An agc-oriented information system fault scenario generation method and system

Abstract

The application discloses an AGC-oriented information system fault scene generation method and system, and belongs to the technical field of power dispatch automation. The method comprises the following steps: establishing an information system fault propagation path; calculating the correlation degree based on the information system fault propagation path and in combination with the spatial distribution characteristics of nodes and branches, wherein the correlation degree comprises node-node correlation degree, node-branch correlation degree and branch-branch correlation degree; calculating the key degree of a single node based on the generalized in-degree and the generalized out-degree, and calculating the comprehensive key degree of a single branch based on the nodes at both ends of the branch and the occurrence frequency; performing key fault point screening on the recursively expanded N-k fault point set based on the comprehensive score screening threshold, so as to obtain an information system N-k fault point set; and constructing a fault scene generation model based on the information system N-k fault point set, so as to generate a fault scene set. The method improves the calculation efficiency, and thus can more accurately cope with the multi-fault scene in a complex power grid.

Description

Technical Field

[0001] This application relates to the field of power dispatch automation technology, and more specifically to a method and system for generating fault scenarios in an information system for AGC (Automatic Generation Control). Background Technology

[0002] Currently, test cases for AGC (Automatic Generation Control) software are primarily designed based on engineering requirements, lacking a comprehensive consideration of information system fault combinations, resulting in insufficient test scenario coverage and theoretical basis. Most fault studies in power grid dispatching information systems focus on N-1 and Nk fault analysis of the physical power grid, primarily employing sensitivity analysis or optimization methods for fault combinations. However, with the continuous development of intelligent and information-based power information systems, their role in power dispatching automation is becoming increasingly important. Faults in information systems can lead to the failure of critical control information systems such as AGC. Existing physical power grid N-1 and Nk fault analysis methods are generally unsuitable for information systems, mainly due to differences in fault propagation mechanisms and logical dependencies. Fault modes in the physical power grid are based on physical equipment connections, and fault impacts propagate through physical phenomena such as power flow redistribution. Their models are typically based on power grid topology and electrical parameters (such as line impedance and generator output). In contrast, fault modes in information systems depend on logical connections and service flows, with fault propagation achieved through service dependency chains or information flow interruptions. Traditional Nk models lack this type of logical analysis capability. Furthermore, the method of traversing and scanning information system faults does not consider the specific relationships between nodes and branches in the information system, resulting in the existence of redundant fault points, which increases the computational burden of fault simulation and makes it impossible to accurately identify key fault points in the information system.

[0003] N-1 and Nk fault modeling methods in physical power grids are widely used to evaluate the stability and operational impact of power grids under single or multiple branch or generator faults. Traditional physical power grid N-1 and Nk fault analysis methods can be extended to information systems, analyzing data flow based on the node-branch correlation matrix of the information system to identify faults in key nodes and branches. However, in the case of multi-node and multi-branch faults (Nk faults), the complexity of fault combinations in the information system increases significantly, making test case writing more difficult. Existing technologies using traversal scanning fault methods require a large amount of computation and are inefficient. Summary of the Invention

[0004] To address the problems existing in current technologies, this application proposes a method and system for generating fault scenarios in information systems for AGC (Automatic Guided Vehicle) systems. By analyzing the data flow of the information system, key fault combinations are identified, redundant fault combinations are avoided, and testing efficiency is improved. Optimizing the selection of fault points significantly reduces unnecessary computation, effectively improving the efficiency of AGC software testing while ensuring fault coverage, thus meeting the needs for fault analysis and rapid response in power dispatch automation.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] Firstly, this application proposes a method for generating information system fault scenarios for AGC, including:

[0007] Obtain the node-branch association matrix of the information system; based on the node-branch association matrix of the information system, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node;

[0008] Based on the fault propagation path of the information system, and combined with the spatial distribution characteristics of nodes and branches, the correlation degree is calculated;

[0009] Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree, and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0010] The criticality of a single node and the comprehensive criticality of a single branch are used as the basis for the initial selection of fault points. The Nk fault point set is recursively expanded based on the greedy search method and the aforementioned correlation degree to obtain the Nk fault point set of the information system.

[0011] Based on the Nk fault point set of the information system, a fault scenario generation model is constructed to generate fault scenario sets.

[0012] As a further improvement to this application, the information system node-branch association matrix is ​​as follows:

[0013] Obtain m nodes and n branches in the information system, where nodes represent devices in the information system and branches represent information transmission channels between devices; construct the node-branch association matrix of the AGC system to describe the topological relationships between the nodes in the information system; the th node in the topological relationship... i line, number j Column matrix elements Represented as:

[0014] .

[0015] As a further improvement to this application, the recursive search for the upstream node set and upstream branch set of each node based on the node-branch association matrix of the information system includes:

[0016] Based on the node-branch association matrix of the information system, the BFS algorithm is used to traverse backwards from the target node to find all upstream branches of the target node. During the backward traversal, all upstream nodes of the target node are found simultaneously, and the set of upstream nodes and the set of upstream branches of each node are obtained based on all upstream nodes of the target node.

[0017] As a further improvement to this application, the calculation of correlation degree based on the fault propagation path of the information system, combined with the spatial distribution characteristics of nodes and branches, includes:

[0018] Based on the set of upstream nodes and the set of upstream branches in the fault propagation path of the information system, arbitrary nodes and branches are obtained.

[0019] For any two nodes, if they both appear in the upstream set of other nodes and there is no direct upstream or downstream dependency, then the number of times they are structurally collinear is counted as one accumulation, and the number of times any two nodes are collinear is used as the node-node association degree.

[0020] For any two branches, count the frequency with which they appear together in the set of upstream branches of a certain node, and there is no direct upstream or downstream path relationship between the target nodes connected by the two branches. The number of times any two branches are collinear is taken as the branch-branch correlation degree.

[0021] Based on the frequency of co-occurrence of any node and its corresponding branch on a certain upstream path, direct dependencies are eliminated, and the number of times a node and its branch are collinear is obtained as the node-branch correlation degree.

[0022] As a further improvement to this application, the step of determining the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculating the criticality of a single node based on the generalized in-degree and generalized out-degree; and calculating the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch, includes:

[0023] Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node. The generalized in-degree of a node is the number of all upstream nodes that can be recursively traced to the node; the generalized out-degree of a node is the number of other nodes that consider the node as an upstream node.

[0024] The larger of the generalized in-degree and generalized out-degree of a node is taken as the criticality of a single node;

[0025] The comprehensive criticality of a single branch is obtained by weighting and fusing path frequency with the criticality of individual nodes:

[0026]

[0027]

[0028]

[0029]

[0030] in, Crit(e j ) branch road The criticality; , All are weighting coefficients; Path frequency term The normalized value, Criterion for a single node The normalized value; For including branches The number of paths, yes The first in k Path ； branch road The sum of the criticality of the two endpoints; It is a side road The ingress node, It is a side road The output node, Indicates the criticality of a single node.

[0031] As a further improvement to this application, the criticality of a single node and the combined criticality of a single branch are used as the basis for initial selection of fault points. Based on a greedy search method and the aforementioned correlation degree, the Nk fault point set is recursively expanded to obtain the information system Nk fault point set, including:

[0032] Obtain the criticality of all failure points;

[0033] Select any fault point, and sort the fault points whose criticality of a single node or the comprehensive criticality of a single branch is greater than the criticality threshold in descending order of criticality. This sorts them as the first fault combination scenario. Based on the greedy search method and the correlation, traverse all non-redundant fault points to generate k fault combination scenarios. Construct a set of selected fault points based on each fault point. S We obtain the Nk fault point set of the information system.

