Key power transmission line dynamic identification method, system, equipment and medium

Abstract

The invention discloses a key power transmission line dynamic identification method, system and device and a medium, and the method comprises the steps: employing a first statistical model to carry out the modeling of wind power output data, and generating a first time sequence; according to the first time sequence, establishing a time sequence cascading failure diagram considering the wind power output fluctuation; establishing a maximum influence theoretical model based on a time sequence cascading failure diagram, analyzing a failure influence propagation path, and identifying a key power transmission line; obtaining the fault influence degree of each key power transmission line through a first influence calculation algorithm, and quantifying the effect in the fault propagation process; and screening the key power transmission lines by adopting a first line identification algorithm in combination with the maximum influence theoretical model to obtain a final identification result. According to the invention, through innovatively integrating time sequence modeling, dynamic network analysis and an intelligent optimization algorithm, accurate identification and risk assessment of the key power transmission line in the wind power fluctuation environment are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of reliable operation and risk assessment of power systems, and in particular to a key transmission line dynamic identification method, system, device and medium. BACKGROUND

[0002] With the adjustment of global energy pattern and the rapid development of renewable energy technology, the proportion of clean energy such as wind energy and solar energy in the power system is gradually rising, and the power grid is moving towards low-carbonization and intelligentization. However, wind power output has significant volatility and uncertainty, which brings new challenges to the operation and management of traditional power systems. In order to adapt to the high proportion of new energy access, the structure and operation characteristics of the power grid have undergone profound changes, and the original dispatching strategy and risk assessment means have been difficult to meet the current demand. Although the introduction of new energy has improved energy utilization efficiency and environmental benefits, the uncontrollability of wind power output has also increased the instability of the system, making the power flow and fault propagation characteristics of the transmission line more complex. In recent years, several large-scale power outages caused by wind power fluctuations have shown that ensuring the reliability and safety of power supply has become a problem to be solved. How to accurately identify the key transmission line that may trigger a chain reaction and predict its fault propagation path is an important research topic for improving the stability of the power grid.

[0003] Current key transmission line identification technology mainly relies on steady-state analysis and complex network theory, which can reveal potential risks in power grid operation to some extent. However, these methods are mostly based on static assumptions and do not fully consider the time series changes and dynamic flow characteristics of wind power output, resulting in limited identification accuracy. In addition, the computational burden of some methods is heavy, making it difficult to meet the real-time application needs of large-scale power grids. Therefore, in the context of high wind power penetration, the existing technology has obvious shortcomings, and more adaptable identification means need to be developed. SUMMARY

[0004] In view of the limitations of the above-mentioned existing methods in the context of wind power access, the present application is proposed.

[0005] Therefore, the present application provides a key transmission line dynamic identification method, system, device and medium, which solves the technical problem that the uncertainty and dynamic flow changes of wind power output increase the risk of power grid operation, making it difficult for traditional static line identification methods to meet the needs of modern power grids.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a key transmission line dynamic identification method, comprising:

[0008] a first statistical model is used to model wind power output data to generate a first time series;

[0009] According to the first time sequence, a time sequence cascading failure graph considering wind power output fluctuation is established;

[0010] A maximum influence theory model based on the time sequence cascading failure graph is established, a failure influence propagation path is analyzed, and a key power transmission line is identified;

[0011] The failure influence degree of each key power transmission line is obtained through a first influence calculation algorithm, and the role in the failure propagation process is quantified;

[0012] A first line identification algorithm is used in combination with the maximum influence theory model to screen the key power transmission line to obtain the final identification result.

[0013] As a preferred scheme of the key power transmission line dynamic identification method, the first time sequence is generated by:

[0014] The first statistical model is parameter trained, and wind power output data is generated based on the trained model;

[0015] It is judged whether the wind power output data meeting the first sequence length has been generated;

[0016] If yes, the final first time sequence is output, otherwise the weather state is updated by the first simulation operation and the wind power output data is continuously generated for judgment.

[0017] The beneficial effects of the preferred technical scheme are that the time sequence characteristics and probability distribution characteristics of the wind power output can be accurately captured, and the prediction error can be significantly reduced compared with the traditional steady-state analysis method.

[0018] As a preferred scheme of the key power transmission line dynamic identification method, the establishment of the time sequence cascading failure graph considering wind power output fluctuation includes:

[0019] According to the first time sequence, a cascading failure chain corresponding to each wind power output time is established;

[0020] The time sequence cascading failure data is mapped by a first mapping function to obtain the time sequence cascading failure graph.

[0021] The beneficial effects of the preferred technical scheme are that by associating the edge weight of the directed graph with the failure propagation time, the space-time coupling analysis of the failure propagation path is first realized, and the influence of wind power fluctuation on failure propagation can be accurately reflected.

[0022] As a preferred scheme of the key power transmission line dynamic identification method, the maximum influence theory model based on the time series cascading failure graph comprises:

[0023] The maximum influence theory model is to find a first line combination in the time series cascading failure graph to maximize the failure influence of the line set.

[0024] As a preferred scheme of the key power transmission line dynamic identification method, the analysis of the failure influence propagation path comprises:

[0025] Set the initial failure time of all lines, and select line u as the initial failure line;

[0026] When the initial failure start time of line u meets the first determination condition, it is possible to cause the failure of neighbor line v, and no matter whether line u can cause the failure of line v, line u will not try to cause the failure of line v in the subsequent round;

[0027] If line v is affected to fail, the failure risk will spread from the new failure line to the entire time series cascading failure graph until there is no new line failure.

[0028] As a preferred scheme of the key power transmission line dynamic identification method, the failure influence degree of each key power transmission line obtained by the first influence calculation algorithm comprises:

[0029] The adjacent lines propagated by the failure line u are sorted in time sequence;

[0030] Initialize the influence set of the failure line u Failure line set Q={u}, line failure time Act=-1 and line influence label l visit =0;

[0031] If Stop calculation and output Otherwise, determine that the failure line u has adjacent lines, and sequentially traverse all adjacent lines v of the failure line u;

[0032] Set adjacent line v=1, if the number of adjacent lines is greater than a first threshold, set And judge Otherwise, make a first determination, calculate the failure start time of the adjacent line v according to the result of the first determination, and update the failure line set and the influence set;

[0033] Repeat the above steps until the failure line set is empty.

