Security verification method and device for power grid dispatching operation, terminal equipment and storage medium
By constructing knowledge graphs and deep learning models, user scheduling operation sequences are decomposed and verified, solving the problem of power grid scheduling operation errors caused by reliance on human experience, and achieving highly accurate safety verification.
Patent Information
- Application Number
- CN202511290176.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-10
- Publication Date
- 2025-12-16
AI Technical Summary
Existing safety verification methods for power grid dispatching and operation rely on the experience and knowledge of dispatchers, which are prone to human error and result in low verification accuracy.
By constructing a knowledge graph and a deep learning model, the user scheduling operation sequence is decomposed into word vectors. The accuracy of the operation sequence is verified based on similarity search and power grid operation model, avoiding reliance on human experience.
It improves the accuracy of safety verification for power grid dispatching and operation, reduces human error, and ensures the standardization and reliability of operation sequences.
Smart Images
Figure CN121144533A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power systems, and in particular to a method, apparatus, terminal equipment, and storage medium for verifying the safety of power grid dispatching and operation. Background Technology
[0002] With the rapid development and increasing intelligence of power systems, the safety verification of power grid dispatching and operation has become particularly important. Safety verification mainly consists of two aspects: system status verification and operating procedure verification. System status verification assesses the power system status through simulation models and software, while operating procedure verification determines whether operations conform to established rules. These rules are typically described in natural language, and understanding and verifying their accuracy is crucial for power grid safety.
[0003] Existing safety verification methods for power grid dispatching operations are usually based on the experience and knowledge of dispatchers, which are prone to human error and result in low accuracy of safety verification for power grid dispatching operations. Summary of the Invention
[0004] This invention provides a method, apparatus, terminal equipment, and storage medium for safety verification of power grid dispatching operations. It can solve the technical problem that verification based on the experience and knowledge of dispatchers in the prior art is prone to human error, resulting in low accuracy of safety verification of power grid dispatching operations.
[0005] This invention provides a security verification method for power grid dispatching operation, comprising:
[0006] Obtain the scheduling execution text and construct a knowledge graph based on the scheduling execution text;
[0007] The user scheduling operation sequence to be verified is decomposed into a set of scheduling operations, and a pre-trained deep learning model is used to construct word vectors based on the set of scheduling operations.
[0008] A similarity search is performed between the word vectors and the entities and relationships in the knowledge graph, and a query set is constructed based on the entities and relationships whose similarity meets the preset conditions.
[0009] Based on the entities and relationships in the query set, determine the standard operation sequence that meets the operation procedure;
[0010] The user scheduling operation sequence to be verified is compared with the standard operation sequence. If the similarity exceeds a preset threshold, the user scheduling operation sequence to be verified is deemed to have passed the verification.
[0011] Furthermore, the step of constructing a knowledge graph based on the scheduled execution text includes:
[0012] Extract entities and relationships from the scheduling and running text, and construct a knowledge graph based on the entities and relationships in the scheduling and running text.
[0013] Furthermore, the step of decomposing the user scheduling operation sequence to be verified into a set of scheduling operations, and constructing word vectors based on the set of scheduling operations using a pre-trained deep learning model, includes:
[0014] The sequence of user scheduling operations to be verified is decomposed into a set of scheduling operations using a large language model;
[0015] A pre-trained deep learning model is used to extract word segments from the scheduling operation set, and a word vector set is constructed based on the word segments.
[0016] Furthermore, determining the standard operation sequence that satisfies the operation procedure based on the entities and relations in the query set includes:
[0017] Based on the entities and relationships in the query set, determine the operation sequence in the knowledge graph;
[0018] The operation sequences are integrated to form a standard operation sequence that meets the operation procedures.
[0019] Furthermore, before comparing the similarity of the user scheduling operation sequence to be verified with the standard operation sequence, and determining that the user scheduling operation sequence to be verified has passed verification when the similarity exceeds a preset threshold, the method further includes:
[0020] Input parameters are extracted from the standard operation sequence and substituted into a pre-built power grid operation model. If the constraints in the power grid operation model are met, the physical feasibility of the standard operation sequence is determined.