[0034] As a further improvement to this application, the greedy search method and the correlation degree are used to traverse all non-redundant fault points to generate k fault combination scenarios; and a set of selected fault points is constructed based on each fault point. S The information system Nk fault point set is obtained, including:

[0035] Obtain the correlation of all fault points;

[0036] For each fault point x that is not included in the selected fault point set S, calculate its evaluation function. This represents the set of fault points and currently selected points. Structural relevance:

[0037]

[0038] in, This indicates the structural coupling degree between each fault point. x' For each fault point, The number of elements in the selected fault point set S;

[0039] All evaluation functions The fault points are added to the selected fault point set S to update the selected fault point set S;

[0040] Repeat the above steps for all expanded fault point sets, and form a recursive candidate expansion process constrained by the aforementioned correlation degree, until each set contains k fault points, and output an Nk fault scenario set that satisfies the specified size of fault points k.

[0041] As a further improvement to this application, the step of constructing a fault scenario generation model based on the Nk fault point set of the information system to generate fault scenarios includes:

[0042] Construct N-1 fault scenarios for the information system, and then obtain Nk fault scenarios; specifically including:

[0043] Construct fault type matching constraints, specifically as follows:

[0044]

[0045] In the formula, A set of scenarios for a certain type of Nk fault; Fault point The set of fault types that it is allowed to occur. j These are index tags for different Nk fault scenario categories under the same k value;

[0046] Based on the fault type matching constraints, an N-1 fault is constructed, represented as:

[0047]

[0048] In the formula, C is the set of all fault points, F is the set of all fault types, and T is the set of all fault time windows;

[0049] The Nk fault combinations derived from N-1 faults are represented as follows:

[0050]

[0051] in, For the k-th fault point, For the k-th fault type, For the k-th fault occurrence time window, Let be the start time of the k-th fault. Let be the end time of the k-th fault; It is a set of combinations containing k fault instances;

[0052] In the Nk fault scenarios, and based on the selected key nodes and branches, any fault scenario is generated by combining the fault point, fault type, and fault occurrence time. ,in, For the first i One fault point, For the first i One type of fault, For the first i A fault occurrence time window The start time, The end time; the fault time setting needs to take into account the measurement cycle, calculation cycle and control cycle in the AGC control process.

[0053] As a further improvement to this application, a fault timing constraint method is also included, including:

[0054] Get the control period window for the current analysis. ,in, It is the start time of the control cycle window for the current analysis; To control the cycle, a fault in a certain information element occurs at the end of the cycle. For the control output to have an impact, the following two conditions must be met:

[0055] 1) The fault has not been resolved, and all components remain in a fault state at the end of the control cycle:

[0056]

[0057] In the formula, For the first i Fault type End time; It is a universal quantifier symbol, representing any one;

[0058] 2) The fault has propagated. If a component fault is to affect control commands, the fault start time must be earlier than the maximum effective trigger time corresponding to the type. ,Right now:

[0059]

[0060] For the first i Fault type The start time;

[0061] in, The value depends on the category to which the component belongs:

[0062]

[0063] In the formula, and These represent the measurement period and the calculation period, respectively. , and These represent information elements of measurement, calculation, and control types, respectively; measurement-type faults must occur before the end of the control cycle. If a computational failure occurs within a certain timeframe, it must be detected beforehand. If a control-related fault occurs within the cycle, it must occur at the beginning of the cycle. In this case, it will propagate to the control output end within the current cycle and affect the final instruction calculation.

[0064] An effective Nk fault combination satisfies the following constraints:

[0065] .

[0066] Secondly, this application also provides an information system fault scenario generation system for AGC, including:

[0067] The fault propagation path establishment module is used to obtain the node-branch association matrix of the information system; based on the node-branch association matrix of the information system, it recursively searches for the upstream node set and upstream branch set of each node, and establishes the information system fault propagation path according to the upstream node set and upstream branch set of each node.

[0068] The correlation calculation module is used to calculate the correlation degree based on the fault propagation path of the information system and the spatial distribution characteristics of nodes and branches.

[0069] The comprehensive criticality calculation module is used to determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0070] The fault point screening module is used to use the criticality of a single node and the comprehensive criticality of a single branch as the basis for initial selection of fault points. Based on the greedy search method and the correlation degree, the Nk fault point set is recursively expanded to obtain the Nk fault point set of the information system.

[0071] The fault scenario generation model construction module is used to construct a fault scenario generation model based on the Nk fault point set of the information system, and to generate a fault scenario set.

[0072] As a further improvement to this application, in the fault propagation path establishment module, the information system node-branch association matrix is ​​as follows:

[0073] Obtain m nodes and n branches in the information system, where nodes represent devices in the information system and branches represent information transmission channels between devices; construct the node-branch association matrix of the AGC system to describe the topological relationships between the nodes in the information system; the th node in the topological relationship... i line, number j Column matrix elements Represented as:

[0074] .

[0075] As a further improvement to this application, in the fault propagation path establishment module, the recursive search for the upstream node set and upstream branch set of each node based on the node-branch association matrix of the information system includes:

[0076] Based on the node-branch association matrix of the information system, the BFS algorithm is used to traverse backwards from the target node to find all upstream branches of the target node. During the backward traversal, all upstream nodes of the target node are found simultaneously, and the set of upstream nodes and the set of upstream branches of each node are obtained based on all upstream nodes of the target node.

[0077] As a further improvement to this application, the correlation calculation module is specifically used for:

[0078] Based on the set of upstream nodes and the set of upstream branches in the fault propagation path of the information system, arbitrary nodes and branches are obtained.

[0079] For any two nodes, if they both appear in the upstream set of other nodes and there is no direct upstream or downstream dependency, then the number of times they are structurally collinear is counted as one accumulation, and the number of times any two nodes are collinear is used as the node-node association degree.

[0080] For any two branches, count the frequency with which they appear together in the set of upstream branches of a certain node, and there is no direct upstream or downstream path relationship between the target nodes connected by the two branches. The number of times any two branches are collinear is taken as the branch-branch correlation degree.

[0081] Based on the frequency of co-occurrence of any node and its corresponding branch on a certain upstream path, direct dependencies are eliminated, and the number of times a node and its branch are collinear is obtained as the node-branch correlation degree.

[0082] As a further improvement to this application, the fault point screening module is specifically used for:

[0083] Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node. The generalized in-degree of a node is the number of all upstream nodes that can be recursively traced to the node; the generalized out-degree of a node is the number of other nodes that consider the node as an upstream node.

[0084] The larger of the generalized in-degree and generalized out-degree of a node is taken as the criticality of a single node;

[0085] The comprehensive criticality of a single branch is obtained by weighting and fusing path frequency with the criticality of individual nodes:

[0086]

[0087]

[0088]

[0089]

[0090] in, Crit(e j ) branch road The criticality; , All are weighting coefficients; Path frequency term The normalized value, Criterion for a single node The normalized value; For including branches The number of paths, yes The first in k Path； branch road The sum of the criticality of the two endpoints; It is a side road The ingress node, It is a side road The output node, Indicates the criticality of a single node.

[0091] As a further improvement to this application, the criticality of a single node and the combined criticality of a single branch are used as the basis for initial selection of fault points. Based on a greedy search method and the aforementioned correlation degree, the Nk fault point set is recursively expanded to obtain the information system Nk fault point set, including:

[0092] Obtain the criticality of all failure points;

[0093] Select any fault point, and sort the fault points whose criticality of a single node or the comprehensive criticality of a single branch is greater than the criticality threshold in descending order of criticality. This sorts them as the first fault combination scenario. Based on the greedy search method and the correlation, traverse all non-redundant fault points to generate k fault combination scenarios. Construct a set of selected fault points based on each fault point. S We obtain the Nk fault point set of the information system.