[0034] The beneficial effect of the preferred technical solution is that the line fault influence calculation reduces the calculation complexity by introducing the fault start time constraint and the neighbor line sorting strategy.

[0035] As a preferred scheme of the key power transmission line dynamic identification method, the key power transmission line screening comprises:

[0036] Initialize the key line set, and arrange all lines in descending order of marginal influence amount;

[0037] Select the first line after sorting as the key line, and add it to the key line set;

[0038] If the second determination condition is met, stop the calculation and output the key line set, otherwise calculate the marginal effect of the second critical line;

[0039] If the marginal effect of the second critical line is greater than or equal to the marginal effect of the third critical line, add the second critical line to the key line set, and judge the second determination condition;

[0040] Otherwise, find a line in the remaining lines whose marginal effect is less than that of the second critical line, recalculate the marginal effect, and arrange the lines in descending order of marginal influence amount;

[0041] Take the first line after rearrangement as the key line, add it to the key line set, and judge the second determination condition.

[0042] The beneficial effect of the preferred technical solution is that the improved key line identification algorithm optimizes the screening process using the characteristics of the submodular function, and the key line identification efficiency is improved by more than 40%.

[0043] In a second aspect, the application provides a key power transmission line dynamic identification system, comprising:

[0044] A sequence generation module is configured to model wind power output data using a first statistical model to generate a first time sequence;

[0045] A fault graph construction module is configured to establish a time sequence cascading failure graph considering wind power output fluctuation based on the first time sequence;

[0046] A maximum influence theory modeling module is configured to establish a maximum influence theory model based on the time sequence cascading failure graph, analyze the fault influence propagation path, and identify the key power transmission line;

[0047] An influence degree quantification module is configured to obtain the fault influence degree of each key power transmission line through a first influence calculation algorithm, and quantify the role in the fault propagation process.

[0048] The line screening identification module is configured to screen the key power transmission line by using a first line identification algorithm combined with the maximum influence theory model to obtain a final identification result.

[0049] In a third aspect, the present application provides an electronic device comprising a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the steps of the method for dynamically identifying a key power transmission line.

[0050] In a fourth aspect, the present application provides a computer readable storage medium storing computer executable instructions, and the computer executable instructions are configured to be executed by a processor to implement the steps of the method for dynamically identifying a key power transmission line.

[0051] Compared with the prior art, the present application has the following beneficial effects:

[0052] ① The present application combines Gaussian Mixture Hidden Markov Model (GMM-HMM) and Monte Carlo simulation to accurately capture the time sequence characteristics and probability distribution of wind power output. Compared with traditional steady-state analysis, the wind power prediction error is reduced from an average of 15% to within 8%, significantly improving data reliability. The GMM-HMM parameters trained by the EM algorithm can accurately describe the wind farm output characteristics under different weather conditions, providing high-quality data for subsequent analysis;

[0053] ② The present application first combines the edge weight of a directed graph with the fault propagation time to realize the spatio-temporal coupling analysis of the fault propagation path. Experimental data show that the TCFG model can accurately reflect the influence of wind power fluctuations on fault propagation, and 3-5 key lines missed by traditional methods are identified in the IEEE-39 node system. Especially when the wind power penetration rate is 30%, the identification accuracy of the present application remains above 92%, while the accuracy of the traditional method decreases to about 65%;

[0054] ③ The LFIC (Line Fault Impact Calculation) algorithm and ILIT (Improved Key Line Identification) algorithm developed by the present application exhibit excellent performance. The LFIC algorithm effectively reduces the computational complexity by introducing a fault start time limit and a neighboring line sorting strategy. The ILIT algorithm optimizes the screening process with the help of the characteristics of the sub-module function, thereby significantly improving the identification efficiency of the key line by more than 40%. Actual application test results show that in the IEEE 118 node system environment, the present research method can accurately identify 15 key lines in only 925 seconds, which is 1.8 times faster than the current optimal algorithm. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0056] Figure 1 The overall flow logic diagram of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0057] Figure 2 The bus system fault topology diagram of the key power transmission line dynamic identification method described in an embodiment of the present application, wherein (a) is a bus power system example topology diagram, and (b) is a time series cascading failure diagram of the bus power system.

[0058] Figure 3 The mapping relationship diagram of the time series cascading failure propagation of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0059] Figure 4 The construction flow diagram of the time series cascading failure diagram (TCFG) of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0060] Figure 5 The fault influence propagation path analysis diagram of the key power transmission line dynamic identification method described in an embodiment of the present application, wherein (a) is an influence propagation process diagram, and (b) is a line L1 influence diagram.

[0061] Figure 6 The improved IEEE 39 node system structure diagram of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0062] Figure 7 The normalized historical wind power output data diagram of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0063] Figure 8 The different topology relationship diagram in the fault propagation network of the key power transmission line dynamic identification method described in an embodiment of the present application, wherein (a) is a topology relationship diagram at t=12, and (b) is a topology relationship diagram at t=24.

[0064] Figure 9 The remaining load increase situation diagram of the IEEE 39 node system after removing the key line of the key power transmission line dynamic identification method described in an embodiment of the present application.

[0065] Figure 10The remaining load increase diagram of the IEEE 118 node system after removing the key transmission line according to the key transmission line dynamic identification method of one embodiment of the present application.

[0066] Figure 11 The remaining load percentage comparison diagram of the key transmission line dynamic identification method of one embodiment of the present application and other methods, wherein (a) is the IEEE 39 node system, and (b) is the IEEE 118 node system. DETAILED DESCRIPTION

[0067] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present application.

[0068] Embodiment 1, reference Figures 1-5 For one embodiment of the present application, a key transmission line dynamic identification method is provided, as shown in the following steps. Figure 1 Specifically, the method comprises the following steps:

[0069] S100: modeling wind power output data by using a first statistical model to generate a first time series;

[0070] S200: establishing a time series cascading failure graph considering wind power output fluctuation according to the first time series;

[0071] S300: establishing a maximum influence theory model based on the time series cascading failure graph to analyze the failure influence propagation path and identify the key transmission line;

[0072] S400: obtaining the failure influence degree of each key transmission line by a first influence calculation algorithm and quantifying the role in the failure propagation process;

[0073] S500: screening the key transmission line by using a first line identification algorithm combined with the maximum influence theory model to obtain the final identification result.

