Intelligent power grid fault deduction method and device based on knowledge graph and electronic equipment

Abstract

The application discloses a kind of based on knowledge graph's intelligent power grid fault deduction method, device and electronic equipment, it is related to power grid fault diagnosis technical field, wherein method includes: receiving the protection action alarm information and circuit breaker tripping alarm information when target intelligent power grid occurs fault;Based on the protection action alarm information and the circuit breaker tripping alarm information utilize preset power grid topology knowledge graph to carry out hypothesis reasoning, obtain suspected fault equipment set;Based on the protection action alarm information and the circuit breaker tripping alarm information utilize preset protection configuration knowledge graph to each suspected fault equipment in suspected fault equipment set carries out deductive reasoning, obtains and each suspected fault equipment corresponding observation event confidence;Based on each observation event confidence to each suspected fault equipment carries out evidence fusion reasoning, obtains the actual fault information of intelligent power grid.The method of the application can improve the accuracy of intelligent power grid fault diagnosis result.

Description

Technical Field

[0001] This invention relates to the field of power grid fault diagnosis technology, and in particular to a method, device, storage medium, and electronic device for smart grid fault deduction based on knowledge graphs. Background Technology

[0002] Modern society is increasingly reliant on a reliable power supply. When a power grid fault occurs, dispatchers need to quickly determine the location and cause of the fault in order to handle it correctly. However, complex power grid faults generate a large amount of event information in a short period, posing significant challenges to dispatchers in quickly and accurately analyzing the fault. Therefore, complex power grid fault diagnosis technology has received widespread attention.

[0003] The development of power grid fault diagnosis technology has largely followed the development path of artificial intelligence (AI) technology. Early fault diagnosis primarily employed rule-based expert systems to handle the uncertainties arising from protection maloperation and failure to operate. However, expert systems have consistently faced significant bottlenecks in knowledge representation, acquisition, and updating, hindering the widespread application of this technology. As AI has entered the data-driven stage, related technologies have also been applied to fault diagnosis. However, current data-driven AI suffers from poor interpretability and requires a large number of training samples. Given the relatively low probability of complex power grid faults, the generalization ability of data-driven methods becomes highly uncertain when samples are insufficient. Existing methods for diagnosing smart grid faults are not accurate in reconstructing accidents. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, storage medium and electronic device for smart grid fault inference based on knowledge graph, the main purpose of which is to solve the problem of inaccurate power grid fault diagnosis.

[0005] To address the aforementioned problems, this application provides a knowledge graph-based method for smart grid fault estimation, comprising:

[0006] Receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid;

[0007] Based on the protection action alarm information and the circuit breaker trip alarm information, a set of suspected faulty devices is obtained by performing hypothetical reasoning using a preset power grid topology knowledge graph.

[0008] Based on the protection action alarm information and the circuit breaker trip alarm information, a preset protection configuration knowledge graph is used to perform deductive reasoning on each suspected faulty device in the suspected faulty device set to obtain the confidence level of the observed event corresponding to each suspected faulty device;

[0009] Based on the confidence level of each observed event, evidence fusion reasoning is performed on each suspected faulty device to obtain the actual fault information of the smart grid.

[0010] Optionally, before performing hypothetical reasoning based on the protection action alarm information and the circuit breaker trip alarm information using a preset power grid topology knowledge graph, the method further includes: constructing a preset power grid topology knowledge graph;

[0011] The construction of the preset power grid topology knowledge graph specifically includes:

[0012] Obtain the topology network structure diagram of the target smart grid;

[0013] Based on the topological network structure diagram, a model is constructed to obtain a CIM / E model corresponding to the target smart grid;

[0014] Based on the CIM / E model, a knowledge graph is constructed to obtain a preset power grid topology knowledge graph that represents the topological connection relationship between devices and corresponds to the target smart grid.

[0015] Optionally, the step of using a preset power grid topology knowledge graph to perform hypothetical reasoning based on the protection action alarm information and the circuit breaker trip alarm information to obtain a set of suspected faulty devices specifically includes:

[0016] Data is extracted based on the protection action alarm information and the circuit breaker tripping alarm information to obtain a set of tripping circuit breakers;

[0017] Construct the mapping relationship between each tripping circuit breaker in the tripping circuit breaker set to obtain each tripping circuit breaker group;

[0018] For each of the tripped circuit breaker groups, a first query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the first suspected fault device corresponding to the shortest path of each of the tripped circuit breaker groups;

[0019] For the primary line equipment in the first suspected faulty equipment, the second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the second suspected faulty equipment corresponding to the shortest path of each of the primary line equipment;

[0020] Perform a union calculation on the first suspected faulty device and the second suspected faulty device to obtain the set of suspected faulty devices;

[0021] The primary equipment includes lines, busbars, and transformers.

[0022] Optionally, before performing deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker trip alarm information using a preset protection configuration knowledge graph, the method further includes: constructing a preset protection configuration knowledge graph;

[0023] The construction of the preset protection configuration knowledge graph specifically includes:

[0024] The SCD file of the target smart grid is parsed to obtain the protection configuration information of the target smart grid. The protection configuration information includes the configuration information of primary equipment main protection and near backup protection, as well as the association between protection and tripping circuit breakers.

[0025] The second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain remote backup protection information;

[0026] Obtain the action time constraints of the backup protection action from the preset protection setting sheet;

[0027] Based on the protection configuration information, the remote backup protection information, and the action time constraints, a knowledge graph is constructed to obtain the preset protection configuration knowledge graph that characterizes the protection configuration and action time constraints of the primary equipment.

[0028] Optionally, the step of using a preset protection configuration knowledge graph to perform deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker trip alarm information to obtain the confidence level of the observed event corresponding to each suspected faulty device specifically includes:

[0029] Based on each of the suspected faulty devices, a third query statement is used to query the preset protection configuration knowledge graph to obtain the expected events corresponding to each of the suspected faulty devices and the time constraints for the occurrence of the expected events;

[0030] Data is filtered based on the protection action alarm information and the circuit breaker trip alarm information to obtain observed events;

[0031] Confidence calculations are performed based on the expected events and the observed events to obtain the confidence scores of the observed events corresponding to each suspected faulty device.

[0032] Optionally, the step of performing confidence calculation based on the expected event set and the observed event set to obtain the confidence score of the observed event corresponding to each suspected faulty device specifically includes:

[0033] The first index value is determined based on the actual time of the observed event and the time constraint of the observed event occurring in the preset protection configuration knowledge graph.

[0034] The second index value is determined based on the type of protection action observed in the event.