[0021] Furthermore, the construction of the power grid operation model includes:
[0022] Construct an objective function based on the total number of units and the deviation adjustment cost of the units, with the goal of minimizing the total system adjustment in multiple stages;
[0023] The power grid operation model is constructed based on the objective function and constraints, wherein the constraints include system operation constraints, new energy unit operation constraints, and power system stability and power flow constraints.
[0024] The present invention also provides a safety verification device for power grid dispatching operation, comprising:
[0025] The knowledge graph construction module is used to obtain the scheduling and execution text and construct a knowledge graph based on the scheduling and execution text.
[0026] The word vector construction module is used to decompose the user scheduling operation sequence to be verified into a set of scheduling operations, and to construct word vectors based on the set of scheduling operations using a pre-trained deep learning model.
[0027] The query set construction module is used to perform similarity search between the word vectors and entities and relations in the knowledge graph, and construct the query set based on entities and relations whose similarity meets preset conditions.
[0028] The standard operation sequence determination module is used to determine the standard operation sequence that satisfies the operation procedure based on the entities and relationships in the query set.
[0029] The operation sequence verification module is used to compare the similarity of the user scheduling operation sequence to be verified with the standard operation sequence, and to determine that the user scheduling operation sequence to be verified has passed the verification when the similarity exceeds a preset threshold.
[0030] Furthermore, the step of constructing a knowledge graph based on the scheduled execution text includes:
[0031] Extract entities and relationships from the scheduling and running text, and construct a knowledge graph based on the entities and relationships in the scheduling and running text.
[0032] The present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the security verification method for power grid dispatch operation as described above.
[0033] The present invention also provides a computer-readable storage medium, comprising: a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the security verification method for power grid dispatch operation as described above.
[0034] The following benefits can be obtained by implementing the present invention:
[0035] This invention constructs word vectors using a deep learning model, determines the query set based on word vector similarity, and further determines a standard operation sequence based on the entities and relationships in the query set. This sequence is then used to verify the user scheduling operation sequence to be verified. The verification of the operation sequence is based on a large amount of data and structured knowledge, eliminating the need to rely on the dispatcher's experience and knowledge, effectively avoiding human error, and thus ensuring the accuracy of the safety verification of power grid dispatching operations.
[0036] Furthermore, this invention constructs a knowledge graph based on the entities and relationships in the scheduling operation text, which can store complex information in power grid scheduling operation in a structured form and form a knowledge graph. Security verification is then performed based on the knowledge graph, without relying on human experience and knowledge, effectively improving the accuracy of verification. Attached Figure Description
[0037] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0038] Figure 1 This is a flowchart illustrating a security verification method for power grid dispatching operation provided in an embodiment of the present invention;
[0039] Figure 2 This is a schematic diagram of the structure of a safety verification device for power grid dispatching and operation provided in an embodiment of the present invention. Detailed Implementation
[0040] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.
[0042] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.
[0043] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0044] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.
[0045] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).
[0046] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.
[0047] See Figure 1 To address the technical problem that existing technologies relying on dispatchers' experience and knowledge for verification are prone to human error, leading to low accuracy in power grid dispatching operation safety verification, an embodiment of the present invention provides a power grid dispatching operation safety verification method, comprising:
[0048] S1. Obtain the scheduling execution text and construct a knowledge graph based on the scheduling execution text;
[0049] In this embodiment of the invention, the scheduling operation text includes existing power dispatching knowledge such as scheduling operation rules, scheduling logs, and scheduling plans. The scheduling operation text can be provided by the user or extracted from the scheduling operation text pre-stored in the database.
[0050] S2. Decompose the user scheduling operation sequence to be verified into a set of scheduling operations, and use a pre-trained deep learning model to construct word vectors based on the set of scheduling operations.
[0051] In this embodiment of the invention, the user scheduling operation sequence can be the operation data input by the user, such as "After the entire 220kV xx substation loses power, disconnect the xx switch of the substation that lost power, quickly check the protection information, force the tripped xx line to be restored, and restore the operation of the substation that lost power". The user scheduling operation sequence to be verified is decomposed into a scheduling operation set A, where the scheduling operation set A can be A = (220kV xx substation, entire substation loses power, disconnect, substation that lost power, xx switch, check, protection information, tripped xx line, force the restoration, restore, xx substation, operation).
