A method and system for identifying key nodes in cascading failures of a power system

Abstract

The present invention relates to the technical field of power systems, and particularly relates to a method and system for identifying key nodes of cascading failures in a power system, including: obtaining historical fault data in the power system, quantifying the historical fault data to obtain initial fault data and final fault scale data; extracting a highly non-linear mapping relationship from the initial fault data to the final fault scale; quantifying the cascading failure risks of all nodes in the power system under different fault types based on the highly non-linear mapping relationship from the initial fault data to the final fault scale; and identifying key nodes using a quantification index according to the cascading failure risks of all nodes under different fault types. The highly non-linear mapping relationship between the initial fault and the final fault scale in historical faults is extracted by a BP neural network optimized by LM, and based on this mapping relationship, the final fault scale is predicted according to the initial fault, considering the method for identifying key nodes of cascading failures.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method and system for identifying key nodes in cascading failures of power systems. Background Art

[0002] The statements in this part only provide background technical information related to the present invention, and do not necessarily constitute prior art.

[0003] While new energy sources such as wind power are being widely utilized, they also bring new challenges to the safe and stable operation of China's power grid. By the end of 2020, the installed capacity of new energy in China had reached 500 million kilowatts, and the proportion of future wind and solar new energy power generation will continue to increase. Along with the rapid growth of new energy capacity, large-scale wind turbine tripping accidents caused by voltage problems have occurred many times in power systems at home and abroad. The power deficit caused by large-scale wind turbine tripping may lead to problems of system frequency stability and affect the stable supply of electricity. Identifying key nodes in the propagation of cascading failures is crucial for ensuring the safe and stable operation conditions of the power grid. Once the key nodes are identified, targeted preventive protection measures can be taken, such as rescheduling generator output, strengthening the stability detection of transmission lines, and increasing energy storage capacity.

[0004] Therefore, studying the cascading failure problem of power systems under high proportion of wind power access and identifying key nodes in the propagation of cascading failures are of great significance for ensuring the safe and stable operation of the power grid. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for identifying key nodes in cascading failures of power systems to solve at least one of the technical problems in the above background art.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] The first aspect of the present invention provides a method for identifying key nodes in cascading failures of power systems, including:

[0008] Obtain historical fault data in the power system, and quantify the historical fault data to obtain initial fault data and final fault scale data;

[0009] Extract the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0010] Quantify the cascading failure risks of all nodes in the power system under different fault types based on the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0011] Identify key nodes according to the cascading failure risks of all nodes under different fault types by using quantification indexes.

[0012] The second aspect of the present invention provides a critical node identification system for cascading failures in a power system, comprising:

[0013] A data acquisition module, configured to acquire historical fault data in the power system, and quantify the historical fault data to obtain initial fault data and final fault scale data;

[0014] A data processing module, configured to extract the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0015] A cascading failure risk calculation module, configured to quantify the cascading failure risks of all nodes in the power system under different fault types based on the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0016] A critical node identification module, configured to identify critical nodes according to the cascading failure risks of all nodes under different fault types by using quantification indexes.

[0017] The third aspect of the present invention provides a computer-readable storage medium.

[0018] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in a method for identifying critical nodes of cascading failures in a power system as described in the first aspect above are implemented.

[0019] The fourth aspect of the present invention provides a computer device.

[0020] A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps in a method for identifying critical nodes of cascading failures in a power system as described in the first aspect above are implemented.

[0021] Advantages of the present invention:

[0022] In the method for identifying critical nodes of cascading failures of the present invention, the highly non-linear mapping relationship between the initial fault and the final fault scale in historical faults is extracted by a BP neural network optimized by LM, and based on this mapping relationship, the final fault scale is predicted according to the initial fault. Further, a cascading failure risk index is proposed, and critical nodes are identified based on this index. The present invention considers the method for identifying critical nodes of cascading failures, which is of great significance for ensuring the safe and stable operation of the power system.

[0023] Additional aspects and advantages of the present invention will be given in part in the following description, which will become apparent from the following description, or will be understood through the practice of the present invention. Description of the Drawings

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0025] Figure 1 Schematic diagram of the example system described in the first embodiment of the present invention;

[0026] Figure 2 Comparison chart of the final fault scale prediction and the actual fault scale when the initial fault matrix of the first embodiment of the present invention is [16 19 1.0 1.0];

[0027] Figure 3 Comparison chart of the final fault scale prediction and the actual fault scale when the initial fault matrix of the first embodiment of the present invention is [19 20 1.0 1.0];

[0028] Figure 4 Calculation result chart of the fault risk index and the key node identification result chart of the first embodiment of the present invention. Detailed implementation manners

[0029] The following details the implementation manners of the present invention. The examples of the implementation manners are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The implementation manners described through the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0030] Those skilled in the art of the present technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used here have the same meaning as the general understanding of those of ordinary skill in the art in the field to which the present invention belongs.