[0035] Based on the first and second index values, a pre-set confidence function is used to calculate and process the observed event confidence, so as to obtain the observed event confidence corresponding to each of the suspected faulty devices.

[0036] Optionally, the step of performing evidence fusion reasoning on each of the suspected faulty devices based on the confidence levels of each of the observed events to obtain the actual fault information of the smart grid fault specifically includes:

[0037] Evidence is added to the confidence scores of each observed event to obtain the target confidence matrix;

[0038] Based on the target confidence matrix, the DS evidence fusion method is used to perform evidence fusion to obtain the evidence fusion result;

[0039] Based on the evidence fusion results, the protection action alarm information when the target smart grid fails, and the circuit breaker tripping alarm information, the fault is reconstructed using the preset protection configuration knowledge graph to obtain the actual fault information of the target smart grid.

[0040] To address the aforementioned issues, this application discloses a knowledge graph-based smart grid fault prediction device, comprising:

[0041] The receiving module is used to receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid;

[0042] The hypothesis reasoning module is used to perform hypothesis reasoning based on the protection action alarm information and the circuit breaker tripping alarm information using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices.

[0043] The deductive reasoning module is used to perform deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker tripping alarm information using a preset protection configuration knowledge graph, and to obtain the confidence level of the observed event corresponding to each suspected faulty device.

[0044] The evidence fusion reasoning module is used to perform evidence fusion reasoning on each of the suspected faulty devices based on the confidence level of each of the observed events, so as to obtain the actual fault information of the smart grid.

[0045] To address the aforementioned issues, this application provides a storage medium storing a computer program that, when executed by a processor, implements the steps of the knowledge graph-based smart grid fault prediction method described above.

[0046] To address the aforementioned issues, this application provides an electronic device, comprising at least a memory and a processor. The memory stores a computer program, and the processor, when executing the computer program in the memory, implements the steps of the aforementioned knowledge graph-based smart grid fault prediction method.

[0047] The beneficial effects of this application are as follows: This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, an accident simulation framework incorporating hypothetical reasoning, deductive reasoning, and fusion reasoning is proposed, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its ability to improve knowledge sharing, scalability, and interpretability through explicit knowledge representation and independent reasoning mechanisms. It also effectively handles time constraints and conflicting multi-source evidence, resulting in more accurate reconstruction of smart grid faults.

[0048] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0049] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0050] Figure 1 The illustration shows a flowchart of a knowledge graph-based smart grid fault inference method provided in an embodiment of this application;

[0051] Figure 2 This illustration shows a flowchart of a knowledge graph-based smart grid fault estimation method according to another embodiment of this application.

[0052] Figure 3 A simple line diagram of the topology of a power grid according to an embodiment of this application is shown.

[0053] Figure 4 This application shows a CIM / E model corresponding to the single-line diagram of the topological network structure according to an embodiment of the present application;

[0054] Figure 5 This application illustrates a power grid topology knowledge graph corresponding to the single-line diagram of the topology network structure described in an embodiment of the present application.

[0055] Figure 6 This application illustrates a knowledge graph of the protection configuration of line L2 according to an embodiment of the present application;

[0056] Figure 7 A structural block diagram of a knowledge graph-based smart grid fault prediction device is shown in another embodiment of this application;

[0057] Figure 8 A schematic diagram of the power grid structure for a single fault mode in a computational power grid according to an embodiment of this application is shown.

[0058] Figure 9 A schematic diagram of the power grid structure for multiple fault modes of a computational power grid according to an embodiment of this application is shown. Detailed Implementation

[0059] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0060] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this application will be apparent to those skilled in the art.

[0061] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present application and, together with the general description of the present application given above and the detailed description of the embodiments given below, serve to explain the principles of the present application.

[0062] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0063] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application.

[0064] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0065] Specific embodiments of this application are described thereafter with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in a variety of substantially any suitable detailed structures.

[0066] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0067] This application provides a method for smart grid fault estimation based on knowledge graphs, such as... Figure 1 As shown, it includes:

[0068] Step S101: Receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid;

[0069] In this step, the fault zero moment is determined through a fault recording device and protection activation information, and protection action alarm information and circuit breaker tripping alarm information are received. Specifically, when a fault occurs in the target smart grid, the fault isolation is ultimately achieved by circuit breaker tripping. The tripping circuit breaker defines the fault area. The fault recording device is an important device in the power system used to record changes in electrical quantities when a fault occurs. When a fault occurs in the power grid, the fault recording device automatically starts, recording the waveforms and effective values ​​of electrical quantities such as current and voltage before and after the fault. This recorded data can be used to analyze the nature and type of the fault, as well as the specific time of the fault occurrence, i.e., the fault zero moment. The fault recording device determines the fault zero moment by detecting sudden changes in current or voltage. The protection device will activate when a fault is detected and issue a protection action signal. Protection activation information includes the action time and action type of the protection device; the circuit breaker will trip when a fault occurs to isolate the fault area. Circuit breaker alarm information includes the tripping time and reason for the tripping.

[0070] Step S102: Based on the protection action alarm information and the circuit breaker trip alarm information, perform hypothetical reasoning using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices;

[0071] In this step, data is extracted based on the protection action alarm information and the circuit breaker tripping alarm information to obtain a set of tripped circuit breakers; a mapping relationship is constructed for each tripped circuit breaker in the set to obtain each tripped circuit breaker group; for each tripped circuit breaker group, a first query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the primary equipment of the shortest path corresponding to each tripped circuit breaker group, wherein the primary equipment includes lines, buses and transformers, to obtain the set of suspected faulty equipment.

[0072] Step S103: Based on the protection action alarm information and the circuit breaker trip alarm information, use a preset protection configuration knowledge graph to perform deductive reasoning on each suspected faulty device in the suspected faulty device set to obtain the confidence level of the observed event corresponding to each suspected faulty device;

[0073] In the specific implementation process of this step, the preset protection configuration knowledge graph is retrieved based on each suspected faulty device to obtain the expected events corresponding to each suspected faulty device and the time constraints for the occurrence of the expected events; data filtering is performed based on the protection action alarm information and the circuit breaker tripping alarm information to obtain the observed events; confidence calculation processing is performed based on the expected events and the observed events to obtain the confidence of the observed events corresponding to each suspected faulty device.

[0074] Step S104: Based on the confidence level of each observed event, perform evidence fusion reasoning on each suspected faulty device to obtain the actual fault information of the smart grid.