[0094] As a further improvement to this application, the greedy search method and the correlation degree are used to traverse all non-redundant fault points to generate k fault combination scenarios; and a set of selected fault points is constructed based on each fault point. S The information system's Nk fault point set is obtained, including:

[0095] Obtain the correlation of all fault points;

[0096] For each fault point x that is not included in the selected fault point set S, calculate its evaluation function. This represents the set of fault points and currently selected points. Structural relevance:

[0097]

[0098] in, This indicates the structural coupling degree between each fault point. x' For each fault point, The number of elements in the selected fault point set S;

[0099] All evaluation functions The fault points are added to the selected fault point set S to update the selected fault point set S;

[0100] Repeat the above steps for all expanded fault point sets, and form a recursive candidate expansion process constrained by the aforementioned correlation degree, until each set contains k fault points, and output an Nk fault scenario set that satisfies the specified size of fault points k.

[0101] As a further improvement to this application, the step of constructing a fault scenario generation model based on the Nk fault point set of the information system to generate fault scenarios includes:

[0102] Construct N-1 fault scenarios for the information system, and then obtain Nk fault scenarios; specifically including:

[0103] Construct fault type matching constraints, specifically as follows:

[0104]

[0105] In the formula, A set of scenarios for a certain type of Nk fault; Fault point The set of fault types that it is allowed to occur. j These are index tags for different Nk fault scenario categories under the same k value;

[0106] Based on the fault type matching constraints, an N-1 fault is constructed, represented as:

[0107]

[0108] In the formula, C is the set of all fault points, F is the set of all fault types, and T is the set of all fault time windows;

[0109] The Nk fault combinations derived from N-1 faults are represented as follows:

[0110]

[0111] in, For the k-th fault point, For the k-th fault type, For the k-th fault occurrence time window, Let be the start time of the k-th fault. Let be the end time of the k-th fault; It is a set of combinations containing k fault instances;

[0112] In the Nk fault scenarios, and based on the selected key nodes and branches, any fault scenario is generated by combining the fault point, fault type, and fault occurrence time. ,in, For the first i One fault point, For the first i One type of fault, For the first i A fault occurrence time window The start time, The end time; the fault time setting needs to take into account the measurement cycle, calculation cycle and control cycle in the AGC control process.

[0113] As a further improvement to this application, a fault timing constraint method is also included, including:

[0114] Get the control period window for the current analysis. ,in, It is the start time of the control cycle window for the current analysis; To control the cycle, a fault in a certain information element occurs at the end of the cycle. For the control output to have an impact, the following two conditions must be met:

[0115] 1) The fault has not been resolved, and all components remain in a fault state at the end of the control cycle:

[0116]

[0117] In the formula, For the first i Fault type End time; It is a universal quantifier symbol, representing any one;

[0118] 2) The fault has propagated. If a component fault is to affect control commands, the fault start time must be earlier than the maximum effective trigger time corresponding to the type. ,Right now:

[0119]

[0120] For the first i Fault type The start time;

[0121] in, The value depends on the category to which the component belongs:

[0122]

[0123] In the formula, and These represent the measurement period and the calculation period, respectively. , and These represent information elements of measurement, calculation, and control types, respectively; measurement-type faults must occur before the end of the control cycle. If a computational failure occurs within a certain timeframe, it must be detected beforehand. If a control-related fault occurs within the cycle, it must occur at the beginning of the cycle. In this case, it will propagate to the control output end within the current cycle and affect the final instruction calculation.

[0124] An effective Nk fault combination satisfies the following constraints:

[0125] .

[0126] Thirdly, this application also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for generating information system fault scenarios for AGC.

[0127] Fourthly, this application also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for generating information system fault scenarios for AGC.

[0128] Fifthly, this application also proposes a computer program product, which includes computer instructions that instruct a computer to execute the above-described method for generating information system fault scenarios for AGC.

[0129] Compared with the prior art, this application has the following advantages:

[0130] This application proposes a method for generating Nk-type fault scenarios in information systems. Based on the optimization of inter-node correlation, node-branch correlation, inter-branch correlation, and criticality, it introduces a criticality screening mechanism for nodes and branches, combined with a redundant fault removal strategy. This effectively reduces unnecessary computation and improves the efficiency of fault modeling and software testing. This method significantly improves the computational efficiency of Nk-type fault modeling in information systems and is applicable to fault analysis and AGC software test case development in large-scale power grid information systems, providing technical support for smart grid safety assessment and scheduling optimization. By optimizing fault point selection and adopting a joint criticality and correlation screening method, redundant fault points can be effectively reduced, improving computational efficiency and thus more accurately addressing multi-fault scenarios in complex power grids. It solves the problems of high computational complexity, difficulty in fault combination optimization, and low efficiency of AGC software testing in traditional Nk-type fault modeling. Attached Figure Description

[0131] Figure 1 Flowchart of the method for generating fault scenarios in an information system for AGC provided in this application;

[0132] Figure 2The overall architecture diagram of the information system fault scenario generation method for AGC provided in this application;

[0133] Figure 3 Flowchart for calculating the upstream node and branch set with distance to simulate fault propagation in an information system;

[0134] Figure 4 Main flowchart for generating a model for the Nk fault point set of an information system;

[0135] Figure 5 This is a schematic diagram of various cycles of the information system during the AGC control process. Detailed Implementation

[0136] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0137] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0138] Terminology Explanation:

[0139] N-1 model: A reliability analysis model that assesses whether an information system can still operate normally after the failure of any single critical component.

[0140] Nk model: A reliability analysis model for assessing the operational status of an information system after any k critical components (k≥1) fail simultaneously.

[0141] AGC system: Automatic Generation Control, is an information platform that supports automatic generation control functions, used to collect data in real time, calculate power deviation, and issue control commands.

[0142] BFS Algorithm: BFS (Breadth-First Search) is a graph traversal algorithm that starts from the starting node and visits adjacent nodes level by level in a priority manner. It is often implemented using a queue.

[0143] like Figure 1 As shown, this application proposes a method for generating information system fault scenarios for AGC, including the following steps:

[0144] S1, obtain the information system node-branch association matrix; based on the information system node-branch association matrix, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node;

[0145] S2, Based on the fault propagation path of the information system, and combined with the spatial distribution characteristics of nodes and branches, calculate the correlation degree;

[0146] S3. Determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0147] S4. The criticality of a single node and the comprehensive criticality of a single branch are used as the basis for the initial selection of fault points. The Nk fault point set is recursively expanded based on the greedy search method and the correlation degree to obtain the Nk fault point set of the information system.

[0148] S5. Based on the fault point set Nk of the information system, a fault scenario generation model is constructed to generate a fault scenario set.

[0149] The aforementioned method, by introducing spatial structural and temporal logic constraints from the AGC control process information system, optimizes the selection of fault points, avoids traversing all fault combinations, effectively reduces the number of fault scenario sets, and improves the feasibility and coverage of fault scenarios. Simultaneously, it significantly enhances the efficiency of fault scenario modeling and software testing. This method can quickly screen key fault points, optimize fault point combinations, and generate fault scenario sets, providing a theoretical basis for software quality control, supporting automated testing, and offering strong technical support for scheduling optimization and self-healing control.

[0150] based on Figure 2 The following is a detailed description of the main process of correlation calculation based on the fault propagation mechanism in the information system, which mainly includes steps S1 to S5.

[0151] S1, obtain the information system node-branch association matrix; based on the information system node-branch association matrix, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node.

[0152] Based on the data acquisition, processing, transmission, and control execution functional modules of the AGC system, a node-branch association matrix is ​​constructed. The BFS algorithm is used to traverse all nodes, recursively searching for the upstream nodes and branch sets of each node, establishing an information system fault propagation path reflecting the information flow propagation characteristics. According to the actual deployment structure of the AGC system, a node-branch association matrix A is constructed, where nodes include data acquisition terminals, regional data concentrators, scheduling data servers, the AGC master station information system, backup control center, communication gateways, historical databases, and real-time databases. Branches represent information transmission channels between nodes. Matrix element A... ij Represents a node i and branch road j The relationships between them include:

[0153] Define the node-branch association matrix of an information system: Suppose there are m nodes and n branches in the information system, where nodes represent devices within the system, and branches represent information transmission channels between devices. (Node-branch association matrix of an information system) Used to describe the topological relationships between nodes in an information system, specifying which branches connect each node to its neighboring nodes. Matrix elements are represented as:

[0154]

[0155] In the formula, It is the matrix number i line, number j The elements of the column.