[0052] The embodiments of the present invention use a deep learning model to transform scheduling operation sequences into word vectors, which can accurately capture semantic information in the operation sequences and help improve the accuracy of security verification.
[0053] S3. Perform similarity search based on word vectors and entities and relationships in knowledge graphs, and construct a query set based on entities and relationships whose similarity meets preset conditions;
[0054] In this embodiment of the invention, similarity search based on word vectors and entities and relationships in the knowledge graph can determine the entities and relationships in the knowledge graph related to the input user scheduling operation, and these entities and relationships constitute the query set.
[0055] In this embodiment of the invention, similarity search can quickly identify entities and relationships related to word vectors in the knowledge graph, and can effectively filter out wiring tubes or entities and relationships with low relevance, thereby effectively improving the efficiency of subsequent security verification.
[0056] S4. Based on the entities and relations in the query set, determine the standard operation sequence that meets the operation procedure;
[0057] In this embodiment of the invention, nodes in the knowledge graph are connected to other nodes through edges. Based on the nodes and relationships in the query set, a single-hop or multi-hop search can be performed in the knowledge graph. Finally, the entities and relationships in the query set form a standard operation sequence, such as "station undervoltage - (post-event operation) -> disconnect station switch - (subsequent operation) -> check protection information - (subsequent operation) -> restore tripped line forced transmission - (subsequent operation) - (subsequent operation) -> check station bus voltage - (subsequent operation) -> ... - (subsequent operation) -> restore station operation".
[0058] S5. Compare the similarity between the user scheduling operation sequence to be verified and the standard operation sequence. If the similarity exceeds the preset threshold, the user scheduling operation sequence to be verified is deemed to have passed the verification.
[0059] In this embodiment of the invention, a preset threshold can be set based on historical dispatch operation sequences. For example, in a dispatch system, the dispatch situation based on historical dispatch operation sequences over the past year can be statistically analyzed. By determining the similarity distribution between historical dispatch operation sequences and their standard operation sequences under successful dispatch conditions, a preset threshold for phase velocity can be determined. For instance, in most successful dispatch conditions, the similarity distribution between historical dispatch operation sequences and their standard operation sequences is between 80% and 95%, so the preset threshold for similarity can be set to 80%, 85%, 90%, etc. When the similarity exceeds the preset threshold, the user dispatch operation sequence to be verified is deemed to have passed verification; when the similarity is below the preset threshold, a detailed risk assessment report and improvement suggestions can be output through retrieval enhancement generation technology, providing dispatchers with comprehensive safety guidance, thereby ensuring the standardization and reliability of power grid dispatch operations.
[0060] This invention constructs word vectors using a deep learning model, determines the query set based on word vector similarity, and further determines a standard operation sequence based on the entities and relationships in the query set. This sequence is then used to verify the user scheduling operation sequence to be verified. The verification of the operation sequence is based on a large amount of data and structured knowledge, eliminating the need to rely on the dispatcher's experience and knowledge, effectively avoiding human error, and thus ensuring the accuracy of the safety verification of power grid dispatching operations.
[0061] In one embodiment, step S1, constructing a knowledge graph based on the scheduled execution text, includes:
[0062] Extract entities and relationships from the scheduling and execution text, and construct a knowledge graph based on the entities and relationships in the scheduling and execution text.
[0063] In this embodiment of the invention, a large language model can be used for few-shot learning, and then entities and relationships can be extracted from the scheduled running text. Entities can be devices, operators, and operation steps, and relationships can be operation order and device association, etc.
[0064] The process of extracting entities and relations through a large language model in this embodiment of the invention can be as follows:
[0065] Given a text block T, a large language model identifies and extracts the triplet t from the text block. i =(e i ,r i ,e i '), where e i and e i 'Is an entity, r i This refers to the relationship between the head entity and the tail entity. This process can be modeled as follows:
[0066]
[0067] In this embodiment of the invention, when new operating conditions and procedural requirements are input, the large language model can automatically extract new text content and update the knowledge graph.