[0075] In this step, evidence is supplemented to improve the confidence of each observed event to obtain a target confidence matrix. Based on the target confidence matrix, the DS evidence fusion method is used to perform evidence fusion to obtain the evidence fusion result. Based on the evidence fusion result, the protection action alarm information and circuit breaker tripping alarm information when the target smart grid fails, the fault is reconstructed using the preset protection configuration knowledge graph to obtain the actual fault information of the target smart grid.

[0076] This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, it proposes an accident simulation framework that includes hypothetical reasoning, deductive reasoning, and fusion reasoning, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its explicit knowledge representation and independent reasoning mechanism, which not only improves the sharing, scalability, and interpretability of knowledge but also effectively handles time constraints and conflicting multi-source evidence. The method in this application provides more accurate reconstruction of smart grid faults.

[0077] Another embodiment of this application provides a different method for smart grid fault estimation based on knowledge graphs, such as... Figure 2 As shown, it includes:

[0078] Step S201: Receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid;

[0079] In this step, the fault zero moment (i.e., the moment when the target smart grid experiences a fault) is determined using a fault recording device and protection activation information. Protection action alarm information and circuit breaker trip alarm information are then received. These alarm information can be collected in real-time using the fault recording device.

[0080] Step S202: Construct a pre-defined power grid topology knowledge graph;

[0081] In the specific implementation process of this step, the topology network structure diagram of the target smart grid is obtained; such as Figure 3 The diagram shown is a simple linear topology diagram of a power grid. Based on this topology diagram, a model is constructed to obtain a CIM / E model corresponding to the target smart grid; for example, when the target smart grid is... Figure 3 When the topology network is shown, such as Figure 4 As shown, this is the CIM / E model corresponding to the target smart grid. Based on the CIM / E model, a knowledge graph is constructed to obtain a knowledge graph representing the topological connections between devices and corresponding to the preset power grid topology of the target smart grid. For example... Figure 5 As shown, this is the power grid topology knowledge graph corresponding to the target smart grid; the dispatching system has widely adopted the CIM / E model to describe the primary system topology. Since there is a clear mapping relationship between the CIM / E model and the power grid topology knowledge graph model, the power grid topology knowledge graph can be automatically constructed based on the CIM / E model. The attached figures of this application are only used to explain this application with a simple topology structure. This application can also be applied to more complex target smart grids. The preset power grid topology knowledge graph of this application is a semantic network that reveals entities and their relationships; the power grid topology knowledge graph constructed in this application is used to describe the topology relationships of the entire network. This application formalizes the preset power grid topology knowledge graph as the following undirected graph G, as shown in the following formula (1):

[0082]

[0083] Where V is the set of vertices and E is the set of edges; the vertex set contains the set of primary equipment (EQ) and the set of electrical connection nodes (CN). In the power grid topology knowledge graph, each vertex represents a primary equipment or electrical connection node, and each edge is an unordered pair (x, y), representing the topological connection relationship between equipment (interconnected through electrical connection nodes).

[0084] Step S203: Based on the protection action alarm information and the circuit breaker trip alarm information, perform hypothetical reasoning using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices;

[0085] In the specific implementation process of this step, data is extracted based on the protection action alarm information and the circuit breaker tripping alarm information to obtain the set of tripped circuit breakers; when a fault occurs in the power grid, the fault isolation is ultimately achieved by the circuit breaker tripping. The scope of the tripped circuit breaker defines the fault area, and the primary equipment within this scope is considered as suspected faulty equipment. Assume that the reasoning task can be formally described as shown in the following formula (2):

[0086]

[0087] in, I represents the set of suspected faulty devices. A mapping relationship is constructed between each tripped circuit breaker in the tripped circuit breaker set to obtain each tripped circuit breaker group. Hypothetical reasoning infers the fault range and forms a fault hypothesis set based on the action results. Its goal is to reduce the search space for subsequent fault reasoning without omitting any actual faulty devices. The tripped circuit breaker information is then aggregated into a set. Where O represents the total number of tripped circuit breakers. This set integrates circuit breaker trip alarm and protection action alarm information; this invention assumes that these two types of information will not be lost simultaneously. Constructing a complete set of tripped circuit breakers... For each of the tripped circuit breaker groups, a first query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain a first suspected faulty device corresponding to the shortest path of each tripped circuit breaker group; for the primary line equipment among the first suspected faulty devices, a second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain a second suspected faulty device corresponding to the shortest path of each primary line equipment; the first suspected faulty device and the second suspected faulty device are then combined to obtain the set of suspected faulty devices. The primary equipment includes lines, busbars, and transformers. The first query statement can be: `MATCH path = shortestpath((:CB{name:“CB3”})-[:CONNECTED_TO*]-(:CB{name:“CB6”}))RETURN path`; the second query statement can be: `MATCH path = shortestpath((:Line{name:'L2'})-[:CONNECTED_TO*]-(l:Line))RETURN l`; Figure 5 For example, assuming the alarm information of the protection and circuit breakers observed after a fault is shown in Table 1, the following set of suspected faulty devices can be obtained through hypothetical reasoning. It can be represented by the following formula (3):

[0088]

[0089] Table 1 Alarm Information for Protection Devices and Circuit Breakers

[0090] Zero-time failure 38ms Protection action alarm information <![CDATA[L2B 2m (50ms), L2B 3m (51ms), L3B 4b (700ms)]]> Circuit breaker trip alarm information <![CDATA[CB3(65ms),CB9(300ms),CB6(725ms)]]>

[0091] Step S204: Construct a knowledge graph of the preset protection configuration;

[0092] In this step, the SCD file of the target smart grid is parsed to obtain the protection configuration information of the target smart grid. This protection configuration information includes the primary equipment main protection, near-backup protection configuration information, and the association between protection and tripping circuit breakers. The protection configuration information can be automatically extracted by parsing the SCD file. A second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the far-backup protection information. The action time constraints of the backup protection actions are obtained from the preset protection setting sheet. Based on the protection configuration information, the far-backup protection information, and the action time constraints, a knowledge graph is constructed to obtain the preset protection configuration knowledge graph representing the protection configuration and action time constraints of the primary equipment. Figure 6 The diagram shows the protection configuration knowledge graph for line L2. The protection configuration knowledge graph constructed in this invention describes the protection configuration and action time constraints of primary equipment. This invention formally represents it as the following directed graph G, which can be represented by the following formula (4):

[0093]