[0074] It should be noted that ① the present application combines Gaussian mixture hidden Markov model (GMM-HMM) and Monte Carlo simulation to accurately capture the time sequence characteristics and probability distribution of wind power output. Compared with traditional steady-state analysis, the present method reduces the wind power prediction error from an average of 15% to within 8%, significantly improving data reliability. The GMM-HMM parameters trained by the EM algorithm can accurately describe the output characteristics of the wind farm under different weather conditions, providing high-quality data for subsequent analysis; ② the present application first combines the edge weight of the directed graph with the fault propagation time to realize the spatio-temporal coupling analysis of the fault propagation path. Experimental data show that the TCFG model can accurately reflect the influence of wind power fluctuations on fault propagation, and identifies 3-5 key lines missed by traditional methods in the IEEE-39 node system. Especially when the wind power penetration rate is 30%, the identification accuracy of the present method remains above 92%, while the accuracy of the traditional method decreases to about 65%; ③ the LFIC (Line Fault Impact Calculation) algorithm and ILIT (Improved Key Line Identification) algorithm developed by the present application exhibit excellent performance. The LFIC algorithm effectively reduces the computational complexity by introducing a fault start time limit and a neighboring line sorting strategy. The ILIT algorithm optimizes the selection process with the help of the submodular function characteristics, thereby significantly improving the identification efficiency of key lines by more than 40%. Actual application test results show that in the IEEE 118 node system environment, the present research method can accurately complete the identification task of 15 key lines in only 925 seconds, which is 1.8 times faster than the current optimal algorithm.

[0075] The specific embodiments of the key transmission line dynamic identification method provided by the present application will be described below. Figure 1 The specific embodiments of the key transmission line dynamic identification method provided by the present application will be described below.

[0076] S100: modeling the wind power output data using a first statistical model to generate a first time sequence;

[0077] It should be noted that in a wind power-containing power system, the fault propagation path can be mapped to a directed graph G T (V,E,T E ) with time sequence attributes, where V represents the node set of all fault transmission lines, E represents the edge set of the fault propagation relationship between fault transmission lines, T E represents the matrix set of when the fault propagation relationship exists between fault transmission lines, T (u,v) represents the time set of the fault propagation relationship between line u and line v, T (u,v) ∈T E . As shown in Figure 2 (a), the wind farm output time sequence of node 1 has 12 time points, Figure 2(b) shows the TCFG of the system. At times 1, 3, and 5, when line L1 fails, it may cause line L4 to fail, but at other times, it will not cause its failure.

[0078] In an optional embodiment, the first statistical model can be a Bayesian dynamic linear model, a dynamic regression model based on a Bayesian framework that describes the output evolution process through state-space equations. Alternatively, the first statistical model can be a Gaussian mixture hidden Markov model, which combines the state transition characteristics of a hidden Markov chain with the probability distribution of a Gaussian mixture model and is suitable for describing the temporal dependence and multimodal characteristics of wind power output.

[0079] In another optional embodiment, the first statistical model may also be a Markov switching autoregressive model, which introduces a Markov state switching mechanism into the autoregressive model and can characterize output characteristics under different operating modes.

[0080] In the embodiment of the present application, a Gaussian mixture hidden Markov model (GMM-HMM) is used to simulate the real-time output of each wind farm to accurately describe the output of multiple wind farms. GMM-HMM can be expressed as:

[0081] λ=(π,A N×N ,b i )

[0082] Where π represents the probability distribution of each meteorological state. Define the finite meteorological state space as {1,2,3,...,N}, and its state transition probability matrix is ​​represented by the N-dimensional probability matrix A N×N Characterization; Use Gaussian mixture model to describe the observed variable b i , assuming b i It is composed of K Gaussian distributions, then b i It can be described as:

[0083]

[0084] Among them, i is the system state at time t, α k is the mixing weight of the k-th Gaussian function, α k ≥0, θ k are the parameters of each Gaussian function, θ k ={μ k ,Σ k}, M is the number of wind farms.

[0085] In the embodiment of the present application, generating a first time series, that is, a real-time wind power output time series of multiple wind farms, includes:

[0086] Parameter training is performed on the first statistical model, and wind power output data is generated based on the trained model;

[0087] It is judged whether wind power output data meeting the first sequence length has been generated;

[0088] If yes, the final first time sequence is output, otherwise the weather state is updated by the first simulation operation and the generation of wind power output data is continued for judgment.

[0089] Specifically, the step of generating wind power output data based on the trained model comprises:

[0090] The historical wind power data O=(o1, o2,..., o T ) is input into the trained model, the probability P(O|λ) of the historical wind power output sequence selected by the model is maximized through the EM algorithm, so as to obtain the model parameters (π, A N×N , θ k ), and the weather state I t＝1 of the system at t=1 is generated; the trained model is used to generate the wind farm output o t＝1 =[p1(t=1), p2(t=2),..., p M (t=1)], wherein p M (t=1) represents the output of the Mth wind farm at t=1.

[0091] In an optional embodiment, the first simulation operation can be generated by Markov chain Monte Carlo sampling method, by constructing a Markov chain, accepting or rejecting sampling points by probability, and gradually converging to the target distribution;

[0092] In another optional embodiment, the first simulation operation can also be generated by Latin hypercube sampling, which divides the multi-dimensional distribution space into layers, and samples only once in each layer, ensuring that the samples uniformly cover all probability intervals.

[0093] In the embodiments of the present application, the Monte Carlo random sampling method is used to generate the first time sequence, i.e. the real-time wind power output time sequence of multiple wind farms; if it is judged that wind power output data meeting the first sequence length has not been generated, the Monte Carlo simulation method is used to randomly sample the system weather state at the next moment, and the steps of generating wind power output data and judgment are repeated, otherwise the final wind power output time sequence is output.