[0156] The node-branch correlation matrix described above reflects the interconnection between nodes and branches in the information system and is the basis for fault propagation calculation.

[0157] As a further improvement, a node-branch association matrix of the AGC system is constructed. The BFS algorithm is used to traverse all nodes, recursively searching for the upstream nodes and branch sets of each node to establish the information system fault propagation path; including:

[0158] Based on the node-branch association matrix of the information system, the BFS algorithm is used to traverse backwards from the target node to find all upstream branches of the target node.

[0159] During the traversal, all upstream nodes of the target node are found simultaneously;

[0160] By recursively calling the BFS algorithm, the search range is expanded upwards layer by layer until the data source node of the AGC system is reached. The upstream nodes and branch set of each node are recursively searched to establish the information system fault propagation path.

[0161] Among them, the calculation of upstream nodes and branch sets in the fault propagation path of the simulated information system is as follows: Figure 3 The diagram shows a flowchart for calculating the upstream nodes and branch sets in a simulated information system fault propagation. In this flowchart, nodes represent transmitted data, and branches represent data transmission paths. Node faults can be caused by: 1) the node itself; or 2) the propagation of faults from upstream nodes and branches. Branch faults, however, are caused solely by their own internal factors and are not caused by faults in upstream nodes or branches. Therefore, fault propagation calculations are used to identify all upstream branches that may affect a given node. All upstream nodes that affect the data source of a certain node The specific process includes the following steps:

[0162] Step 1: Locate the target node All upstream branches ;

[0163] target node The upstream branch is all that makes Side road:

[0164]

[0165] That is, searching the matrix The Okay, find all branch indices with a value of 1. These branch roads are the upstream branch roads.

[0166] Step 2: Locate the target node All upstream nodes ;

[0167]

[0168] That is, for each upstream branch , search matrix The Column, find all row indices with a value of -1 The nodes corresponding to these rows are the upstream nodes.

[0169] Step 3: Use the BFS algorithm to recursively search for all upstream nodes and branches;

[0170] If you need to find all possible causes To identify faulty upstream nodes and branches, we need to examine not only the directly connected upstream branches but also continue recursively upwards.

[0171] Specifically, the recursive iteration process in the above scheme is as follows:

[0172]

[0173]

[0174]

[0175] when When the iteration stops, the final set is obtained:

[0176]

[0177] in, For the target node All upstream nodes The initial value of the set; target node All upstream branches gather; and Represent the target node in the t-th iteration. All upstream nodes Set, all upstream branches gather.

[0178] The recursive search steps using the Breadth-First Search (BFS) algorithm are as follows:

[0179] Initialization: Target Node upstream node set upstream branch collection Pending queue

[0180] Then, an iterative search is performed, the specific process of which is as follows:

[0181] when When the queue is not empty:

[0182] Retrieve the current node :

[0183]

[0184] Search All upstream branches :

[0185]

[0186] Update the global upstream branch set :

[0187] Distance is denoted as ;

[0188] Find these branch roads upstream node set :

[0189]

[0190] Update the global upstream node set :

[0191] The distance is denoted as dist+1;

[0192] Add the newly discovered upstream node to the queue :

[0193]

[0194] Termination condition: when At that time, all possible upstream nodes and branches have been found.

[0195] After the termination condition is met, the final output is: all upstream nodes: All upstream branches: .

[0196] Step 4: Traverse all nodes in the information system and calculate their upstream nodes and branch sets:

[0197] For each node in the information system The BFS recursive search method is used to calculate all possible data sources that may affect it, i.e., the set of upstream nodes. and upstream branch road collection .

[0198] ① Traverse all nodes, and use each node as the target node in turn. For each node, perform a BFS recursive search to find all its upstream nodes and upstream branches;

[0199] ② Record the search results, including each node. upstream node set and upstream branch road collection Store using a suitable data structure.

[0200] S2, based on the fault propagation path of the information system and combined with the spatial distribution characteristics of nodes and branches, calculate the correlation degree, which includes node-node correlation degree, node-branch correlation degree, and branch-branch correlation degree. This step is based on the upstream node set. and upstream branch road collection Calculate the spatial correlation between nodes in an information system to measure the degree of influence between nodes. The specific process includes the following steps:

[0201] Step 1: Define the node-node structural correlation matrix; for any two nodes If both appear in the upstream set of other nodes and there is no direct upstream or downstream dependency, then the number of times they are structurally collinear is counted as one accumulation:

[0202]

[0203] Define the branch-branch structural correlation matrix; for any two branches Count the frequency of their co-occurrence in the upstream branch set of a certain node, and construct the structural coupling degree between branches if there is no direct upstream or downstream path relationship between the target nodes connected by the two branches:

[0204]

[0205] Define the node-branch structural correlation matrix; nodes and branch road The structural correlation is defined as the frequency of co-occurrence of elements in a given upstream path, excluding direct dependencies:

[0206] .

[0207] Step 2: Initialize the correlation calculation; Input data: Set of all upstream nodes Set of all upstream branches .

[0208] Step 3: Calculate the correlation degree. The specific process is as follows:

[0209] Node-to-node correlation; for any pair of nodes If both appear at at least one node If two elements are in the upstream set of a given set and there is no direct upstream-downstream dependency between them, then the degree of collinearity is defined as:

[0210]

[0211] Branch-branch correlation degree; for any two branch pairs If its target node does not constitute a direct upstream or downstream path, and it also appears in the upstream branch set of a certain node, then its co-occurrence degree is defined as follows:

[0212]

[0213] Node-branch correlation degree; for any node and branch roads ,like Appears at a certain node In the set of upstream nodes, and It appears in its upstream branch set, and and It does not constitute a direct connection (neither inflow nor outflow), nor is it a The upstream node of the target node, then:

[0214] ;

[0215] Step 4: Matrix Normalization

[0216] To eliminate the influence of quantity dimensions, the above matrix is ​​normalized by the maximum value of each row, and each element is identified as its most relevant other elements:

[0217]

[0218] Output result: , , .

[0219] S3. Determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches. Calculate the criticality of a single node based on the generalized in-degree and generalized out-degree, and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0220] In the above scheme, the core of N-1 fault modeling in information systems lies in identifying the impact of the failure of critical nodes or branches on the functions of the information system, while the criticality analysis of nodes and branches provides quantitative support for fault modeling. The specific steps are as follows:

[0221] (1) The criticality of a node is calculated, and the specific process is as follows:

[0222] 1) Generalized in-degree: Defines a node The generalized in-degree is the number of all upstream nodes that a node can recursively trace, i.e.:

[0223]

[0224] in, for The recursive set of upstream nodes satisfies:

[0225]

[0226] After normalization, we get:

[0227] ;

[0228] 2) Generalized out-degree: Defines a node The generalized out-degree is the number of other nodes that consider that node as an upstream node, i.e.:

[0229] ;

[0230] in, For all nodes, their recursive upstream set contains The set of nodes:

[0231] ;

[0232] After normalization, we get:

[0233] ;

[0234] 3) Criticality of a single node: Considering the different characteristics of information nodes in the AGC control process, the nodes are defined... The criticality is the larger of its generalized in-degree and generalized out-degree, as shown in the following formula.