[0094] Among them, E EQ E CB and E PR These represent the primary equipment fault event set, the circuit breaker tripping and closing event set, and the protection action event set, respectively; each vertex v represents one event from the above sets; each edge represents an ordered pair of events.<p,q> This indicates that events p and q have a causal relationship; T maps the event set E to the set of real numbers. That is, each set of events<p,q> Assign a real number as the weight of the edge to represent the time constraint between p and q. Suppose that event p causes event q, and the time of occurrence of q, relative to p, must fall within the interval [t]. min ,t max In ], t min ,t max ≥0. To represent the above causal relationships and time constraints in the protected configuration knowledge graph, this invention sets the weights according to the following rules: the weight of the directed edge from p to q is set to t. max The weight of the directed edge from q to p is set to -t. min Therefore, the protection configuration knowledge graph is a directed graph containing negative-weighted edges. Still using... Figure 3 Taking a simple power grid as an example, Figure 6 A knowledge graph of protection configurations related to line L2 is displayed, with L2 as the root vertex. The graph uses the convention: L2B. 3m This indicates the protection for line L2 near busbar B3; the subscripts m, f, and b represent the main protection, failure protection, and backup protection, respectively. For example, Figure 6 Mid-vertices L2 and L2B 3m The two directed edges between them indicate that if L2 fails, the operating time constraint of the main protection of the line near B3 is [10,40]ms.

[0095] Step S205: Based on each of the suspected faulty devices, use a third query statement to query the preset protection configuration knowledge graph to obtain the expected events corresponding to each of the suspected faulty devices and the time constraints for the occurrence of the expected events;

[0096] In the specific implementation process of this step, after obtaining a set of suspected faulty devices... Then, for each suspected faulty device Deductive reasoning is performed. Its formal mathematical description can be represented by the following formula (5):

[0097]

[0098] Where p→q is the protection rule in equation (4), which is represented by the protection configuration knowledge graph; If For faulty equipment, the expected set of protection and circuit breaker actions according to rule p→q is referred to as the expected event set in this invention. Thus, the accident simulation problem is transformed into a typical deductive reasoning problem, which can then utilize a knowledge graph. Through retrieval and... Related preset protection configuration knowledge graph (e.g.) Figure 5 By combining this with the all-correct time constraint table constructed using the third query statement, we can obtain... The time constraints are then determined. This completes the deductive reasoning of equation (5). A third query statement is used to query the all-pairs time constraints between events in the protected configuration knowledge graph. This query can be completed using the All-Pairs Shortest Path algorithm; this invention uses the Floyd-Warshall algorithm, which supports negative weight edges. Figure 6Taking the protection configuration knowledge graph as an example, the result of the third query statement Query3 can form the all-time constraint table in Table 2. For example, from the table, we can see that the shortest distance from line L2 to circuit breaker CB6 is 1140ms, while the shortest distance from CB6 to L2 is -620ms. This means that if line L2 fails, the time constraint for the remote tripping event of circuit breaker CB6 is [620, 1140]ms. Using Table 2, we can quickly determine whether the time constraints between any two events are met. This provides an important basis for implementing fault reasoning.

[0099] Table 2. Time Constraints for All Correct Answers (ms)

[0100] vertex <![CDATA[L2]]> <![CDATA[L2B 2m ]]> <![CDATA[L2B 3m ]]> <![CDATA[L3B 4b ]]> <![CDATA[CB3]]> <![CDATA[CB6]]> <![CDATA[L2]]> 0 40 40 1100 80 1140 <![CDATA[L2B 2m ]]> -10 0 30 1090 40 1130 <![CDATA[L2B 3m ]]> -10 30 0 1090 70 1130 <![CDATA[L3B 4b ]]> -600 -560 -560 0 -520 40 <![CDATA[CB3]]> -30 -20 10 1070 0 1110 <![CDATA[CB6]]> -620 -580 -580 -20 -540 0

[0101] Step S206: Based on the protection action alarm information and the circuit breaker trip alarm information, perform data filtering to obtain observed events;

[0102] In the specific implementation process of this step, due to It may not be a genuine faulty device, and is also affected by complex factors such as protection malfunctions and missing information. Possibly related to the actual observed event set Not entirely consistent. Data filtering is performed based on the protection action alarm information and the circuit breaker trip alarm information to obtain observed events.

[0103] Step S207: Perform confidence calculation processing based on the expected event and the observed event to obtain the confidence of the observed event corresponding to each suspected faulty device;

[0104] In the specific implementation process of this step, due to It may not be a genuine faulty device, and is also affected by complex factors such as protection malfunctions and missing information. Possibly related to the actual observed event set Not entirely consistent. This invention assesses the confidence level of the consistency between the two, a calculation process that can be formalized as follows: (6)

[0105]

[0106] Where, m ij Indicating the fault assumption Below, observed events The confidence level is determined based on the actual time of the observed event and the time constraint between the observed event and the time of the observed event in the preset protection configuration knowledge graph. The observed event is judged according to the all-pair time constraint table. Does it meet the time constraint? When the time constraint is met, the first index value r is determined to be 1; If a value appears in the full time constraint table but exceeds the time constraint limit (possibly due to slow protection or circuit breaker operation), the first index value r will be set to 0.5. If an event does not appear in the full time constraint table or is below the lower limit of the time constraint, the first indicator value r is determined to be 0. Based on the protection action type of the observed event, a second indicator value is determined; the second indicator value D is used to assign weights according to the protection action type: specifically, for a main protection action or associated circuit breaker tripping, the second indicator value D is determined to be 1; for a failure protection action or associated circuit breaker tripping, the second indicator value D is determined to be 2; for a backup protection action or associated circuit breaker tripping, the second indicator value D is determined to be 3. Based on the first and second indicator values, a pre-set confidence function is used to calculate the confidence level of the observed event, thus obtaining the confidence level m of the observed event corresponding to each suspected faulty device. ij The confidence level m of the observed event ij The mathematical formula for calculating it can be shown in the following formula (7):

[0107]

[0108] by Figure 3 Taking Table 1 as an example, the calculated m ij By placing these values ​​in a two-dimensional table of size I×J, we can obtain a summary table of confidence levels representing the confidence matrix, as shown in Table 3.