[0068] In this embodiment of the invention, supernodes can be introduced to represent complex scheduling operation relationships. A supernode can contain multiple sub-entities and relationships, thus forming a multi-level nested structure. For example, a supernode can represent a complex operation flow, which includes multiple operation steps and device associations. Supernode h * The definition is as follows:
[0069] h * = (h1, r1, h2), where h * h1, h2 ∈ E, r1 ∈ R
[0070] Here, h1 and h2 represent the head entity and tail entity, respectively. They can be simple entities or other triplet supernodes, and h1 and h2 can be any of them. * Recursively containing further nested triplet supernodes h i :
[0071]
[0072] In this embodiment of the invention, the extracted entities, relationships, and supernodes are stored in a graph database to form a knowledge graph. The storage structure of the knowledge graph is as follows:
[0073] G={t=(h,r,h′)|h,h′∈E,r∈R}
[0074] Where G is the knowledge graph, E is the entity set, R is the relation set, t is the triple in the knowledge graph, h and h' are the head entity and tail entity respectively, which can be ordinary entities or supernodes containing multiple sub-entities and relations, and r is the relation between the head entity and the tail entity.
[0075] This invention constructs a knowledge graph based on entities and relationships in the scheduling operation text. This allows complex information in power grid scheduling operations to be stored in a structured form and formed into a knowledge graph. Security verification is then performed based on the knowledge graph, eliminating the need to rely on human experience and knowledge, and effectively improving the accuracy of verification.
[0076] In one embodiment, step S2, decomposing the user scheduling operation sequence to be verified into a set of scheduling operations, and using a pre-trained deep learning model to construct word vectors based on the set of scheduling operations, includes:
[0077] S21. The sequence of user scheduling operations to be verified is decomposed into a set of scheduling operations using a large language model.
[0078] In this embodiment of the invention, the scheduling operation set is used for retrieval in the knowledge graph.
[0079] In this embodiment of the invention, the process of decomposing into scheduling operations can be represented as follows:
[0080] A = LLM(q)
[0081] Where A is the set of scheduled operations, which is automatically generated by the Large Language Model (LLM) based on the user operation sequence q.
[0082] S22. Use a pre-trained deep learning model to extract word segments from the scheduling operation set, and construct a word vector set based on the word segments.
[0083] In this embodiment of the invention, a pre-trained deep learning model can be used to perform word segmentation on the scheduling operation set A, extract the word segments from the scheduling operation set, and construct a word vector set based on these word segments. The deep learning model can be a Word Transformer model.
[0084] The embodiments of the present invention can decompose complex scheduling operation sequences into finer-grained operation sets through a large language model, and extract word segments and construct word vectors through a deep learning model, thereby capturing semantic information in the operation sequence, accurately understanding the meaning of the operation, and effectively improving the accuracy and reliability of the safety verification of power grid scheduling operation.
[0085] In one embodiment, step S3, performing a similarity search based on word vectors and entities and relationships in a knowledge graph, and constructing a query set based on entities and relationships whose similarity meets preset conditions, can be represented as:
[0086] N=VectorSearch(WordTransformer(A))
[0087] Where N is the set to be queried, VectorSearch is the function used to perform vector search, and WordTransformer is the function used to convert input data (such as text) into vector form.
[0088] In this embodiment of the invention, a similarity search is performed based on word vectors and entities and relationships in the knowledge graph to determine entities and relationships in the knowledge graph that meet preset conditions, such as a similarity of more than 80%, and a query set is constructed based on these entities and relationships.
[0089] In one embodiment, step S4, determining the standard operation sequence that satisfies the operation procedure based on the entities and relationships in the query set, includes:
[0090] S41. Determine the operation sequence in the knowledge graph based on the entities and relations in the query set;
[0091] In this embodiment of the invention, the Cypher query language can be used to query relevant operation sequences in the knowledge graph based on the entities and relations in the query set N. These relevant operation sequences are either directly or indirectly related. This process can be represented as follows:
[0092] Path = CypherQuery(N)
[0093] Where N is the set to be queried, and Path is the sequence of operations obtained after querying the knowledge graph using the Cypher query language.