[0235] ;

[0236] This criticality can balance the dependency and impact of nodes. For example, an RTU measurement node may have a generalized in-degree of 0, but its data anomalies can be amplified and propagated downstream along the path; while a control command output node may have a generalized out-degree of 0, but its anomalies can directly lead to deviations in the control results of the physical information system.

[0237] (2) Branch Criticality: The criticality of an information branch during AGC control is mainly determined by its frequency of occurrence in the information propagation path and the structural importance of the nodes it connects to. To comprehensively evaluate the criticality of each branch, this embodiment defines the branch criticality based on a weighted fusion of path frequency and the criticality of individual nodes, as follows:

[0238]

[0239] ;

[0240] ;

[0241] ;

[0242] in, Crit(e j ) branch road The criticality; , All are weighting coefficients; Path frequency term The normalized value, Criterion for a single node The normalized value; For including support road The road number of diameters, yes middle The k Path ； branch road The sum of the criticality of the two endpoints; It is a side road The ingress node, It is a side road The output node, Indicates the criticality of a single node;

[0243] Taking the branch that sends control commands from the PLC to the unit as an example, as the terminal channel for sending control commands, it appears in multiple complete AGC control paths from measurement acquisition to control command issuance; and the upstream and downstream nodes it connects, "PLC command node" and "unit command node", both have high generalized in-degree, so this branch is in a critical position in path propagation and structural connection.

[0244] Specifically, the criticality analysis of nodes and branches can provide a basis for N-1 fault modeling of information systems. Failure of highly critical nodes may lead to widespread functional abnormalities and should be given special attention; failure of highly critical branches may affect multiple critical nodes and is a risk point at the information system level.

[0245] S4, using the criticality of a single node and the combined criticality of a single branch as the basis for initial selection of fault points, and recursively expanding the Nk fault point set based on the greedy search method and the aforementioned correlation degree, to obtain the Nk fault point set of the information system; as Figure 4 The diagram shows the main flowchart of the information system Nk fault point set generation model.

[0246] To construct a representative, low-redundancy Nk information system failure scenario, this embodiment proposes a key failure point combination optimization strategy based on a greedy search approach. This strategy uses criticality as the initial ranking criterion and introduces a structural redundancy suppression mechanism. Through iterative optimization, it selects the k failure points with the greatest systemic impact from the set of failure points. The specific process is as follows:

[0247] (1) Input parameters:

[0248] 1) The criticality of all failure points (including information system nodes and branches);

[0249] 2) The correlation matrices between the three types of fault points include: node-node correlation matrix Rpp; node-branch correlation matrix Rpe; and branch-branch correlation matrix Ree.

[0250] (2) Setting parameters: Fault combination scale k: The number of fault points contained in each fault scenario;

[0251] Initialize the fault scenario set S={}, and the ultimate goal is to generate multiple fault combination scenarios S1,S2,…,Sn, where each scenario Si={c1,c2,…,ck}.

[0252] (3) Outer loop: Sort all fault points (nodes or branches) with criticality greater than the criticality threshold (e.g., the criticality threshold is set to 0) in descending order of criticality, and use them as the starting point c1 of the fault scenario. This process traverses all non-redundant fault points to avoid using redundant points (with zero criticality) that are directly connected to the grounding node as the starting point.

[0253] (4) Inner loop: For each starting point c1, execute the following recursive steps to gradually expand to k fault points. The specific steps are as follows:

[0254] 1) Current stage: The set of selected fault points is determined as Si={c1,c2,…,cm}, denoted as... S Wherein, the number of currently selected fault points is m <k;

[0255] 2) Fault point evaluation: For each fault point x that has not been added to the selected fault point set Si, calculate the evaluation function. This represents the set of fault points and currently selected points. Structural relevance:

[0256]

[0257] in, Indicates the structural coupling degree between various types of fault points; The number of elements in set S; filter all evaluation functions. The fault points are combined with the currently selected fault point set Si to generate a new extended fault point set; x' For each fault point;

[0258] Nodes and nodes Nodes and branches Branch roads and branch roads .

[0259] 3) Selection mechanism: Filter all The fault points are combined with the current Si to generate a new set of extended fault points;

[0260] 4) Iterative expansion: Repeat the above steps for all expanded fault point sets until each set contains k fault points.

[0261] This process constitutes a recursive candidate expansion process with correlation constraints, which can effectively filter out unrelated combinations and prioritize the selection of fault points with high criticality and structural coupling for combination.

[0262] 5) Output results: The final output is a set of typical Nk fault scenarios S that meet the specified size k and have high criticality and low redundancy. It can be used as the input set for subsequent fault propagation simulation and test case generation.

[0263] The above-mentioned Nk fault scenario generation method can flexibly adapt to two types of testing needs: in the manual testing stage, by prioritizing criticality and eliminating redundancy, high-risk combinations that have a greater impact on the AGC control process are selected first; in the automated testing stage, candidate test sets can be generated in batches according to the full combination generation or coverage priority strategy, thereby improving the efficiency of test integrity and information system robustness verification.

[0264] S5. Based on the fault point set Nk of the information system, a fault scenario generation model is constructed to generate a fault scenario set.

[0265] Based on the screening of key failure points, a failure scenario generation model is further constructed to systematically generate a set of failure scenarios that conform to the actual engineering logic constraints. Each failure scenario... From {fault point Fault type Fault time window The triplet definition requires that the scenario be representative, reasonable, and triggerable. The problem of generating a set of fault scenarios is modeled as a combinatorial problem satisfying multiple constraints. For example... Figure 5 The diagram shown illustrates various cycles of the information system during the AGC control process.

[0266] Given the selected key nodes and branches, any fault scenario is composed of the fault point, fault type, and the time of fault occurrence. ,in, For the first i One fault point, For the first i One type of fault, For the first i A fault occurrence time window The start time, The end time; the fault time setting needs to consider the measurement cycle, calculation cycle, and control cycle in the AGC control process, such as... Figure 5As shown. Due to the strong dependence of the AGC control process on information flow, it is essential to ensure that the generated scenarios are representative, reasonable, and triggerable. Therefore, the fault scenario generation problem is modeled as a combinatorial problem satisfying multiple constraints. Specifically, the process is as follows:

[0267] Step 1: Definition of N-1 and Nk Fault Scenarios in Information Systems; In the AGC system, an N-1 fault refers to a type of fault occurring at any node or branch within the information system during a certain time period. An N-1 fault is represented as:

[0268]

[0269] In the formula, C is the set of all fault points, F is the set of all fault types, and T is the set of all fault time windows;

[0270] In an information system, if any k components simultaneously or sequentially experience a certain type of failure, they form Nk failure combinations, denoted as:

[0271]

[0272] in, For the k-th fault point, For the k-th fault type, For the k-th fault occurrence time window, Let be the start time of the k-th fault. Let be the end time of the k-th fault; It is a combined set containing k fault instances;

[0273] Step 2: The fault type matching constraint is:

[0274]

[0275] In the formula, A set of scenarios for a certain type of Nk fault; Fault point The set of fault types that it is allowed to occur. j It is an index marker for different Nk fault scenario categories under the same k value.

[0276] This constraint indicates that for a certain fault point... The selected fault type when generating fault scenarios. It must be one of the types of faults that can occur at the fault point itself.

[0277] Examples of the execution cycles of various modules in the AGC control process are as follows: Figure 5As shown, the system mainly comprises three modules: measurement, calculation, and control issuance, with their time cycles illustrated using typical parameters from the provincial dispatch center. To ensure that faults can effectively interfere with the calculation and issuance of control commands, the start and end times of faults for various components must be rationally designed to ensure their effectiveness within the current control cycle. Therefore, a fault timing constraint method is proposed, which, combined with the multi-level periodic structure and functional transmission links of the information system, uniformly constrains the occurrence and recovery times of each component in the fault combination.