[0109] Table 3 Confidence Level Summary Table

[0110] Suspected faulty equipment <![CDATA[L2B 2m ]]> <![CDATA[L2B 3m ]]> <![CDATA[L3B 4b ]]> <![CDATA[CB3]]> <![CDATA[CB6]]> <![CDATA[CB9]]> <![CDATA[B2]]> 0 0 0 0 0 0 <![CDATA[B3]]> 0 0 0.125 0 0.125 0 <![CDATA[L2]]> 1 1 0.125 1 0.125 0 <![CDATA[L3]]> 0 0 0 0 0 0 <![CDATA[L4]]> 0 0 0 0 0 0

[0111] Step S208: Supplement the confidence scores of each observed event with evidence to obtain the target confidence matrix;

[0112] In this step, the confidence summary table representing the confidence matrix is ​​supplemented to address cases of insufficient evidence, and evidence is added to the confidence scores that cannot be attributed to any faulty device to obtain the target confidence matrix. The target confidence scores of the target confidence matrix are summarized as shown in Table 4 below:

[0113] Table 4 Summary of Target Confidence Levels

[0114]

[0115] Step S209: Based on the target confidence matrix, perform evidence fusion using the DS evidence fusion method to obtain the evidence fusion result;

[0116] In the specific implementation process of this step, if each column in Table 4 is regarded as an independent source of evidence, the next question is: how to integrate J sources of evidence to determine the fault of I suspected faulty equipment. To this end, this application introduces the DS evidence fusion theory for final reasoning.

[0117] To address the issues raised in this application, the identification framework Θ for the evidence fusion problem is a set of suspected faulty devices. Within this identification framework, the confidence function is m:2 Θ →[0,1] should satisfy the following mathematical expression (8):

[0118]

[0119] in: Let A be a proposition, and m(A) be the confidence level of A.

[0120] Without loss of generality, suppose there are two mutually exclusive sources of evidence, m1 and m2. Then the fusion result for proposition A is as shown in the following formula (9):

[0121]

[0122] in: The 1-K in the denominator is to ensure that the fusion result still satisfies the normalization condition of equation (8). The fusion of Table 4 using the above formula is shown in Table 5. The additional rows at the bottom of the table are assigned confidence levels for failure to be attributed to any faulty device, indicating cases of insufficient evidence.

[0123] Table 5 Evidence Fusion Results

[0124]

[0125] Step S210: Based on the evidence fusion results, the protection action alarm information when the target smart grid fails, and the circuit breaker trip alarm information, the fault is reconstructed using the preset protection configuration knowledge graph to obtain the actual fault information of the target smart grid.

[0126] In the specific implementation process of this step, after obtaining the evidence fusion results in Table 5, it can be combined with the original alarm information in Table 1 and... Figure 6 The protection configuration knowledge graph provides a complete explanation of the accident process. The specific implementation steps are as follows:

[0127] Step 1: Identify the faulty equipment. Based on the "Evidence Fusion Result" column in Table 5, determine the suspected equipment corresponding to the maximum value as the actual faulty equipment. If the maximum value is lower than the evidence fusion result, the evidence fusion is deemed invalid, and the faulty equipment cannot be identified.

[0128] Step Two: Determine the Action of Protection and Circuit Breaker. After confirming the actual faulty equipment, further analyze the action of the protection and circuit breaker:

[0129] ①False alarm event: If an observed event is assigned a confidence level of 0, then the event is determined to be a false alarm;

[0130] ② Failure to operate event: Based on the protection configuration knowledge graph of the actual faulty equipment, if a protection or circuit breaker should have operated but was not observed, it is determined to be a failure to operate.

[0131] Step 3: Reconstruct the accident process. Finally, by combining the operating timing of the protection devices and circuit breakers, the entire accident process is reconstructed. Details are as follows:

[0132] Line L2 failed at 38ms, L2B 2m L2B 3m The circuit breakers operated at 50ms and 51ms respectively, with CB3 tripping at 65ms, while CB4 failed to operate; L3B 4b CB6 tripped at 725ms, while CB9 malfunctioned at 300ms.

[0133] This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, it proposes an accident simulation framework that includes hypothetical reasoning, deductive reasoning, and fusion reasoning, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its explicit knowledge representation and independent reasoning mechanism, which not only improves the sharing, scalability, and interpretability of knowledge but also effectively handles time constraints and conflicting multi-source evidence. The method in this application provides more accurate reconstruction of smart grid faults.

[0134] The following example illustrates the derivation process of this application:

[0135] This section establishes a fault simulation example based on the actual topology and protection configuration of four 500kV substations in the XX power grid, namely "Area A – Area B – Area C – Area D". The topology structure of the example is as follows: Figure 8 , 9 As shown in the diagram, substation identifiers have been added, and the protection adopts a dual configuration. For example, L1S1 indicates the S1 side of the substation for line L1, and B1m,1 indicates the first set of main protection for bus B1.

[0136] exist Figure 6 In this scenario, a fault occurs on line L2. The system determines the fault zero moment through the fault recording device and protection activation information, and receives the protection and circuit breaker alarm information shown in Table 6.

[0137] Table 6 Alarm information of protection and circuit breakers during a single fault.

[0138] Zero-time failure 50ms Protection action alarm information <![CDATA[L2S 1m，1 (62ms),L2S 2m,1 (63ms),L2S 1m，2 (64ms),L2S 2m2 (65ms),5033 f (356ms)]]> Circuit breaker trip alarm information <![CDATA[CB 5031 (85ms),CB 5061 (86ms),CB 5062 (87ms),CB 5081 (377ms),CB 5013 (380ms),CB 5023 (382ms)]]>

[0139] Step 1: Hypothetical Reasoning: Based on the power grid topology knowledge graph of the example power grid and Table 6, a set of suspected faulty devices is obtained through hypothetical reasoning.

[0140] Step 2, Deductive Reasoning: Based on Table 6 and the knowledge graph of protection configuration of suspected faulty equipment, the confidence level of each event is obtained by deductive reasoning, as shown in Table 7.

[0141] Table 7 Confidence Level Summary Table

[0142] Suspected faulty equipment <![CDATA[L2S 1m，1 ]]> <![CDATA[L2S 2m，1 ]]> <![CDATA[L2S 1m,2 ]]> <![CDATA[L2S 2m,2 ]]> <![CDATA[5033 f ]]> <![CDATA[CB 5032 ]]> <![CDATA[CB 5061 ]]> <![CDATA[CB 5062 ]]> <![CDATA[CB 5081 ]]> <![CDATA[CB 5013 ]]> <![CDATA[CB 5023 ]]> <![CDATA[B2]]> 0 0 0 0 0.5 0 0 0 0 0.5 0.5 <![CDATA[B5]]> 0 0 0 0 0 0 0 0 0.5 0 0 <![CDATA[L2]]> 1 1 1 1 0.5 1 1 1 0 0.5 0.5 <![CDATA[L3]]> 0 0 0 0 0 0 0 0 0.5 0 0.5 <![CDATA[L4]]> 0 0 0 0 0 1 0 0 0.5 0 0

[0143] Step 3: Evidence fusion reasoning: Evidence fusion is performed on Table 7, and the results are shown in Table 8.