[0094] It should be noted that the specific value of the first sequence length depends on the task requirements and data characteristics, and common choices include: fixed value: such as 1 hour, 6 hours, etc.; dynamic value: dynamically adjusted according to data changes or model requirements; periodic length; length driven by statistical characteristics: such as autocorrelation decay time or spectral analysis results.

[0095] S200: establishing a time series cascading failure diagram considering wind power output fluctuations based on the first time series;

[0096] It should be noted that when establishing a time series cascading failure diagram that takes into account wind power output fluctuations, the present invention mainly considers line failures caused by transmission line overload, and does not take into account the impact of system transients on transmission line voltage and current. Therefore, the following assumptions are made:

[0097] 1) Transmission Line Disconnection Principle: When simulating cascading failures, a disconnection of the initial fault line, Line 1, causes a change in the system's operating structure. This change in power flow may trigger new faults in other lines, which in turn may lead to multiple line failures due to a cascading effect. At this point, obtaining subsequent fault information is difficult, and high-power systems also face issues such as dimensionality hazards. Therefore, this paper comprehensively considers the time-varying characteristics of wind power and the time-domain characteristics of fault propagation, focusing on the lines with the highest risk of subsequent disconnection and overload as the research object. The overload risk of Line 1 is as follows:

[0098]

[0099] Among them, F l is the power flow of line l when other lines fail, F l normal is the power flow of line l when the system is operating normally at the current moment, F max,l is the power flow limit of line l.

[0100] It should be noted that the purpose of introducing the exponential term is to highlight the failure risk when the line is severely overloaded.

[0101] 2) Fault chain termination criteria: The termination of a fault chain is affected by many factors, mainly the system load reduction percentage and the length of the fault chain. To simplify the model, only the system load reduction percentage is used as the termination criterion, and its calculation formula is as follows:

[0102]

[0103] The DC OPF is used to calculate the load reduction percentage to improve the calculation efficiency of the model. The DC OPF calculation process is as follows:

[0104]

[0105] in, and are the loads at node d before and after the system failure at time t, KL is the node line matrix, KP is the node generation matrix, KD is the node load matrix, D0 is the initial load matrix, m is the fault chain expansion level, w l,m It is the status indicator of line 1, which is 1 when the line is open, and 0 otherwise.

[0106] In the embodiments of the present application, when the wind power output fluctuation continues to increase, the occurrence path of cascading failures also changes, and therefore a time series cascading failure graph considering the characteristics of wind power output fluctuation needs to be established, as shown in Figure 4 The specific content includes:

[0107] According to the first time series, a cascading failure chain corresponding to each wind power output moment is established;

[0108] The time series cascading failure data is mapped by using a first mapping function to obtain a time series cascading failure graph G T (V, E, T E ).

[0109] In an optional embodiment, the first mapping function can be an overload probability-based weighted mapping function, the first mapping function can be a time window-based time series aggregation function, and the first mapping function can also be a complex network centrality-based influence mapping function.

[0110] In the embodiments of the present application, the time series cascading failure data is mapped by using a function , so as to obtain a time series cascading failure graph G T (V, E, T E ), as shown in Figure 3 , wherein the fault line l is mapped to the corresponding node of the cascading failure graph G T , the fault propagation relationship is mapped to the edge of the fault graph G T , and the time when the fault propagation relationship exists is mapped to the weight of the edge.

[0111] It should be noted that when the cascading failure cannot be handled in time, the failure will spread step by step, the system load shedding amount will rapidly expand, and the system operation state will be seriously affected. Therefore, the importance of the line closer to the end of the failure chain is lower. In order to illustrate this feature, the overload risk of the line l needs to be improved. Assuming that there are h failure chains containing the fault propagation relationship (u, v) in the time series cascading failure graph TCFGat a certain moment, the improved risk is:

[0112]

[0113] Wherein, m is the number of stages of the fault propagation relationship (u, v) in the failure chain.

[0114] S300: Establish a maximum influence theory model based on a time series cascading failure graph, analyze the failure influence propagation path, and identify a key power transmission line;

[0115] It should be noted that the set of all failure lines caused by the failure of the line u in the time series cascading failure graph TCFGis represented as , that is, the failure influence set. For example,Figure 5 As shown, when t = 1, line L1 is broken due to fault, leading to change of system structure, re-distribution of power flow of the system, and further leading to overload and breakage of line L2; then the output of the wind farm is increased to 12 MW at t = 4, further leading to overload and fault of line L7; correspondingly, due to the step-by-step propagation of L1 fault, line L3 also has overload fault leading to breakage at t = 9. In summary, after the fault of line L1, due to the fluctuation characteristics of wind power output, lines L2, L3 and L7 all have the risk of breakage, and therefore the fault influence set at this time is When analyzing the fault influence propagation path of the static cascading graph, the starting time of the fault caused by line influence does not need to be considered, while in the TCFG, the influence of the starting time needs to be considered, and the starting time of line fault is defined as follows:

[0116] The time at which the fault of line v is caused due to the fault of line u is recorded as the fault starting time Act v of line v. v = min{t | (t ∈ T (u,v) & t ≥ Act u )}. When the initial fault line L1 leads to the fault of line L3, Act L3 = min{9, 11} = 9, and line L3 will have influence on adjacent lines L4, L5, L6 and L7. When t ≥ 9, the line fault will only affect L4 and L5, and will not propagate to lines L6 and L7.

[0117] In the embodiment of the present application, on the basis of the traditional static cascading fault graph influence propagation path analysis method, the TCFG influence propagation path analysis method is proposed, so that the fault risk propagates in the TCFG. In the initial system network, the initial fault time Act = -1 of all lines is set, indicating that the system is normally running at the initial time. After the initial fault of line u, the fault propagation process is as follows:

[0118] (1) It is assumed that after the fault of line u, there is only one probability to cause the fault of its adjacent line v;

[0119] (2) Only when the initial fault starting time of line u satisfies the first determination condition, i.e. Act u ≤ max(T (u,v) ), can the adjacent line v be caused to have fault;

[0120] (3) Regardless of whether the fault can cause the fault of line v, it will not cause the fault of line v at the subsequent time;

[0121] (4) Line v is caused to have fault due to the influence of the previous line fault, and the fault starting time Act v = t.(u,v) where t (u,v) ∈ T (u,v) and Act u ≤ t (u,v) ≤ max(T (u,v) );

[0122] (5) The risk of failure spreads from the new failure line to the entire TCFG until there are no new line failures.