[0094] S42. Integrate the operation sequences to form a standard operation sequence that meets the operating procedures.
[0095] In this embodiment of the invention, the order of the operation sequence can be analyzed based on the operation procedure, and the order of the operation sequence can be sorted, duplicate operation steps can be removed, and necessary operation steps can be added according to the analysis results, so as to integrate and obtain a standard operation sequence that meets the operation procedure.
[0096] The embodiments of the present invention can accurately determine the relevant operation sequences by using entities and relationships in the knowledge graph, ensuring the standardization and accuracy of the final determined standard operation sequence, thereby effectively improving the accuracy and reliability of security verification.
[0097] In one embodiment, before step S5, which compares the similarity of the user scheduling operation sequence to be verified with the standard operation sequence, and determines that the user scheduling operation sequence to be verified has passed verification when the similarity exceeds a preset threshold, the method further includes:
[0098] Input parameters are extracted from the standard operating sequence and substituted into a pre-built power grid operation model. If the constraints in the power grid operation model are met, the physical feasibility of the standard operating sequence is determined.
[0099] In this embodiment of the invention, a computer tool interface can be encapsulated, a computing tool can be determined based on the tool description information, and input parameters can be passed to the computing tool for calculation to determine whether the standard operation sequence is physically feasible. The computing tool is a model used to perform relevant calculations based on the input parameters.
[0100] In this embodiment of the invention, the functional description, input parameters, and output results of the computing tool can be described in words, enabling the large language model to accurately determine the functional positioning of the computing tool, thereby facilitating accurate identification of the computing tool. This embodiment of the invention standardizes the encapsulation of the computing tool, clarifying the encapsulation criteria and framework structure. The computing tool includes multiple aspects such as power flow calculation, load forecasting, and autonomous generation. Autonomous generation needs to consider the temporal coupling of different scheduling moments, minimizing the total system adjustment amount across multiple stages by optimizing the generation plan and making real-time adjustments while meeting the constraints of safe system operation.
[0101] The embodiments of the present invention, by substituting the extracted input parameters into the power grid operation model, can accurately calculate and verify the physical feasibility of the standard operation sequence, ensuring that the operation sequence is safe and feasible in actual power grid operation. Furthermore, through model verification, interference from human factors can be effectively avoided, thereby effectively improving the accuracy of safety verification of power grid dispatching and operation.
[0102] In one embodiment, the construction of the power grid operation model includes:
[0103] Construct an objective function based on the total number of units and the deviation adjustment cost of the units, with the goal of minimizing the total system adjustment in multiple stages;
[0104] In this embodiment of the invention, the expression for the objective function F is as follows:
[0105]
[0106] Where I is the total number of generating units, T is the number of time periods, and C is the number of time periods. i,t Let t be the deviation adjustment cost for unit i during time period t.
[0107] A power grid operation model is constructed based on the objective function and constraints, including system operation constraints, new energy unit operation constraints, and power system stability and power flow constraints.
[0108] In this embodiment of the invention, the expression for the system operation constraint is as follows:
[0109]
[0110] Among them, P i TG (t) and This represents the real-time active power output and real-time reactive power output of a traditional thermal power unit i at time t. and These represent the downward and upward ramping constraints of the power of traditional thermal power unit i, respectively; and These represent the upper and lower limits of the active power output of a traditional thermal power unit i, respectively. and These represent the upper and lower limits of reactive power output of traditional thermal power unit i, respectively.
[0111] The expressions for the operating constraints of new energy generating units are as follows:
[0112]
[0113] In the formula, P i RE (t) represents the real-time active power output of the new energy unit i at time t. This indicates the upper limit of the active power output of the new energy unit i.