[0144] Table 8. Results of Evidence Fusion

[0145]

[0146] Step 4: Explanation of the Accident Process: Based on the evidence fusion results in Table 8, it can be seen that: Line L2 experienced a fault at 50ms, and the main protection L2S... 1m,1 and L2S 2m,1 L2S operates at 62ms and 63ms. 1m,2 and L2S 2m,2 Operating at 64ms and 65ms; Circuit breaker CB 5032 CB 5061 CB 5062 Tripping at 85ms, 86ms, and 87ms, CB 5033 Refusal to operate; Malfunction protection 5033 f The circuit breaker CB operated at 356ms. 5013 and CB 5023 The circuit breaker CB tripped at 380ms and 382ms. 5081 It malfunctioned at 377ms.

[0147] The method in this application can also be used to simulate accidents with multiple failures, as shown in the following specific examples:

[0148] exist Figure 9 During the process, faults occurred successively on lines L2 and L4. After the faults ended, the system received the protection and circuit breaker alarm information shown in Table 9. Through fault recording and protection activation information, two fault zero moments were detected.

[0149] Table 9. Alarm information for protection and circuit breaking during multiple faults.

[0150]

[0151] Step 1: Hypothetical Reasoning: Based on the power grid topology knowledge graph of the example power grid and Table 9, a set of suspected faulty devices is obtained through hypothetical reasoning.

[0152] Step 2, Deductive Reasoning: Based on the set of suspected faulty devices, deductive reasoning is performed at two specific moments when the fault occurs.

[0153] 1) Taking 40ms as the zero time of the fault, and combining the protection configuration knowledge graph and Table 9, the reasoning results in Table 10 are obtained.

[0154] 2) Taking 776ms as the zero time of the fault, and combining the protection configuration knowledge graph and Table 9, the reasoning results in Table 11 are obtained.

[0155] Table 10: Summary of Reliability of First Failure

[0156] Suspected faulty equipment <![CDATA[B 2m，1 ]]> <![CDATA[B 2m，2 ]]> <![CDATA[L4S 1m，1 ]]> <![CDATA[L4S 3m，1 ]]> <![CDATA[L4S 1m，2 ]]> <![CDATA[L4S 3m，2 ]]> <![CDATA[5033 f ]]> <![CDATA[CB 5013 ]]> <![CDATA[CB 5023 ]]> <![CDATA[CB 5032 ]]> <![CDATA[CB 5081 ]]> <![CDATA[CB 5061 ]]> <![CDATA[CB 5062 ]]> <![CDATA[CB 5031 ]]> <![CDATA[CB 5071 ]]> <![CDATA[CB 5072 ]]> <![CDATA[B2]]> 1 1 0 0 0 0 0.5 1 1 0.5 0 0.5 0.5 0 0.125 0.125 <![CDATA[B5]]> 0 0 0 0 0 0 0 0 0 0.5 0.5 0 0 0.25 0.5 0.25 <![CDATA[L2]]> 0 0 0 0 0 0 0.5 0 0 0 0 0.5 0.5 0.125 0.25 0.25 <![CDATA[L3]]> 0 0 0 0 0 0 0 0 0 0 0.5 0 0 0.125 0.125 0.125 <![CDATA[L4]]> 0 0 0.5 0.5 0.5 0.5 0 0 0 0.5 0 0.25 0.25 0.5 0.5 0.5

[0157] Table 11 Summary of the Reliability of the Second Failure

[0158] Suspected faulty equipment <![CDATA[B 2m，1 ]]> <![CDATA[B 2m,2 ]]> <![CDATA[L4S 1m，1 ]]> <![CDATA[L4S 3m，1 ]]> <![CDATA[L4S 1m，2 ]]> <![CDATA[L4S 3m，2 ]]> <![CDATA[5033 f ]]> <![CDATA[CB 5013 ]]> <![CDATA[CB 5023 ]]> <![CDATA[CB 5032 ]]> <![CDATA[CB 5081 ]]> <![CDATA[CB 5061 ]]> <![CDATA[CB 5062 ]]> <![CDATA[CB 5031 ]]> <![CDATA[CB 5071 ]]> <![CDATA[CB 5072 ]]> <![CDATA[B2]]> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <![CDATA[B5]]> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.5 0 <![CDATA[L2]]> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <![CDATA[L3]]> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 <![CDATA[L4]]> 0 0 1 1 1 1 0 0 0 0 0 0 0 1 1 1

[0159] Step 3: Integration Reasoning: Based on Tables 10 and 11, Tables 12 and 13 are obtained by integrating the evidence.

[0160] Table 12 Evidence fusion results of the first failure

[0161]

[0162] Table 13 Evidence fusion results of the second fault

[0163]

[0164] Step 4: Explanation of the Accident Process: Based on the evidence fusion results in Table 12, busbar B2 is determined to be the first actual fault device; based on Table 13, line L4 is determined to be the second actual fault device. Circuit breaker CB 5081 The confidence level for both busbar B2 and line L4 is 0, indicating a false trip event. Based on the protection configuration knowledge graph for busbar B2, circuit breaker CB... 5033 The circuit breaker did not trip, indicating a failure to operate event. The complete fault process is as follows:

[0165] Initial fault: A fault occurred on bus B2 at 40ms, triggering the main protection B. 2m,1 and B 2m,2 It operates at 52ms and 54ms; Circuit breaker CB 5013 CB 5023 Tripping at 75ms and 76ms, CB 5033Refusal to operate; Malfunction protection 5033 f The circuit breaker CB tripped at 350ms. 5032 CB 5061 and CB 5062 It tripped at 377ms, 577ms and 579ms.

[0166] Second fault: Line L4 experienced a fault at 776ms, triggering the main protection L4S. 1m,1 and L4S 3m,1 Actions at 790ms and 791ms, L4S 1m,2 and L4S 3m,2 It operates at 794ms and 795ms; Circuit breaker CB 5031 CB 5071 and CB 5072 It tripped at 815ms, 817ms, and 818ms.

[0167] Circuit breaker CB 5081 It malfunctioned after 300ms.

[0168] Another embodiment of this application provides a smart grid fault prediction device based on a knowledge graph, such as... Figure 7 The following are included:

[0169] Receiver module 1 is used to receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid;

[0170] Hypothesis reasoning module 2 is used to perform hypothesis reasoning based on the protection action alarm information and the circuit breaker tripping alarm information using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices;

[0171] The deductive reasoning module 3 is used to perform deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker tripping alarm information using a preset protection configuration knowledge graph, and to obtain the confidence level of the observed event corresponding to each suspected faulty device.

[0172] The evidence fusion reasoning module 4 is used to perform evidence fusion reasoning on each of the suspected faulty devices based on the confidence level of each of the observed events, so as to obtain the actual fault information of the smart grid.