[0031] It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art and will not be interpreted with an idealized or overly formal meaning unless defined as here.

[0032] Those skilled in the art of the present technology can understand that, unless specifically stated, the singular forms "a", "an", "the" and "said" used here may also include the plural forms. It should be further understood that the term "comprising" used in the description of the present invention means the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or their groups.

[0033] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0034] For the convenience of understanding the present invention, the following further explains the present invention with specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation to the embodiments of the present invention.

[0035] Those skilled in the art should understand that the drawings are only schematic diagrams of the embodiments, and the components in the drawings are not necessarily essential for implementing the present invention.

[0036] Embodiment 1

[0037] As Figures 1-4 shown, this embodiment provides a method for identifying key nodes in a power system cascading failure, including:

[0038] Obtain historical fault data in the power system, and quantify the historical fault data to obtain initial fault data and final fault scale data;

[0039] Extract the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0040] Based on the highly non-linear mapping relationship from the initial fault data to the final fault scale, quantify the cascading failure risks of all nodes in the power system under different fault types;

[0041] According to the cascading failure risks of all nodes under different fault types, use quantitative indicators to identify key nodes.

[0042] As Figure 1 shown, this paper uses the IEEE 39-bus system to simulate and verify the effectiveness of the proposed method. In Figure 1In the system shown, the total load is 6254 MW. In System 39, 175 wind turbines with a rated capacity of 1.5 MW are installed at Nodes 10 - 16, 19 - 20, and 22 - 24, and the wind power penetration is approximately 50%. To simulate cascading failures in a weak power grid, the lengths of all lines in System 39 are increased to 1.5 times their original lengths. The initial fault is set as a three - phase balanced short - circuit fault with a random line and random fault severity in the N - 2 scenario. The fault lines are selected with medium probability from among Lines 10 - 11, 11 - 12, 12 - 13, 13 - 14, 14 - 15, 15 - 16, 16 - 24, 16 - 19, 19 - 20, 22 - 23, and 23 - 24. The fault severity is characterized by the degree of node voltage reduction and is set according to a uniform probability between 0 - 100%.

[0043] 1. Fault data quantization and neural network construction:

[0044] 1) Characterize and digitize the fault data obtained from the simulation:

[0045] 11) Initial fault quantization:

[0046] Take the vector f om containing the initial fault information as the input, and the vector F em containing the probability distribution of the final fault scale as the output. f om represents the node number where the initial fault occurs and the severity of the initial fault,

[0047] f om = [i m , j m , ΔU im , ΔU jm T (1)

[0048] where i m , j m are the node numbers of the fault location during the m - th simulation of the initial N - 2 three - phase balanced short - circuit fault. Denote the vector [i m , j m as S om ; ΔU im , ΔU jm are the degrees of voltage drop at nodes i m , j m during the m - th simulation of the initial N - 2 three - phase balanced short - circuit fault. Denote the vector [ΔU im , ΔU jm as l om .

[0049] l o = [ΔU im , ΔU jm T (2)​​

[0050] 12) Final failure scale quantification

[0051] The scale of the accident is characterized by the proportion of the final load loss to the total load, denoted as

[0052]

[0053] where P loss is the final load loss, and P all is the total load. The probability distribution matrix of r e is obtained through multiple simulation results.

[0054] 2) Neural network construction and training:

[0055] 21) Replace the traditional BP neural network with the BP neural network optimized by the LM algorithm:

[0056] In the error backpropagation of the traditional BP neural network, the gradient descent method is usually used to adjust the weights to reduce the error, which has problems such as slow convergence speed and low efficiency. To solve the above problems, the Levenberg-Marquardt (LM) algorithm is usually used to improve the BP neural network. Compared with the traditional BP neural network, the BP neural network optimized by LM uses the method of finding the minimum value of the error function e to iterate the network weights and thresholds, and its iterative process can be expressed as:

[0057] w(k + 1) = w(k) - [J T J + μI] -1 J T e(4)

[0058] b(k + 1) = b(k) - [J T J + μI] -1 J T e(5)

[0059] where k represents the number of iterations, b(k) and w(k) represent the weight matrix and bias matrix of the network at the k-th iteration respectively, μ represents the learning rate, I represents the identity matrix, e is the error function, and is the difference between the vector F em output by the network and the probability distribution matrix of the true value r e , and J represents the Jacobian matrix of the error function.