[0123] The maximum impact theory model is to find the first line combination in the time series cascading failure graph, so that the failure impact of the line set is maximized, that is:

[0124]

[0125] where S = {u1, u2,..., u k}, u i is the critical transmission line, and k is the total number of transmission lines in set S, which is specifically expressed as follows:

[0126]

[0127] where ω is a specific scenario, Ω is all possible scenarios that cause failure propagation due to the previous impact, p ω is the probability of the occurrence of scenario ω, is the failure impact of line u when scenario ω occurs, and infs ω (u) is the marginal effect of line u.

[0128] It should be noted that the marginal effect refers to the number of other line failures that may be caused when each additional failure line is added to the system, and |infs ω (u)| represents the size of the marginal effect.

[0129] It should be noted that step S300 constructs a maximum impact theory model (IMTG) based on a time series cascading failure graph (TCFG), which aims to evaluate the propagation impact range when a transmission line fails. The IMTG model incorporates the time characteristics of wind power, builds a dynamic failure propagation network, and determines the lines that have a significant impact on failure propagation by using the influence maximization method. By calculating the impact degree of the critical transmission line and optimizing the calculation process of the failure propagation path, the accuracy of the failure propagation modeling under wind power conditions is enhanced.

[0130] S400: Obtain the failure impact degree of each critical transmission line by a first impact calculation algorithm, and quantify the role in the failure propagation process;

[0131] In an optional embodiment, the first influence calculation algorithm can be a fast evaluation algorithm based on power flow sensitivity, which uses the direct current power flow sensitivity matrix B to calculate the power flow transfer distribution after the line is disconnected.

[0132] In another optional embodiment, the first influence calculation algorithm can also be a graph neural network (GNN) influence propagation algorithm, which inputs the TCFG into a graph neural network, aggregates neighbor information through a message passing mechanism, and outputs the line influence score.

[0133] In the embodiments of the present application, the first influence calculation algorithm is a line fault influence calculation algorithm, which calculates the propagation range of line faults based on a time sequence cascading fault graph through Monte Carlo simulation and dynamic traversal.

[0134] The overload risk w u,v As the basis for judging whether the line v is faulty, the mathematical expectation of the influence of the line u fault can be expressed as In view of the challenge of direct calculation , the Monte Carlo approximation method is introduced to solve it, and thus the LFIC algorithm is designed, which fully considers the time sequence characteristics of the fault propagation relationship between lines in the calculation process. In analyzing the time sequence characteristics, it can be found that when Act L3 = 9, the lines L6 and L7 will not be affected by the line L3 fault, so they do not need to be included in the traversal range. Based on this finding, for each fault line, its adjacent lines are arranged in descending order of max(T (u,v) ). In the traversal process, if the starting fault time of a line is later than the maximum contact time in the already traversed lines, the traversal operation of the subsequent adjacent lines is immediately stopped, so the calculation efficiency can be greatly improved.

[0135] Specifically, the steps of the LFIC algorithm include:

[0136] Given the TCFG, the fault line u, and the time sequence graph influence propagation path analysis.

[0137] (1) According to the time sequence, the surrounding lines triggered by the fault line u are sorted;

[0138] (2) The related set of the fault line u is initialized, specifically: set the influence set Fault line set Q = {u}, line fault time Act = -1, line influence flag l visit = 0;

[0139] (3) If , stop calculation and output Otherwise, go to (4);

[0140] (4) If the fault line u has adjacent lines, then all adjacent lines v are traversed in turn;

[0141] (5) Set v = 1, if the number of adjacent lines is greater than the first threshold, i.e. v > Num (preset number of adjacent lines), then go to (10), otherwise go to (6);

[0142] (6) If v visit = 0, then v visit = 1, generate a random number w, go to (7), otherwise go to (9);

[0143] (7) If w ≤ w u,v and Act u ≤ max (T (u,v) ), then go to (8), otherwise go to (9);

[0144] (8) Calculate Act v = min (t | t ∈ T (u,v) , Δt ≥ Act u ), update Q ← Q ∪ {v},

[0145] (9) Let v ← v + 1, go to (5);

[0146] (10) Set go to (3);

[0147] It should be noted that the first judgment step is to check whether the adjacent line v has been visited, if not, mark it as visited, and generate a random number w for probability judgment; judge whether the random number w is less than or equal to the fault propagation weight wu,v from line u to v, and whether the fault start time of the current line u exceeds the effective propagation time window of (u, v), if the conditions are met, then trigger the fault propagation. Calculate the fault start time of line v, take the minimum time point in T(u, v) that satisfies Δt ≥ Actu, and add v to the fault set Q to be processed and the influence set .

[0148] S500: Adopt a first line identification algorithm combined with a maximum influence theory model to screen critical transmission lines to obtain a final identification result;

[0149] In an optional embodiment, the first line identification algorithm can be a power grid critical node algorithm based on PageRank, regarding TCFG as a directed graph, and quantifying the "influence" of the line through an improved PageRank algorithm;

[0150] In another optional embodiment, the first link identification algorithm may also be a greedy maximum cover algorithm, which converts the critical link identification into a set cover problem. In each round, a candidate link that covers the most unprotected links is selected, that is, the set of uncovered links is initialized to all links. In each round, a link u is selected to maximize |φ(u)∩U|, and U←U\φ(u) is updated. This process is repeated until the critical links are covered or the number k is reached.

[0151] In the embodiment of the present application, the first line identification algorithm is an improved key line identification algorithm.

[0152] It should be noted that a basic method for identifying critical lines based on IMTG is proposed according to the marginal effect, where S is the critical line set, It is the set of fault lines caused by these critical line failures. The marginal effect of each line is Select the line with the largest marginal effect As the first critical line u1, then S={u1}, After that, recalculate the marginal effects of all lines, and select the line with the largest marginal effect at this time as the second critical line u2, and so on, until k critical lines are determined.