[0173] In specific implementation, the device further includes a preset power grid topology knowledge graph construction module, which is specifically used for: obtaining the topology network structure diagram of the target smart grid; constructing a model based on the topology network structure diagram to obtain a CIM / E model corresponding to the target smart grid; and constructing a knowledge graph based on the CIM / E model to obtain the preset power grid topology knowledge graph corresponding to the target smart grid, representing the topological connection relationship between devices.

[0174] In the specific implementation process, the hypothesis reasoning module 2 is specifically used for: extracting data based on the protection action alarm information and the circuit breaker tripping alarm information to obtain a set of tripped circuit breakers; constructing a mapping relationship between each tripped circuit breaker in the set to obtain each tripped circuit breaker group; for each tripped circuit breaker group, using a first query statement to perform a shortest path query on the preset power grid topology knowledge graph to obtain a first suspected faulty device corresponding to the shortest path of each tripped circuit breaker group; for the primary equipment of the line in the first suspected faulty device, using a second query statement to perform a shortest path query on the preset power grid topology knowledge graph to obtain a second suspected faulty device corresponding to the shortest path of each primary equipment of the line; performing a union calculation on the first suspected faulty device and the second suspected faulty device to obtain the set of suspected faulty devices; wherein, the primary equipment includes lines, buses, and transformers.

[0175] In specific implementation, the device further includes: a preset protection configuration knowledge graph construction module, which is specifically used for: parsing the SCD file of the target smart grid to obtain the protection configuration information of the target smart grid, the protection configuration information including the primary equipment main protection, near backup protection configuration information, and the association between protection and tripping circuit breakers; using a second query statement to perform a shortest path query on the preset power grid topology knowledge graph to obtain far backup protection information; obtaining the action time constraint of backup protection action from the preset protection setting sheet; and constructing a knowledge graph based on the protection configuration information, the far backup protection information, and the action time constraint to obtain the preset protection configuration knowledge graph used to characterize the protection configuration and action time constraint of the primary equipment.

[0176] In the specific implementation process, the deductive reasoning module 3 is specifically used to: query the preset protection configuration knowledge graph based on each of the suspected faulty devices using a third query statement to obtain the expected events corresponding to each of the suspected faulty devices and the time constraints for the occurrence of the expected events; perform data filtering based on the protection action alarm information and the circuit breaker tripping alarm information to obtain the observed events; and perform confidence calculation processing based on the expected events and the observed events to obtain the confidence of the observed events corresponding to each of the suspected faulty devices.

[0177] In the specific implementation process, the deductive reasoning module 3 is also used to: determine a first index value based on the actual time of the observed event and the time constraint of the observed event in the preset protection configuration knowledge graph; determine a second index value based on the protection action type of the observed event; and calculate the observed event confidence level by using a preset confidence function based on the first index value and the second index value, so as to obtain the observed event confidence level corresponding to each of the suspected faulty devices.

[0178] In the specific implementation process, the evidence fusion reasoning module 4 is specifically used to: supplement the confidence of each observation event with evidence to obtain a target confidence matrix; perform evidence fusion using the DS evidence fusion method based on the target confidence matrix to obtain the evidence fusion result; and reconstruct the fault using the preset protection configuration knowledge graph based on the evidence fusion result, the protection action alarm information when the target smart grid fails, and the circuit breaker tripping alarm information to obtain the actual fault information of the target smart grid.

[0179] This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, it proposes an accident simulation framework that includes hypothetical reasoning, deductive reasoning, and fusion reasoning, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its explicit knowledge representation and independent reasoning mechanism, which not only improves the sharing, scalability, and interpretability of knowledge but also effectively handles time constraints and conflicting multi-source evidence. The method in this application provides more accurate reconstruction of smart grid faults.

[0180] Another embodiment of this application provides a storage medium storing a computer program, which, when executed by a processor, implements the following method steps:

[0181] Step 1: Receive protection action alarm information and circuit breaker tripping alarm information when a fault occurs in the target smart grid;

[0182] Step 2: Based on the protection action alarm information and the circuit breaker trip alarm information, perform hypothetical reasoning using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices;

[0183] Step 3: Based on the protection action alarm information and the circuit breaker trip alarm information, use a preset protection configuration knowledge graph to perform deductive reasoning on each suspected faulty device in the suspected faulty device set to obtain the confidence level of the observed event corresponding to each suspected faulty device;

[0184] Step 4: Based on the confidence level of each observed event, perform evidence fusion reasoning on each suspected faulty device to obtain the actual fault information of the smart grid.

[0185] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0186] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0187] The specific implementation process of the above method steps can be found in any of the above embodiments of the knowledge graph-based smart grid fault inference method, and will not be repeated here.

[0188] This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, it proposes an accident simulation framework that includes hypothetical reasoning, deductive reasoning, and fusion reasoning, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its explicit knowledge representation and independent reasoning mechanism, which not only improves the sharing, scalability, and interpretability of knowledge but also effectively handles time constraints and conflicting multi-source evidence. The method in this application provides more accurate reconstruction of smart grid faults.

[0189] Another embodiment of this application provides an electronic device, which can be a server. The electronic device includes a processor, a memory, a network interface, and a database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the program is executed by the processor, it implements the functions or steps of a knowledge graph-based smart grid fault prediction method on the server side.

[0190] In one embodiment, an electronic device is provided, which can be a client. The electronic device includes a processor, memory, a network interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with an external server via a network connection. When the program is executed by the processor, it implements client-side functions or steps of a knowledge graph-based smart grid fault prediction method.

[0191] Another embodiment of this application provides an electronic device, including at least a memory and a processor. The memory stores a computer program, and the processor, when executing the computer program in the memory, performs the following method steps:

[0192] Step 1: Receive protection action alarm information and circuit breaker tripping alarm information when a fault occurs in the target smart grid;

[0193] Step 2: Based on the protection action alarm information and the circuit breaker trip alarm information, perform hypothetical reasoning using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices;

[0194] Step 3: Based on the protection action alarm information and the circuit breaker trip alarm information, use a preset protection configuration knowledge graph to perform deductive reasoning on each suspected faulty device in the suspected faulty device set to obtain the confidence level of the observed event corresponding to each suspected faulty device;

[0195] Step 4: Based on the confidence level of each observed event, perform evidence fusion reasoning on each suspected faulty device to obtain the actual fault information of the smart grid.

[0196] The specific implementation process of the above method steps can be found in any of the above embodiments of the knowledge graph-based smart grid fault inference method, and will not be repeated here.