[0060] 22) Extract the input-output relationship using the optimized BP neural network:

[0061] Take the preprocessed vector f om as the input of the neural network, and the vector F emAs the output, a non - linear mapping relationship between the input and output is obtained using a neural network. Using this mapping relationship, any initial fault [S o ,l o T The probability matrix F of the final fault scale distribution after the occurrence e .

[0062] Based on the proposed cascading fault analysis model, the simulation results of 0.7×10 6 times are used as training samples, 0.15×10 6 is used as the validation sample to verify the training results of the deep neural network, and the remaining 0.15×10 6 is used as the test sample to finally test the accuracy of the deep neural network.

[0063] During the neural network training process, considering the balance between over - fitting and under - fitting, finally, the number of hidden layers is selected as 10, the final value of epoch is 40, and the final value of mini - batch is 128.

[0064] 2. Prediction of the final fault scale of the system:

[0065] Figure 2 and Figure 3 are the comparison charts of the predicted final fault scale and the actual fault scale when the initial fault matrices are [16 19 1.0 1.0] and [19 20 1.0 1.0] respectively. As Figure 2 shown, there is an obvious self - organized critical phenomenon between the accident scale and the accident probability of cascading faults. When the initial fault occurs at nodes 16 and 19, the occurrence probability when the accident scale exceeds 20% will decrease rapidly; while when the initial fault occurs at nodes 19 and 20, the occurrence probability when the accident scale exceeds 25% will decrease rapidly.

[0066] The comparison between the training results and the test results shows that the training results are basically in agreement with the test results, and the improved BP neural network can well capture the highly non - linear relationship of cascading faults.

[0067] 3. Identification of cascading fault risk indicators and key nodes

[0068] According to the obtained prediction results, the R i value of each node can be calculated as Figure 3 shown.

[0069]

[0070] R i is the quantitative value of the severe fault risk of node i; S is the vector of all possible initial fault positions containing node i; l is all possible in S o ​Initial voltage drop degree vector at the place; Index function I(S o ,l o ,r e ) is defined as:

[0071]

[0072] α is the defined value for severe faults. In this embodiment, the value is 0.25, and r e value higher than α is regarded as a severe fault. p(S o ,l o ,r e ) is the probability that the initial fault [S o ,l o evolves to the final fault scale r e . The probability p(S o ,l o ) is the probability that the initial fault [S o ,l o occurs, which can be calculated from the historical fault data of the power system. In this embodiment, it is assumed that the occurrence probabilities of each initial fault are the same.

[0073] By comparing the R i values of each node, the criticality of each node during the propagation of cascading faults can be compared. The higher the R i value, the more critical the corresponding node i is in the occurrence and propagation of cascading faults.

[0074] As Figure 4 shown, according to the different R i values of the nodes, the nodes can be divided into three groups, namely the most critical node group (including nodes 16, 19, 20), the sub-critical node group (including nodes 15, 23, 24), and the non-critical node group (10, 11, 12, 13, 14, 22). Among them, the critical node group satisfies R i ≥30, the sub-critical node group satisfies 3≤R i <30, and the non-critical node group satisfies R i <3. That is, the nodes whose mathematical expectation of the final fault scale caused by random initial faults at the nodes is higher than 30% are defined as critical nodes.

[0075] Taking targeted preventive protection measures for the identified critical nodes, such as rescheduling the generator output, strengthening the stability detection of transmission lines, and increasing the energy storage capacity, can ensure the safe and stable operation of the power grid.

[0076] Embodiment 2

[0077] This embodiment provides a system for identifying critical nodes in cascading faults of a power system, including:

[0078] A data acquisition module, configured to acquire historical fault data in a power system, and quantify the historical fault data to obtain initial fault data and final fault scale data;

[0079] A data processing module, configured to extract a highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0080] A cascading fault risk calculation module, configured to quantify the cascading fault risks of all nodes in the power system under different fault types based on the highly non-linear mapping relationship from the initial fault data to the final fault scale;

[0081] A critical node identification module, configured to identify critical nodes according to the cascading fault risks of all nodes under different fault types by using quantification indexes.

[0082] It should be noted here that the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the content disclosed in the first embodiment above. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer executable instructions.

[0083] Embodiment Three

[0084] This embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in a method for identifying critical nodes of cascading faults in a power system as described in the first embodiment above are implemented.

[0085] Embodiment Four

[0086] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps in a method for identifying critical nodes of cascading faults in a power system as described in the first embodiment above are implemented.