[0153] The maximum influence function in the f(·) time series graph is a monotone submodular function, namely:

[0154] f(S∪{u})-f(S)≥f(T∪{u})-f(T)

[0155] where S and T are arbitrary sets, and Considering that the marginal effect of each line in the system satisfies the sub-model characteristic, that is, as the number of critical lines increases, the marginal effect will gradually decrease, the ILIT algorithm is proposed in combination with the BCLI algorithm.

[0156] The main difference between the ILIT algorithm and the BCLI algorithm lies in the identification process of k-1 critical lines, such as u2~u k As an example, the specific process is as follows:

[0157] Assume that the first critical line is u1. When using the BCLI algorithm to identify u1, the second and third lines of marginal effect are x and y respectively, and we have:

[0158] |infs(x)|≥|infs({u1,x})-infs({u1})|

[0159] |infs(y)|≥|infs({u1,y})-infs({u1})|

[0160] If |infs({u1, x}) - infs({u1})| ≥ |infs(y)|, then directly determine the straight line x as u2. Otherwise, find the first line z that satisfies the above condition, and re-calculate the marginal effect of each line in the range of x~z, and finally select the line with the largest marginal effect as u2.

[0161] In the embodiments of the present application, the processing steps of improving the key line identification algorithm include:

[0162] Given TCFG, fault influence set and the number of key lines k to be identified.

[0163] (1) Perform initialization operation on the key line set S;

[0164] (2) Arrange all lines u1, u2,..., u n in descending order according to the size of marginal influence;

[0165] (3) Select the first line in the sorted order as the key line, and include it in the set S, that is, S = S∪u1;

[0166] (4) If |S| = k, terminate the operation and output the key line set S, otherwise continue to execute (5);

[0167] (5) Calculate

[0168] (6) If |infs(u2)| ≥ |infs(u3)|, execute (7), otherwise execute (9);

[0169] (7) Set S←S∪u2, return to (4);

[0170] (8) Return to (4);

[0171] (9) Find the first line u i that satisfies |infs(u i )| < |infs(u2)| (in the remaining lines);

[0172] (10) Recalculate |infs(u m )|, where u m ∈{u3,...,u i}, and reorder u2, u3,..., u i according to |infs(u)|, forming a sequence: u i1 , u i2 ,..., u n-|S| ;

[0173] (11) S←S∪u i1, return (4);

[0174] It should be noted that the second determination condition is |S|=k.

[0175] It should be noted that the present application proposes LFIC (Line Fault Impact Calculation) algorithm and ILIT (Improved Key Line Identification) algorithm, which aims to evaluate the impact degree of each transmission line in the system after the fault occurs, so as to accurately identify the key transmission line. The LFIC algorithm is mainly used to calculate the influence range of each transmission line in the fault propagation process, and quantitatively evaluates the fault degree of the line in the fault chain. And the ILIT algorithm is based on the IMTG model, which optimizes the influence propagation path and improves the accuracy of key transmission line identification. The algorithm in the present application can effectively adapt to the wind power fluctuation environment, enhance the stability of power grid operation, and reduce the risk of cascading reaction caused by transmission line fault.

[0176] Embodiment 2, refer to Figures 6-11 Based on the last embodiment, the present embodiment provides an application example of a key transmission line dynamic identification method, system, device and medium, which verifies the technical effects adopted in the present method.

[0177] The present embodiment demonstrates based on MATLAB software simulation as a benchmark result, and analyzes the example by using IEEE 39 node system and IEEE 118 node system to verify the performance and applicability of the present method. The IEEE 39 node system includes 39 buses, 46 transmission lines, 10 generators and 10 wind farms, with a total load of about 6.19GW, mainly used to evaluate the key transmission line identification effect in small-scale power grid environment. The IEEE 118 node system is composed of 118 buses, 186 transmission lines, 54 traditional generators and 10 wind farms, which is suitable for evaluating the adaptability and calculation efficiency of the present method in large-scale power grid.

[0178] The present embodiment takes the historical output data of the wind farm as input, and sets the sampling time interval to 5 minutes to capture the time sequence characteristics of wind power fluctuation. In the case analysis, four scenarios of wind power penetration rate of 0%, 10%, 20% and 30% are set to evaluate the operation state of the transmission line under different wind power output conditions. In the experiment, the post-fault state is solved by DC-OPF (Direct Current Optimal Power Flow), considering power balance, power generation operation and power flow constraints to ensure the stability of the system under different wind power conditions.

[0179] To verify the effectiveness of the method more comprehensively, the application compares and analyzes three key transmission line identification methods: (1) identifying key lines based on the topological structure of complex networks, which mainly focuses on the topological connection relationship of the power grid, but does not adequately consider the dynamic changes of wind power output in the analysis process; (2) the second method identifies key lines based on risk assessment theory, which fails to effectively integrate the time series variation characteristics of wind power output when calculating the risk level of the transmission line; (3) the third method is a dynamic identification method for key transmission lines based on the maximum impact theory, which combines the time series cascading failure graph (TCFG) and the improved key line identification algorithm (ILIT), and can consider the influence of wind power fluctuations on transmission lines, thereby improving the accuracy of key line identification.

[0180] In this embodiment, the 10-fold cross-validation method is used to evaluate the stability and generalization ability of the proposed method. The historical database is randomly divided into 10 training sets and 10 test sets to ensure fair comparison of different methods under the same data set. All methods are implemented using MATLAB and Gurobi and run on an Intel i5-10500@3.10GHz, 16GB RAM computer.