[0278] Let the control period window of the current analysis be... ,in To control the cycle, and to ensure that a fault in a certain information component occurs at the end of the cycle. For the control output to have an impact, the following two conditions must be met:

[0279] 1) Situation where the fault has not yet been resolved

[0280] All components remain in a fault state at the end of the control cycle:

[0281]

[0282] In the formula, For the first i Fault type End time; It is a universal quantifier symbol, representing any one;

[0283] 2) The fault has completed propagation.

[0284] If a component failure is to affect control commands, its failure start time must be earlier than the maximum effective trigger time corresponding to its type. ,Right now:

[0285]

[0286] For the first i Fault type The start time;

[0287] in, The value depends on the category to which the component belongs, and is defined as follows:

[0288]

[0289] in, and These represent the measurement period and the calculation period, respectively. , and These represent information elements of measurement, calculation, and control types, respectively.

[0290] This formula embodies the sufficient condition for the measurement-calculation-control three-layer propagation chain to close within the control cycle, namely, measurement-related faults must occur before the end of the control cycle. If a computational failure occurs within a certain timeframe, it must be detected beforehand. Control-related faults must occur at the beginning of the cycle in order to propagate to the control output and affect the final instruction calculation within the current cycle.

[0291] In summary, an effective Nk fault combination should satisfy the following constraints:

[0292] .

[0293] This constraint not only ensures that the start and end times of the fault conform to their respective periodic characteristics, but also guarantees that the impact of the fault can be transmitted down to the control output within the current control cycle, thus realizing the formation of an effective interference path.

[0294] In the process of generating the fault scenario model, based on the identified key fault components and Nk fault component combinations, the above three hard constraints are superimposed to generate fault types and fault times, thus forming a complete fault scenario. This method generates a set of executable, high-quality fault scenarios through structured combination and constraint verification, and is suitable for evaluating the functional integrity of AGC software, assessing error prevention mechanisms, and verifying the vulnerability of power information systems.

[0295] This application proposes a fault propagation simulation mechanism based on upstream influence, considering the spatial structural dependencies of information systems. This addresses the issue of indiscriminate information flow in network centrality analysis methods, enabling more accurate identification of fault propagation origins and ranges, and providing precise data support for the analysis of key nodes and branches in AGC software information systems. It also proposes a method for quantifying the association of information system components based on path collinearity, a method for quantifying the key components of information system components based on generalized in-degree and generalized out-degree, and a method for combining Nk fault components based on greedy search and association constraints. This improves the efficiency of fault component combination, avoids the computational complexity of traditional full traversal modeling, and achieves key quantification and accurate selection of AGC software information components. Furthermore, it proposes a triplet definition for N-1 and Nk fault scenarios in information systems, and considers the temporal logic of the AGC control process. Through fault type constraints and fault timing constraints, it enhances the universality and feasibility of fault scenario set modeling. This method can support automated testing and significantly improve the testing efficiency of AGC software.

[0296] Secondly, this application discloses an information system fault scenario generation system for AGC, which is based on the above-mentioned information system fault scenario generation method for AGC, including: a fault propagation path establishment module, a correlation calculation module, a comprehensive criticality calculation module, a fault point screening module, and a fault scenario generation model construction module.

[0297] The fault propagation path establishment module is used to obtain the node-branch association matrix of the information system; based on the node-branch association matrix of the information system, it recursively searches for the upstream node set and upstream branch set of each node, and establishes the information system fault propagation path according to the upstream node set and upstream branch set of each node.

[0298] The correlation calculation module is used to calculate the correlation degree based on the fault propagation path of the information system and the spatial distribution characteristics of nodes and branches.

[0299] The comprehensive criticality calculation module is used to determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0300] The fault point screening module is used to use the criticality of a single node and the comprehensive criticality of a single branch as the basis for initial selection of fault points. Based on the greedy search method and the correlation degree, the Nk fault point set is recursively expanded to obtain the Nk fault point set of the information system.

[0301] The fault scenario generation model construction module is used to construct a fault scenario generation model based on the Nk fault point set of the information system, and to generate a fault scenario set.

[0302] Therefore, this application constructs N-1 and Nk fault scenario generation models for the power grid AGC control process. These models can accurately identify critical information system fault propagation paths, avoiding the computational explosion problem caused by full-traversal calculations in traditional fault modeling, and significantly improving the efficiency and relevance of information system fault scenario modeling. Considering the temporal logic of the AGC control process, by optimizing the fault scenario constraint mechanism, a low-redundancy, high-coverage injectable fault scenario set is generated. This technology can significantly improve the testing efficiency of AGC software. Because it can accurately locate key fault points and reduce unnecessary calculations and tests, the fault diagnosis and verification process of AGC software becomes more efficient, reducing the redundant workload caused by full-coverage testing. This provides a theoretical basis for the experimental verification of control software, serves as theoretical support for automated testing, and can be extended to other scheduling automation software such as AGC and scheduling plans, providing strong technical support for the cyber-physical security protection and scheduling optimization of smart grids.

[0303] The third objective of this application is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for generating information system fault scenarios for AGC.

[0304] The method for generating fault scenarios in an AGC-oriented information system, implemented when the processor executes a computer program, specifically includes:

[0305] S1, obtain the information system node-branch association matrix; based on the information system node-branch association matrix, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node;

[0306] S2, based on the fault propagation path of the information system and combined with the spatial distribution characteristics of nodes and branches, calculate the correlation degree, which includes node-node correlation degree, node-branch correlation degree, and branch-branch correlation degree.

[0307] S3. Determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0308] S4. Based on the N-1 fault model of the information system, the criticality of a single node and the comprehensive criticality of a single branch are used as the basis for the initial selection of fault points. The Nk fault point set is recursively expanded based on the greedy search method and the aforementioned correlation degree. A comprehensive scoring rule for the fault point set is defined, and a comprehensive scoring screening threshold is given. Based on the comprehensive scoring screening threshold, the Nk fault point set after recursion is expanded is used to screen for critical fault points, thus obtaining the Nk fault point set of the information system.

[0309] S5, based on the Nk fault point set of the information system, constructs a fault scenario generation model to generate fault scenario sets.

[0310] The fourth objective of this application is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for generating information system fault scenarios oriented towards AGC.

[0311] When a computer program is executed by a processor, the method for generating fault scenarios in an AGC-oriented information system includes:

[0312] S1, obtain the information system node-branch association matrix; based on the information system node-branch association matrix, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node;

[0313] S2, based on the fault propagation path of the information system and combined with the spatial distribution characteristics of nodes and branches, calculate the correlation degree, which includes node-node correlation degree, node-branch correlation degree, and branch-branch correlation degree.

[0314] S3. Determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0315] S4. Based on the N-1 fault model of the information system, the criticality of a single node and the comprehensive criticality of a single branch are used as the basis for the initial selection of fault points. The Nk fault point set is recursively expanded based on the greedy search method and the aforementioned correlation degree. A comprehensive scoring rule for the fault point set is defined, and a comprehensive scoring screening threshold is given. Based on the comprehensive scoring screening threshold, the Nk fault point set after recursion is expanded is used to screen for critical fault points, thus obtaining the Nk fault point set of the information system.

[0316] S5, based on the Nk fault point set of the information system, constructs a fault scenario generation model to generate fault scenario sets.

[0317] The fifth objective of this application is to provide a computer program product, which includes computer instructions that instruct a computer to execute an information system fault scenario generation method for AGC.

[0318] A method for generating fault scenarios for an AGC-oriented information system, which is instructed by computer commands to be executed by the computer, specifically includes:

[0319] S1, obtain the information system node-branch association matrix; based on the information system node-branch association matrix, recursively search for the upstream node set and upstream branch set of each node, and establish the information system fault propagation path according to the upstream node set and upstream branch set of each node;

[0320] S2, based on the fault propagation path of the information system and combined with the spatial distribution characteristics of nodes and branches, calculate the correlation degree, which includes node-node correlation degree, node-branch correlation degree, and branch-branch correlation degree.