[0087] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories and optical memories, etc.) containing computer-usable program code.

[0088] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0089] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce a means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0090] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device to perform a series of operation steps on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0092] Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, it is not a limitation on the protection scope of the present invention. Those skilled in the art should understand that, based on the technical solutions disclosed in the present invention, various modifications or deformations that can be made by those skilled in the art without creative efforts should be covered within the protection scope of the present invention.

Claims

1. A method for identifying key nodes in cascading failures of a power system, characterized in that, Including: Obtain historical fault data in the power system, and quantify the historical fault data to obtain initial fault data and final fault scale data; Extract the non-linear mapping relationship from the initial fault data to the final fault scale; Quantify the cascading fault risks of all nodes in the power system under different fault types based on the non-linear mapping relationship from the initial fault data to the final fault scale; Identify key nodes using a quantification index according to the cascading fault risks of all nodes under different fault types.

2. The key node identification method for cascading faults in a power system according to claim 1, wherein The initial fault data f om is expressed as: f om = [S om , l om T ​ S om = [i m , …, j m represents the set of node numbers where the fault location is located at the m-th fault, and l om represents the set of fault severities at different nodes at the m-th fault.

3. A method for identifying key nodes of cascading faults in a power system according to claim 1, characterized in that, The final fault scale data is the probability distribution of the final fault scale under the initial fault data, specifically as follows: First, quantify the final fault scale as the proportion of the final load loss to the total load under the initial fault data: Among them, P loss is the final load loss, and P all is the total load; Then, obtain the probability distribution of the final fault scale under the initial fault data based on the statistics of historical fault data.

4. A method for identifying key nodes in cascading failures of a power system according to claim 1, characterized in that The extraction of the non-linear mapping relationship from the initial fault data to the final fault scale specifically includes: Use the LM algorithm to optimize the BP neural network to obtain an improved BP neural network; Take the initial fault data as the input of the improved BP neural network, and the predicted probability distribution vector of the final fault scale as the output, and use the neural network to obtain the non-linear mapping relationship between the input and output; Use this mapping relationship to estimate the final fault scale distribution probability matrix after any initial fault occurs.

5. A method for identifying key nodes of cascading faults in a power system according to claim 1, characterized in that, Adopt the discrete mathematical expectation to define the quantification index of the severe fault risk of node i: R i Quantification value of the severe failure risk for node i; S is the vector of all possible initial fault locations containing node i; F e is any initial fault [S o, l o T The final fault scale distribution probability matrix after the occurrence; The index function I(S o , l o , r e ) is defined as:​ α is the defined value for a severe fault, r e A value higher than α is regarded as a severe fault, p(S o ,l o ,r e ) is the probability that the initial fault [S o ,l o evolves to the final fault scale r e after occurrence, and p(S o ,l o ) is the probability of the occurrence of the initial fault [S o ,l o ; Compare each node R i That is, compare the criticality of each node during the propagation of cascading failures.

6. The method for identifying key nodes of cascading faults in a power system according to claim 4, characterized in that, The LM-optimized BP neural network uses the method of finding the minimum value of the error function e to iterate the network weights and thresholds, and its iteration process can be expressed as: w(k + 1)=w(k)-[J T J + μI] -1 J T e b(k + 1)=b(k)-[J T J + μI] -1 J T e where k represents the number of iterations, b(k) and w(k) respectively represent the weight matrix and bias matrix of the network at the k-th iteration, μ represents the learning rate, I represents the identity matrix, and e is the error function for the vector F output by the network em and the true value r e is the difference of the probability distribution matrix, and J represents the Jacobian matrix of the error function 7. A method for identifying key nodes in cascading faults of a power system according to claim 1, characterized in that The fault types that can be identified include three-phase balanced short-circuit faults and three-phase balanced open-circuit faults.

8. A critical node identification system for cascading failures in a power system, characterized in that, Including: A data acquisition module configured to obtain historical fault data in the power system, and quantify the historical fault data to obtain initial fault data and final fault scale data; A data processing module configured to extract the non-linear mapping relationship from the initial fault data to the final fault scale; A cascading fault risk calculation module configured to quantify the cascading fault risks of all nodes in the power system under different fault types based on the non-linear mapping relationship from the initial fault data to the final fault scale; A key node identification module configured to identify key nodes using a quantification index according to the cascading fault risks of all nodes under different fault types.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps in a method for identifying key nodes of cascading faults in a power system as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in a method for identifying key nodes of cascading faults in a power system as described in any one of claims 1-7.

Images