[0181] To verify the effectiveness of the method, two typical fault scenarios are analyzed: (1) key line identification under single transmission line fault; (2) key line identification under four transmission line faults. The dynamic influence degree of the key transmission line is calculated for each scenario to evaluate the calculation efficiency and identification accuracy of each method. Figure 6 The improved IEEE-39 node system structure is shown. Using the hourly wind power output historical data of a certain region in the third quarter, Table 1 shows the wind power installed capacity, and the normalized wind power output data is shown in Figure 7

[0182] Table 1: Wind power installed capacity

[0183]

[0184] In this embodiment, the TCFG graph model of the power system is not affected by the change of system load reduction percentage, and the TCFG is constructed under the condition of setting Δ = 30%, which is considered as a large enough parameter value. Figure 8 ​The TCFG of the IEEE 39-bus system at t = 12 and t = 24 is shown. The TCFG abstracts the power system with spatial geographical relationship into the fault propagation network with time series relationship. The green thick lines represent the different topological relationships in the TCFG at t = 12 and t = 24. Due to the fluctuation of wind power output, the fault propagation path changes at different times. The TCFG has time series characteristics, which can comprehensively analyze the change of the fault propagation path under the influence of wind power output fluctuation. The scale-free characteristic of the TCFG makes it have strong robustness under random attacks, but it shows great vulnerability under malicious attacks. The small-world characteristic of the power system reflects the mutual influence between the fault lines when a line fails. The time series characteristic and the scale-free characteristic of the TCFG reflect the vulnerability of the line and the robustness of the network structure when considering the influence of wind power output fluctuation.

[0185] To explore whether the proposed model can efficiently identify the critical lines of the power system, the time complexity is analyzed, mainly covering three parts: the time consumption of constructing the TCFG, the time consumption of evaluating the fault influence degree of the line, and the time consumption of identifying the critical transmission line. According to the algorithm in the foregoing, the time consumption of calculating the marginal effect v of the non-critical line is To identify a critical line, the calculation needs to be performed for all non-critical lines The time consumption of sorting is O (|V|log2|V|). In summary, the total time consumption of identifying a critical line is

[0186] Assuming that the time interval is 15 minutes, it is divided into 4 time periods, and the time required by each part is shown in Table 2. For the IEEE-39-bus system, it only takes 211.953 seconds to identify 10 critical lines; while for the IEEE-118-bus system, it takes 925.728 seconds to identify 15 critical lines. In the IEEE-39-bus system, it takes 305.892 seconds to identify the same number of critical lines, while in the IEEE-118-bus system, it takes 1295.327 seconds. It can be seen that the method proposed in the present application can provide feedback information for power system operators in a timely manner.

[0187] Table 2: Calculation efficiency of the model

[0188]

[0189] The application adopts a method based on TCFG to identify critical lines of a wind power system. For an IEEE-39 node system, the first 10 critical lines are shown in Table 3. To analyze the influence of wind power penetration on line identification, the results in Table 3 are divided into four groups: 1) no wind power access, penetration rate is 0%; 2) wind power replaces the conventional generator of bus 35, penetration rate is 10%; 3) wind power replaces the conventional generator of buses 32 and 35, penetration rate is 20%; 4) wind power replaces the conventional generator of buses 32, 35 and 38, penetration rate is 30%.

[0190] Table 3: Top 10 critical lines of IEEE 39 node system under different wind power penetration

[0191]

[0192] From the results in Table 3, it can be seen that as the wind power penetration rate increases, the ranking of critical lines changes. This is because as the proportion of wind power replacing conventional generators increases, the system is more affected by wind power fluctuations, and the power flow of the transmission line changes significantly.

[0193] Currently, the research on the accuracy of critical line identification is mainly from the perspective of attack and defense. The application identifies from the perspective of attack, and verifies its effectiveness through random attack and deliberate attack. In random attack, the line in the system is randomly cut off each time, and the load reduction percentage is calculated; in deliberate attack, the corresponding line is cut off according to the identification result, and the load reduction percentage is calculated. The application adopts a synchronous attack method to cut off the corresponding lines of IEEE-39 and IEEE-118 node systems at the same time.

[0194] To evaluate the response ability of the identification result to subsequent wind power fluctuations, 8 time instants of wind power output are selected for two attack experiments, and the results are shown in Figure 9 and Figure 10

[0195] In the IEEE-39 node system, the remaining load of the system is shown in the following table. Figure 9 It can be seen that except for t=13, the remaining load of the system decreases rapidly, and the remaining load of the system decreases rapidly, which is consistent with the robustness characteristics of the system. Under deliberate attack, the remaining load of the system decreases significantly, showing vulnerability. By comparing the loss of the remaining load of the system after attack of different lines, the critical lines identified by the application can effectively respond to wind power fluctuations. From the following table, Figure 9 ​(b) can be seen that when the number of attack lines is less than 3, the system residual load loss at t = 19 and t = 22 changes little, and the overall downward trend is relatively slow, which is due to the high wind power output at this time, and the system power flow changes little. For the IEEE-118 node system, due to the distribution of multiple wind farms (8 conventional generators are replaced) in different areas, the wind power output fluctuation is reduced, resulting in small differences in critical lines at each time.

[0196] Figure 11 It is shown that when considering wind power output fluctuation, the comprehensive identification effect of the method is better. If the influence of wind power fluctuation is ignored, some time may miss the key line, such as time 1, 4, 7, 10, etc., resulting in poor identification effect. The method of the application performs better at most times because it comprehensively considers wind power fluctuation, but it may not be as good as the other two mainstream methods at individual times (such as Figure 11 (a) time 22 and Figure 11 (b) time 19).

[0197] Overall, the method of the application can effectively cope with wind power output fluctuation, which is consistent with the foregoing comparison results. However, when constructing the TCFG, the time interval of wind power output needs to be determined according to the specific situation, and 15 minutes is usually recommended to meet the time requirements of emergency analysis.

[0198] In example 3, a key transmission line dynamic identification system is provided, which comprises:

[0199] A sequence generation module is configured to model wind power output data using a first statistical model to generate a first time sequence;

[0200] A fault graph construction module is configured to establish a time sequence cascading failure graph considering wind power output fluctuation according to the first time sequence;

[0201] A maximum influence theory modeling module is configured to establish a maximum influence theory model based on the time sequence cascading failure graph, analyze the fault influence propagation path, and identify the key transmission line;

[0202] An influence degree quantification module is configured to obtain the fault influence degree of each key transmission line by a first influence calculation algorithm, and quantify the role in the fault propagation process;

[0203] A line screening and identification module is configured to screen the key transmission line by using a first line identification algorithm combined with the maximum influence theory model to obtain the final identification result.