[0321] S3. Determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch.

[0322] S4. Based on the N-1 fault model of the information system, the criticality of a single node and the comprehensive criticality of a single branch are used as the basis for the initial selection of fault points. The Nk fault point set is recursively expanded based on the greedy search method and the aforementioned correlation degree. A comprehensive scoring rule for the fault point set is defined, and a comprehensive scoring screening threshold is given. Based on the comprehensive scoring screening threshold, the Nk fault point set after recursion is expanded is used to screen for critical fault points, thus obtaining the Nk fault point set of the information system.

[0323] S5, based on the Nk fault point set of the information system, constructs a fault scenario generation model to generate fault scenario sets.

[0324] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0325] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0326] This application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, readable storage media, optical storage, etc.) containing computer-usable program code.

[0327] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0328] Obviously, the described embodiments are only some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort should fall within the scope of protection of this application.

[0329] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for generating fault scenarios in an information system oriented towards AGC, characterized in that, include: Obtain the node-branch association matrix of the information system; Based on the node-branch association matrix of the information system, the upstream node set and upstream branch set of each node are recursively searched, and the information system fault propagation path is established according to the upstream node set and upstream branch set of each node. Based on the fault propagation path of the information system, and combined with the spatial distribution characteristics of nodes and branches, the correlation degree is calculated, including: Based on the set of upstream nodes and the set of upstream branches in the fault propagation path of the information system, arbitrary nodes and branches are obtained. For any two nodes, if they both appear in the upstream set of other nodes and there is no direct upstream or downstream dependency, then the number of times they are structurally collinear is counted as one accumulation, and the number of times any two nodes are collinear is used as the node-node association degree. For any two branches, count the frequency with which they appear together in the set of upstream branches of a certain node, and there is no direct upstream or downstream path relationship between the target nodes connected by the two branches. The number of times any two branches are collinear is taken as the branch-branch correlation degree. Based on the frequency of co-occurrence of any node and its corresponding branch in a certain upstream path, direct dependencies are eliminated, and the number of times a node and its branch are collinear is obtained as the node-branch correlation degree. Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree, and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch. The criticality of a single node and the combined criticality of a single branch are used as the basis for initial selection of fault points. Based on a greedy search method and the aforementioned correlation degree, the Nk fault point set is recursively expanded to obtain the Nk fault point set of the information system; including: Obtain the criticality of all failure points; Select any fault point, and sort the fault points whose criticality of a single node or the combined criticality of a single branch is greater than the criticality threshold in descending order of criticality. This sorts them as the first fault combination scenario. Based on the greedy search method and the correlation, traverse all non-redundant fault points to generate k fault combination scenarios. Construct a set of selected fault points based on each fault point. S Thus, the Nk fault point set of the information system is obtained; Based on the Nk fault point set of the information system, a fault scenario generation model is constructed to generate fault scenario sets.

2. The method for generating fault scenarios in an information system oriented towards AGC according to claim 1, characterized in that, The node-branch association matrix of the information system is as follows: Obtain m nodes and n branches in the information system, where nodes represent devices in the information system and branches represent information transmission channels between devices; construct the node-branch association matrix of the AGC system to describe the topological relationships between the nodes in the information system; the th node in the topological relationship... i line, number j Column matrix elements Represented as: 。 3. The method for generating fault scenarios in an information system oriented towards AGC according to claim 2, characterized in that, The recursive search for the upstream node set and upstream branch set of each node based on the node-branch association matrix of the information system includes: Based on the node-branch association matrix of the information system, the BFS algorithm is used to traverse backwards from the target node to find all upstream branches of the target node. During the backward traversal, all upstream nodes of the target node are found simultaneously, and the set of upstream nodes and the set of upstream branches of each node are obtained based on all upstream nodes of the target node.

4. The method for generating fault scenarios in an information system oriented towards AGC according to claim 1, characterized in that, The generalized in-degree and generalized out-degree of each node are determined based on the set of upstream nodes and the set of upstream branches of each node. The criticality of a single node is calculated based on generalized in-degree and generalized out-degree. The comprehensive criticality of a single branch is calculated based on the criticality of the nodes at both ends of the branch and their frequency of occurrence, including: Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node. The generalized in-degree of a node is the number of all upstream nodes that can be recursively traced to the node; the generalized out-degree of a node is the number of other nodes that consider the node as an upstream node. The larger of the generalized in-degree and generalized out-degree of a node is taken as the criticality of a single node; The comprehensive criticality of a single branch is obtained by weighting and fusing path frequency with the criticality of individual nodes: in, Crit(e j ) branch road The criticality; , All are weighting coefficients; Path frequency term The normalized value, Criterion for a single node The normalized value; For including branches The number of paths, yes The first in k Path ； branch road The sum of the criticality of the two endpoints; It is a side road The ingress node, It is a side road The output node, Indicates the criticality of a single node.

5. The method for generating fault scenarios in an information system oriented towards AGC according to claim 1, characterized in that, The greedy search method and the correlation degree are used to traverse all non-redundant fault points to generate k fault combination scenarios; Construct a set of selected fault points based on each fault point. S The information system Nk fault point set is obtained, including: Obtain the correlation of all fault points; For each fault point x that is not included in the selected fault point set S, calculate its evaluation function. This represents the set of fault points and currently selected points. Structural relevance: in, This indicates the structural coupling degree between each fault point. x' For each fault point, The number of elements in the selected fault point set S; All evaluation functions Fault points that meet the evaluation threshold are added to the selected fault point set S to update the selected fault point set S; Repeat the above steps for all expanded fault point sets, and form a recursive candidate expansion process constrained by the aforementioned correlation degree, until each set contains k fault points, and output an Nk fault scenario set that satisfies the specified size of fault points k.

6. The method for generating fault scenarios in an information system oriented towards AGC according to claim 5, characterized in that, Based on the Nk fault point set of the information system, a fault scenario generation model is constructed to generate fault scenarios, including: Construct N-1 fault scenarios for the information system, and then obtain Nk fault scenarios; specifically including: Construct fault type matching constraints, specifically as follows: In the formula, A set of scenarios for a certain type of Nk fault; Fault point The set of fault types that it is allowed to occur. j These are index tags for different Nk fault scenario categories under the same k value; Based on the fault type matching constraints, an N-1 fault is constructed, represented as: In the formula, C is the set of all fault points, F is the set of all fault types, and T is the set of all fault time windows; The Nk fault combinations derived from N-1 faults are represented as follows: in, For the k-th fault point, For the k-th fault type, For the k-th fault occurrence time window, Let be the start time of the k-th fault. Let be the end time of the k-th fault; It is a set of combinations containing k fault instances; In the Nk fault scenarios, and based on the selected key nodes and branches, any fault scenario is generated by combining the fault point, fault type, and fault occurrence time. ,in, For the first i One fault point, For the first i One type of fault, For the first i A fault occurrence time window The start time, The end time; the fault time setting needs to take into account the measurement cycle, calculation cycle and control cycle in the AGC control process.