[0197] This application establishes a power grid topology knowledge graph and a protection configuration knowledge graph, constituting a power grid accident simulation knowledge graph. Based on this, it proposes an accident simulation framework that includes hypothetical reasoning, deductive reasoning, and fusion reasoning, achieving fault scope delineation, hypothesis verification of suspected faults, and fusion of multi-source evidence. The advantage of the knowledge graph-driven accident simulation framework proposed in this application lies in its explicit knowledge representation and independent reasoning mechanism, which not only improves the sharing, scalability, and interpretability of knowledge but also effectively handles time constraints and conflicting multi-source evidence. The method in this application provides more accurate reconstruction of smart grid faults.

[0198] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for fault prediction in smart grids based on knowledge graphs, characterized in that, include: Receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid; Based on the protection action alarm information and the circuit breaker trip alarm information, a set of suspected faulty devices is obtained by performing hypothetical reasoning using a preset power grid topology knowledge graph. Based on the protection action alarm information and the circuit breaker trip alarm information, a preset protection configuration knowledge graph is used to perform deductive reasoning on each suspected faulty device in the suspected faulty device set to obtain the confidence level of the observed event corresponding to each suspected faulty device; Based on the confidence level of each observed event, evidence fusion reasoning is performed on each suspected faulty device to obtain the actual fault information of the smart grid. The process of using a preset power grid topology knowledge graph to perform hypothetical reasoning based on the protection action alarm information and the circuit breaker trip alarm information to obtain a set of suspected faulty devices specifically includes: Data is extracted based on the protection action alarm information and the circuit breaker tripping alarm information to obtain a set of tripping circuit breakers; Construct the mapping relationship between each tripping circuit breaker in the tripping circuit breaker set to obtain each tripping circuit breaker group; For each of the tripped circuit breaker groups, a first query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the first suspected fault device corresponding to the shortest path of each of the tripped circuit breaker groups; For the primary line equipment in the first suspected faulty equipment, the second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain the second suspected faulty equipment corresponding to the shortest path of each of the primary line equipment; Perform a union calculation on the first suspected faulty device and the second suspected faulty device to obtain the set of suspected faulty devices; The primary equipment includes lines, busbars, and transformers; Before performing deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker trip alarm information using a preset protection configuration knowledge graph, the method further includes: constructing a preset protection configuration knowledge graph; The construction of the preset protection configuration knowledge graph specifically includes: The SCD file of the target smart grid is parsed to obtain the protection configuration information of the target smart grid. The protection configuration information includes the configuration information of primary equipment main protection and near backup protection, as well as the association between protection and tripping circuit breakers. The second query statement is used to perform a shortest path query on the preset power grid topology knowledge graph to obtain remote backup protection information; Obtain the action time constraints of the backup protection action from the preset protection setting sheet; Based on the protection configuration information, the remote backup protection information, and the action time constraints, a knowledge graph is constructed to obtain the preset protection configuration knowledge graph used to characterize the protection configuration and action time constraints of the primary equipment. The method of using a preset protection configuration knowledge graph to perform deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker trip alarm information to obtain the confidence level of the observed event corresponding to each suspected faulty device specifically includes: Based on each of the suspected faulty devices, a third query statement is used to query the preset protection configuration knowledge graph to obtain the expected events corresponding to each of the suspected faulty devices and the time constraints for the occurrence of the expected events; Data is filtered based on the protection action alarm information and the circuit breaker trip alarm information to obtain observed events; Confidence calculations are performed based on the expected events and the observed events to obtain the confidence scores of the observed events corresponding to each suspected faulty device.

2. The method as described in claim 1, characterized in that, Before performing hypothetical reasoning based on the protection action alarm information and the circuit breaker trip alarm information using a preset power grid topology knowledge graph, the method further includes: constructing a preset power grid topology knowledge graph; The construction of the preset power grid topology knowledge graph specifically includes: Obtain the topology network structure diagram of the target smart grid; Based on the topological network structure diagram, a model is constructed to obtain a CIM / E model corresponding to the target smart grid; Based on the CIM / E model, a knowledge graph is constructed to obtain a preset power grid topology knowledge graph that represents the topological connection relationship between devices and corresponds to the target smart grid.

3. The method as described in claim 1, characterized in that, The confidence calculation process based on the expected event set and the observed event set to obtain the confidence score of the observed event corresponding to each suspected faulty device specifically includes: The first index value is determined based on the actual time of the observed event and the time constraint of the observed event occurring in the preset protection configuration knowledge graph. The second index value is determined based on the type of protection action observed in the event. Based on the first and second index values, a pre-set confidence function is used to calculate and process the observed event confidence, so as to obtain the observed event confidence corresponding to each of the suspected faulty devices.

4. The method as described in claim 1, characterized in that, The process of performing evidence fusion reasoning on each of the suspected faulty devices based on the confidence levels of each of the observed events to obtain the actual fault information of the smart grid specifically includes: Evidence is added to the confidence scores of each observed event to obtain the target confidence matrix; Based on the target confidence matrix, the DS evidence fusion method is used to perform evidence fusion to obtain the evidence fusion result; Based on the evidence fusion results, the protection action alarm information when the target smart grid fails, and the circuit breaker tripping alarm information, the fault is reconstructed using the preset protection configuration knowledge graph to obtain the actual fault information of the target smart grid.

5. A knowledge graph-based smart grid fault estimation device, used to implement the knowledge graph-based smart grid fault estimation method as described in any one of claims 1-4, characterized in that, include: The receiving module is used to receive protection action alarm information and circuit breaker trip alarm information when a fault occurs in the target smart grid; The hypothesis reasoning module is used to perform hypothesis reasoning based on the protection action alarm information and the circuit breaker tripping alarm information using a preset power grid topology knowledge graph to obtain a set of suspected faulty devices. The deductive reasoning module is used to perform deductive reasoning on each suspected faulty device in the suspected faulty device set based on the protection action alarm information and the circuit breaker tripping alarm information using a preset protection configuration knowledge graph, and to obtain the confidence level of the observed event corresponding to each suspected faulty device. The evidence fusion reasoning module is used to perform evidence fusion reasoning on each of the suspected faulty devices based on the confidence level of each of the observed events, so as to obtain the actual fault information of the smart grid.

6. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the knowledge graph-based smart grid fault inference method according to any one of claims 1-4.

7. An electronic device, characterized in that, It includes at least a memory and a processor, wherein the memory stores a computer program, and the processor, when executing the computer program in the memory, implements the steps of the knowledge graph-based smart grid fault inference method according to any one of claims 1-4.

Images