[0204] It should be noted that the technical scheme of the key power transmission line dynamic identification system is the same as the technical scheme of the key power transmission line dynamic identification method described above, and the details of the technical scheme of the key power transmission line dynamic identification system in the embodiment are not described in detail, which can be seen from the description of the technical scheme of the key power transmission line dynamic identification method.

[0205] The above-mentioned unit modules can be embedded in or independent of the processor in the electronic device in hardware form, or can be stored in the memory in the electronic device in software form, so as to call and execute the operations corresponding to the above-mentioned modules by the processor.

[0206] The embodiment also provides an electronic device, which comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, operator network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a key power transmission line dynamic identification method. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen. The input device of the electronic device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0207] The embodiment also provides a computer readable storage medium, which stores a computer program. The program is executed by the processor to implement the method proposed in the above-mentioned embodiment.

[0208] The storage medium proposed in the embodiment belongs to the same inventive concept as the method proposed in the above-mentioned embodiment, and the technical details not described in detail in the embodiment can be seen from the above-mentioned embodiment, and the embodiment has the same beneficial effects as the above-mentioned embodiment.

[0209] Those skilled in the art can clearly understand the present application by the description of the above embodiments, and the present application can be realized by software and necessary general hardware, and of course, can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH memory, a hard disk, or an optical disc, etc., and includes a number of instructions to make an electronic device (which can be a personal computer, a server, or a network device, etc.) execute the method of the embodiments of the present application.

[0210] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application, and although the present application 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 application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method for dynamic identification of key transmission lines, characterized in that: include: Modeling wind power output data using a first statistical model to generate a first time series; Based on the first time series, establishing a time series cascading failure diagram taking into account wind power output fluctuations; Establish a maximum impact theoretical model based on the time series cascading failure diagram, analyze the fault impact propagation path, and identify critical transmission lines; Obtaining the fault impact of each of the key transmission lines through a first impact calculation algorithm and quantifying its role in the fault propagation process; The first line identification algorithm is combined with the maximum impact theoretical model to screen key transmission lines to obtain a final identification result.

2. A method for dynamic identification of key transmission lines according to claim 1, characterized in that: Generating the first time series includes: performing parameter training on the first statistical model, and generating wind power output data based on the trained model; Determining whether wind power output data meeting the first sequence length has been generated; If so, the final first time series is output; otherwise, the first simulation operation is used to update the meteorological state and continue to generate wind power output data for judgment.

3. A method for dynamic identification of key transmission lines according to claim 2, characterized in that: The establishment of a time series cascading failure diagram considering wind power output fluctuations includes: Establishing a chain of cascading failures corresponding to each wind power output moment according to the first time series; Fault mapping is performed on the time series cascading failure data using a first mapping function to obtain the time series cascading failure graph.

4. A method for dynamic identification of key transmission lines according to claim 3, characterized in that: The establishing of the maximum impact theory model based on the time series cascading failure graph comprises: The maximum impact theoretical model is to find a first line combination in the time series cascading failure graph so as to maximize the failure impact of the line set.

5. A method for dynamic identification of key transmission lines according to claim 4, characterized in that: The analysis of the fault impact propagation path includes: Set the initial fault time of all lines and select line u as the initial fault line; When the initial fault start time of line u meets the first judgment condition, it is possible to cause the neighboring line v to fail. Regardless of whether line u can cause line v to fail, line u will not attempt to cause line v to fail in subsequent rounds. If line v is affected and fails, the failure risk will spread from the new faulty line to the entire time series cascading failure graph until there are no new line failures.

6. A method for dynamic identification of key transmission lines according to claim 5, characterized in that: The obtaining of the fault impact degree of each of the key transmission lines by using the first impact calculation algorithm includes: Sort the adjacent lines propagated by the fault line u in chronological order; Initialize the impact set of the fault line u Fault line set Q = {u}, line fault time Act = -1 and line impact mark l visit =0; like Then stop the calculation and output Otherwise, determine that the fault line u has adjacent lines, and traverse all adjacent lines v of the fault line u in sequence; Set adjacent line v = 1, if the number of adjacent lines is greater than the first threshold, then set and judge Otherwise, a first judgment is performed, and the fault start time of the adjacent line v is calculated according to the result of the first judgment, and the fault line set and the impact set are updated; Repeat the above steps until the fault line set is empty.

7. A method for dynamic identification of key transmission lines according to claim 6, characterized in that: The screening of key transmission lines includes: Initialize the critical path set and sort all paths in descending order of marginal impact; Selecting the first path after sorting as the key path and adding it to the key path set; If the second judgment condition is met, the calculation is stopped and the critical path set is output; otherwise, the marginal effect of the second critical path is calculated; If the marginal effect of the second critical path is greater than or equal to the marginal effect of the third critical path, the second critical path is added to the critical path set, and a second determination condition is determined; Otherwise, find a line among the remaining lines whose marginal effect of the first line is less than the marginal effect of the second critical line, recalculate the marginal effect, and sort the lines in descending order according to the size of the marginal effect; The first rearranged path is taken as a key path, added to the key path set, and the second determination condition is determined.

8. A key transmission line dynamic identification system, using a key transmission line dynamic identification method according to any one of claims 1 to 7, characterized in that: include: A sequence generation module, configured to model the wind power output data using a first statistical model to generate a first time series; a fault diagram construction module, configured to construct a time series chain fault diagram taking into account wind power output fluctuations based on the first time series; A maximum impact theory modeling module is used to establish a maximum impact theory model based on the time series cascading failure diagram, analyze the fault impact propagation path, and identify critical transmission lines; An impact degree quantification module is used to obtain the fault impact degree of each of the key transmission lines through a first impact calculation algorithm, and quantify the role of the fault in the fault propagation process; The line screening and identification module is used to adopt the first line identification algorithm in combination with the maximum impact theoretical model to screen the key transmission lines to obtain the final identification result.

9. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and the processor implements the steps of a method for dynamic identification of key transmission lines according to any one of claims 1 to 7 when executing the computer-executable instructions.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer executable instructions are executed by a processor, the steps of a method for dynamically identifying key transmission lines according to any one of claims 1 to 7 are implemented.

Images