7. The method for generating fault scenarios in an information system for AGC according to claim 6, characterized in that, It also includes fault timing constraint methods, including: Get the control period window for the current analysis. ,in, It is the start time of the control cycle window for the current analysis; To control the cycle, a fault in a certain information element occurs at the end of the cycle. For the control output to have an impact, the following two conditions must be met: 1) The fault has not been resolved, and all components remain in a fault state at the end of the control cycle: In the formula, For the first i Fault type End time; It is a universal quantifier symbol, representing any one; 2) The fault has propagated. If a component fault is to affect control commands, the fault start time must be earlier than the maximum effective trigger time corresponding to the type. ,Right now: in, For the first i Fault type The start time; The value depends on the category to which the component belongs: In the formula, and These represent the measurement period and the calculation period, respectively. , and These represent information elements of measurement, calculation, and control types, respectively; measurement-type faults must occur before the end of the control cycle. It happened within a certain time period; An effective Nk fault combination satisfies the following constraints: 。 8. A fault scenario generation system for information systems oriented towards AGC, characterized in that, include: The fault propagation path establishment module is used to obtain the node-branch association matrix of the information system; Based on the node-branch association matrix of the information system, the upstream node set and upstream branch set of each node are recursively searched, and the information system fault propagation path is established according to the upstream node set and upstream branch set of each node. The correlation calculation module is used to calculate the correlation degree based on the fault propagation path of the information system and the spatial distribution characteristics of nodes and branches; it includes: Based on the set of upstream nodes and the set of upstream branches in the fault propagation path of the information system, arbitrary nodes and branches are obtained. For any two nodes, if they both appear in the upstream set of other nodes and there is no direct upstream or downstream dependency, then the number of times they are structurally collinear is counted as one accumulation, and the number of times any two nodes are collinear is used as the node-node association degree. For any two branches, count the frequency with which they appear together in the set of upstream branches of a certain node, and there is no direct upstream or downstream path relationship between the target nodes connected by the two branches. The number of times any two branches are collinear is taken as the branch-branch correlation degree. Based on the frequency of co-occurrence of any node and its corresponding branch in a certain upstream path, direct dependencies are eliminated, and the number of times a node and its branch are collinear is obtained as the node-branch correlation degree. The comprehensive criticality calculation module is used to determine the generalized in-degree and generalized out-degree of each node based on the set of upstream nodes and the set of upstream branches; calculate the criticality of a single node based on the generalized in-degree and generalized out-degree; and calculate the comprehensive criticality of a single branch based on the criticality and frequency of occurrence of the nodes at both ends of a single branch. The fault point screening module uses the criticality of a single node and the comprehensive criticality of a single branch as the basis for initial fault point selection. Based on a greedy search method and the aforementioned correlation degree, it recursively expands the Nk fault point set to obtain the Nk fault point set of the information system; including: Obtain the criticality of all failure points; Select any fault point, and sort the fault points whose criticality of a single node or the combined criticality of a single branch is greater than the criticality threshold in descending order of criticality. This sorts them as the first fault combination scenario. Based on the greedy search method and the correlation, traverse all non-redundant fault points to generate k fault combination scenarios. Construct a set of selected fault points based on each fault point. S Thus, the Nk fault point set of the information system is obtained; The fault scenario generation model construction module is used to construct a fault scenario generation model based on the Nk fault point set of the information system, and to generate a fault scenario set.

9. A fault scenario generation system for an information system oriented towards AGC according to claim 8, characterized in that, In the fault propagation path establishment module, the information system node-branch association matrix is ​​as follows: Obtain m nodes and n branches in the information system, where nodes represent devices in the information system and branches represent information transmission channels between devices; construct the node-branch association matrix of the AGC system to describe the topological relationships between the nodes in the information system; the th node in the topological relationship... i line, number j Column matrix elements Represented as: 。 10. The information system fault scenario generation system for AGC according to claim 9, characterized in that, In the fault propagation path establishment module, the recursive search for the upstream node set and upstream branch set of each node based on the node-branch association matrix of the information system includes: Based on the node-branch association matrix of the information system, the BFS algorithm is used to traverse backwards from the target node to find all upstream branches of the target node. During the backward traversal, all upstream nodes of the target node are found simultaneously, and the set of upstream nodes and the set of upstream branches of each node are obtained based on all upstream nodes of the target node.

11. The information system fault scenario generation system for AGC according to claim 8, characterized in that, The fault point screening module is specifically used for: Based on the set of upstream nodes and the set of upstream branches for each node, determine the generalized in-degree and generalized out-degree of the node. The generalized in-degree of a node is the number of all upstream nodes that can be recursively traced to the node; the generalized out-degree of a node is the number of other nodes that consider the node as an upstream node. The larger of the generalized in-degree and generalized out-degree of a node is taken as the criticality of a single node; The comprehensive criticality of a single branch is obtained by weighting and fusing path frequency with the criticality of individual nodes: in, Crit(e j ) branch road The criticality; , All are weighting coefficients; Path frequency term The normalized value, Criterion for a single node The normalized value; For including branches The number of paths, yes The first in k Path; branch road The sum of the criticality of the two endpoints; It is a side road The ingress node, It is a side road The output node, Indicates the criticality of a single node.

12. The information system fault scenario generation system for AGC according to claim 8, characterized in that, The greedy search method and the correlation degree are used to traverse all non-redundant fault points to generate k fault combination scenarios; Construct a set of selected fault points based on each fault point. S The information system Nk fault point set is obtained, including: Obtain the correlation of all fault points; For each fault point x that is not included in the selected fault point set S, calculate its evaluation function. This represents the set of fault points and currently selected points. Structural relevance: in, This indicates the structural coupling degree between each fault point. x' For each fault point, The number of elements in the selected fault point set S; All evaluation functions Fault points that meet the evaluation threshold are added to the selected fault point set S to update the selected fault point set S; Repeat the above steps for all expanded fault point sets, and form a recursive candidate expansion process constrained by the aforementioned correlation degree, until each set contains k fault points, and output an Nk fault scenario set that satisfies the specified size of fault points k.

13. The information system fault scenario generation system for AGC according to claim 12, characterized in that, Based on the Nk fault point set of the information system, a fault scenario generation model is constructed to generate fault scenarios, including: Construct N-1 fault scenarios for the information system, and then obtain Nk fault scenarios; specifically including: Construct fault type matching constraints, specifically as follows: In the formula, A set of scenarios for a certain type of Nk fault; Fault point The set of fault types that it is allowed to occur. j These are index tags for different Nk fault scenario categories under the same k value; Based on the fault type matching constraints, an N-1 fault is constructed, represented as: In the formula, C is the set of all fault points, F is the set of all fault types, and T is the set of all fault time windows; The Nk fault combinations derived from N-1 faults are represented as follows: in, For the k-th fault point, For the k-th fault type, For the k-th fault occurrence time window, Let be the start time of the k-th fault. Let be the end time of the k-th fault; It is a set of combinations containing k fault instances; In the Nk fault scenarios, and based on the selected key nodes and branches, any fault scenario is generated by combining the fault point, fault type, and fault occurrence time. ,in, For the first i One fault point, For the first i One type of fault, For the first i A fault occurrence time window The start time, The end time; the fault time setting needs to take into account the measurement cycle, calculation cycle and control cycle in the AGC control process.

14. The information system fault scenario generation system for AGC according to claim 13, characterized in that, It also includes fault timing constraint methods, including: Get the control period window for the current analysis. ,in, It is the start time of the control cycle window for the current analysis; To control the cycle, a fault in a certain information element occurs at the end of the cycle. For the control output to have an impact, the following two conditions must be met: 1) The fault has not been resolved, and all components remain in a fault state at the end of the control cycle: In the formula, For the first i Fault type End time; It is a universal quantifier symbol, representing any one; 2) The fault has propagated. If a component fault is to affect control commands, the fault start time must be earlier than the maximum effective trigger time corresponding to the type. ,Right now: For the first i Fault type The start time; in, The value depends on the category to which the component belongs: In the formula, and These represent the measurement period and the calculation period, respectively. , and These represent information elements of measurement, calculation, and control types, respectively. An effective Nk fault combination satisfies the following constraints: 。 15. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the information system fault scenario generation method for AGC as described in any one of claims 1-7.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the information system fault scenario generation method for AGC as described in any one of claims 1-7.

17. A computer program product, the computer program product comprising computer instructions, characterized in that, The computer instructions instruct the computer to execute the information system fault scenario generation method for AGC as described in any one of claims 1-7.

Images