, and have important innovation and practical application value. Through the dynamic hierarchical vulnerability evaluation method, the expansion and pruning of branches are more scientific, which can reflect the impact of changes in the power grid structure on vulnerability in real time. By co-optimizing the stability of the power grid, it is ensured that the power grid not only avoids the increase of vulnerability risks during the optimization process, but also further improves the overall stability, providing a more reliable guarantee for the safe operation of the power grid.

[0147] Further, verify the high stability topology and perform data logging, including,

[0148] Apply the optimal topology to the power grid, start the power grid operation and collect key indicator data in real time, and calculate the adjusted hierarchical vulnerability indicators and power grid stability values ​​in real time;

[0149] Compare the hierarchical vulnerability index and the grid stability value before and after the adjustment. If the change value of the hierarchical vulnerability index is less than or equal to 0, and the change value of the grid stability value is less than or equal to 0, it indicates an optimization error.

[0150] Summarize the changes in vulnerability indicators and stability indicators in all tests and create a comprehensive report for recording and preservation.

[0151] By applying the optimal topology to the power grid, key indicator data is collected and calculated in real time, and the vulnerability indicators and power grid stability values ​​before and after the adjustment are immediately compared, so as to dynamically verify the optimization effect during operation, and ensure that the optimization results can truly improve the stability of the power grid and reduce vulnerability. Through the real-time dynamic feedback mechanism and the ability to instantly identify errors, it is ensured that there are no invalid or harmful measures in the optimization process, thereby improving the safety and reliability of power grid operation. Through the systematic recording and preservation of comprehensive reports, a scientific basis is provided for power grid optimization decisions, and the transparency and traceability of the optimization process are improved, providing data support for future optimization improvements.

[0152] Example 2

[0153] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0154] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0155] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0156] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0157] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for constructing a multi-level vulnerability quantitative index based on topological power grid operation, characterized in that: include: Collecting topological data of the power grid, the topological data including node data and edge data, forming a hierarchical matrix according to the node data, and analyzing the correlation between nodes at each level; Preliminarily divide the node into subgroups, calculate the modularity of different levels and across levels, calculate the incremental change of modularity of node movement based on Louvain algorithm, and determine the subgroup structure through repeated iterations, determine the hierarchical vulnerability index based on node data, and calculate the quantitative index of multi-level vulnerability of the power grid; Draw the vulnerability index curve, set up the early warning mechanism, analyze the grid stability according to the grid fluctuation amplitude, frequency deviation and power stability, expand the topology branch with the hierarchical vulnerability index to determine the high-stability topology structure, verify the high-stability topology structure and record the data.

2. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 1, characterized in that: The forming of a hierarchical matrix based on the node data and analyzing the association between nodes at each level includes preprocessing the topological structure data and constructing an adjacency matrix of the power grid based on whether there are edges between the nodes. ; Count the number of nodes directly connected to node i as the node degree ; Based on the voltage level of the power grid, the hierarchy is divided into three levels: high voltage level, medium voltage level and low voltage level; the high voltage level includes all substations and transmission lines of 220kV and above, the medium voltage level includes all substations and transmission lines between 35kV and 220kV, and the low voltage level includes all distribution stations and distribution lines of 35kV and below; the voltage levels of the nodes based on the adjacency matrix are allocated according to the high voltage level, medium voltage level and low voltage level to form a hierarchical matrix , the level matrix Includes high voltage level matrix , medium voltage level matrix and low voltage level matrix ; For each level matrix , use the Floyd-Warshall algorithm to calculate the shortest path length matrix between any two nodes ; Topological distance matrix for each level Analyze and calculate the betweenness centrality value of each node; analyze the correlation between nodes at each level based on whether two nodes belong to different levels and are directly connected, and establish a cross-level connection relationship matrix .

3. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 2, characterized in that: The preliminary division of node subgroups, calculation of different levels and cross-level modularity, calculation of the incremental change of modularity of node movement based on Louvain algorithm, and repeated iteration to determine the subgroup structure include: If two nodes belong to the same level, they are considered as subgroups. For each node , by calculating the shortest path length matrix , calculate the average path length of all nodes as the length threshold, and divide the two nodes whose shortest path length is less than the length threshold into the same subgroup; calculate the modularity within the level based on the adjacency matrix of each level of the power grid, expressed as: , in, Indicates the modularity of the corresponding power grid level, Represents the total number of edges in the network at this level, Representation Node The degree of is the indicator function, Representation Node The degree of and They represent the subgroups to which the nodes belong respectively; At the same time, the cross-level modularity is calculated and expressed as: , in, represents cross-level modularity; Based on the Louvain algorithm, the network structure is constructed according to the divided subsets, and the incremental change of the modularity of each node when it moves to the adjacent node is calculated. , expressed as: , , in, represents the total number of edges across levels, Representation Node and Direct connection between represents the total number of connected edges, It indicates the change in the internal connection of the atomic group after node i is removed from the atomic group. Represents node i and node The connection status, and Respectively represent nodes and nodes The degree of For each node , determine the maximum The largest neighboring node , and the node Move to In the same subgroup, the subgroup structure is merged and recalculated; Repeat the iteration to merge and recalculate the subgroup structure. Based on the shortest path length matrix value between nodes, calculate the standard deviation and multiply it with the sensitivity constant as the subgroup threshold. Real-time monitoring of modularity increments in each iteration , and in When it is less than the subgroup threshold, the iteration stops; Based on the iterated subgroup structure, the modularity within each level and the cross-level modularity are recalculated and normalized.

4. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 3, characterized in that: Determining the hierarchical vulnerability index based on the node data and calculating the multi-level vulnerability quantitative index of the power grid includes constructing the node vulnerability index based on the node and counting the betweenness centrality value and node degree of the node; The average distance between a node and other nodes is calculated based on the shortest path length matrix, and the average distance is used as the node closeness; the inverse of the maximum fluctuation amplitude of the node voltage is used as the voltage stability index of each node; the sum of the comprehensive betweenness centrality value, node degree, node closeness and voltage stability index is used as the node vulnerability index of the node; the modularity of each level is divided by the total number of subgroups of the corresponding level, and the vulnerability index of each level is calculated considering the vulnerability index of the level node, which is expressed as: ； in, Represents the vulnerability index of each level, represents the modularity of each level, Represents the node set of the corresponding level, represents the node vulnerability index of node i, It represents the number of nodes at the corresponding level. The cross-level modularity is divided by the sum of the number of node connections across different voltage levels to obtain the cross-level vulnerability index. The vulnerability index of each level and the cross-level vulnerability index are weighted and comprehensively calculated to form the final multi-level vulnerability quantitative index of the power grid.

5. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 4, characterized in that: The drawing of the vulnerability index curve and setting of the early warning mechanism include drawing the vulnerability index curve based on real-time collection of operation data during the operation of the power grid, wherein the operation data include voltage, power, frequency and load data; Preprocess the collected data and use a median filter to eliminate noise in the data; Calculate the node vulnerability index based on the real-time collected operation data , Vulnerability indicators at the corresponding level , based on the MATLAB tool, a curve graph is drawn for visual display; according to historical data and the operation requirements of the power grid, the sum of the mean value of the historical vulnerability index of the node and twice the standard deviation of the historical vulnerability index is calculated as the node vulnerability threshold; The sum of the mean of the historical vulnerability index at different levels and twice the standard deviation of the historical vulnerability index is calculated as the level vulnerability threshold; If the node vulnerability index If the node vulnerability threshold is greater than the node vulnerability threshold, the node warning alarm is triggered. If the vulnerability indicators at different levels If it is greater than the layer vulnerability threshold, a layer warning alarm is triggered.

6. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 5, characterized in that: The analysis of grid stability according to grid fluctuation amplitude, frequency deviation and power stability includes, based on the maximum voltage fluctuation amplitude value, comprehensively calculating the average inverse value of voltage fluctuation of all nodes as the voltage stability value; The frequency deviation value is calculated based on the grid frequency, and the absolute value of the frequency deviation is averaged to represent the frequency stability, which is expressed as: , in, Indicates time The frequency deviation, Represents the total number of sampling points, Indicates frequency stability; Based on the power output and total power demand of each node, the power stability is calculated and expressed as: , in, Indicates power stability, represents the total power output of the node, represents the total power demand of the system to which the node belongs, Indicates the power imbalance; the voltage stability value, frequency stability value and power stability value are normalized and added together as the stability value of the power grid.

7. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 6, characterized in that: The method of expanding the topological branch in coordination with the hierarchical vulnerability index to determine the high-stability topological structure includes: based on the early warning alarm of the node vulnerability threshold and the hierarchical vulnerability threshold, confirming the nodes and corresponding subgroups that exceed the threshold, visually displaying the topological structure data through the branch and bound algorithm and the MATLAB picture tool, generating a branch topological structure by deleting the connection edges of the high vulnerability index nodes, and repeatedly expanding the branches; Calculate the corresponding hierarchical vulnerability index for each new topology until the vulnerability index of the hierarchy in the new topology is lower than the hierarchical vulnerability threshold and the grid stability continues to increase, then continue to expand the branch, and stop expanding the branch when the grid stability stops increasing; The boundary assessment mechanism is used in conjunction with the power grid simulation tool PSS / E to simulate and verify the scenario of each generated branch. The branches whose adjustment results cannot meet the requirements for reducing the vulnerability index are pruned, and the topology with a low vulnerability index and the highest power grid stability is selected as the high-stability structure.

8. The method for constructing a multi-level vulnerability quantitative index based on topological power grid operation according to claim 7, characterized in that: The verifying the high stability topology structure and recording data includes applying the optimal topology structure to the power grid, starting the power grid operation and collecting key indicator data in real time, and calculating the adjusted hierarchical vulnerability index and power grid stability value in real time; comparing the hierarchical vulnerability index and power grid stability value before and after adjustment, if the change value of the hierarchical vulnerability index is less than or equal to 0, and the change value of the power grid stability value is less than or equal to 0, it indicates an optimization error; Summarize the changes in vulnerability indicators and stability indicators in all tests and create a comprehensive report for recording and preservation.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the method for constructing a multi-level vulnerability quantitative indicator based on topological power grid operation as described in any one of claims 1-8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for constructing a multi-level vulnerability quantitative indicator based on topology power grid operation are implemented as described in any one of claims 1 to 8.

Images