[0114] The expressions for power system stability and power flow constraints are as follows:
[0115]
[0116] In the formula, U i (t), θ i (t) represents the voltage value and phase of node i at scheduling time t, respectively; θ ij (t) represents the phase difference between nodes i and j; U i,max U i,min Indicates the upper and lower limits of the allowable voltage; θ i,max θ i,min Indicates the upper and lower limits of phase tolerance; G ij and B ij Let i and j represent the conductance and susceptance between nodes i and j, respectively.
[0117] Implementing the embodiments of the present invention has the following beneficial effects:
[0118] This invention constructs word vectors using a deep learning model, determines the query set based on word vector similarity, and further determines a standard operation sequence based on the entities and relationships in the query set. This sequence is then used to verify the user scheduling operation sequence to be verified. The verification of the operation sequence is based on a large amount of data and structured knowledge, eliminating the need to rely on the dispatcher's experience and knowledge, effectively avoiding human error, and thus ensuring the accuracy of the safety verification of power grid dispatching operations.
[0119] Furthermore, in this embodiment of the invention, a knowledge graph is constructed based on the entities and relationships in the scheduling operation text. This allows complex information in power grid scheduling operation to be stored in a structured form and formed into a knowledge graph. Security verification is then performed based on the knowledge graph, eliminating the need to rely on human experience and knowledge, and effectively improving the accuracy of verification.
[0120] like Figure 2 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided;
[0121] An embodiment of the present invention provides a safety verification device for power grid dispatching operation, comprising:
[0122] The knowledge graph construction module 10 is used to obtain the scheduling execution text and construct a knowledge graph based on the scheduling execution text.
[0123] The word vector construction module 20 is used to decompose the user scheduling operation sequence to be verified into a set of scheduling operations, and to construct word vectors based on the set of scheduling operations using a pre-trained deep learning model.
[0124] The query set construction module 30 is used to perform similarity search based on word vectors and entities and relationships in the knowledge graph, and constructs the query set based on entities and relationships whose similarity meets preset conditions.
[0125] The standard operation sequence determination module 40 is used to determine the standard operation sequence that meets the operation procedure based on the entities and relations in the query set.
[0126] The operation sequence verification module 50 is used to compare the similarity of the user scheduling operation sequence to be verified with the standard operation sequence. When the similarity exceeds a preset threshold, the user scheduling operation sequence to be verified is determined to have passed the verification.
[0127] In one embodiment, constructing a knowledge graph based on scheduled execution text includes:
[0128] Extract entities and relationships from the scheduling and execution text, and construct a knowledge graph based on the entities and relationships in the scheduling and execution text.
[0129] In one embodiment, the sequence of user scheduling operations to be verified is decomposed into a set of scheduling operations. A pre-trained deep learning model is used to construct word vectors based on the set of scheduling operations, including:
[0130] The user scheduling operation sequence to be verified is decomposed into a set of scheduling operations using a large language model;
[0131] A pre-trained deep learning model is used to extract word segments from the scheduling operation set, and a set of word vectors is constructed based on the word segments.
[0132] In one embodiment, a standard sequence of operations that satisfies the operational procedure is determined based on the entities and relationships in the query set, including:
[0133] Determine the sequence of operations in the knowledge graph based on the entities and relationships in the query set;
[0134] The operation sequences are integrated to form a standard operation sequence that meets the operating procedures.
[0135] In one embodiment, before comparing the user scheduling operation sequence to be verified with a standard operation sequence, and determining that the user scheduling operation sequence to be verified has passed verification when the similarity exceeds a preset threshold, the method further includes:
[0136] Input parameters are extracted from the standard operating sequence and substituted into a pre-built power grid operation model. If the constraints in the power grid operation model are met, the physical feasibility of the standard operating sequence is determined.
[0137] In one embodiment, the construction of the power grid operation model includes:
[0138] Construct an objective function based on the total number of units and the deviation adjustment cost of the units, with the goal of minimizing the total system adjustment in multiple stages;
[0139] A power grid operation model is constructed based on the objective function and constraints, including system operation constraints, new energy unit operation constraints, and power system stability and power flow constraints.
[0140] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the safety verification method for power grid dispatching and operation provided by any of the above-described method embodiments of the present invention.
[0141] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.
[0142] Based on the above embodiments of the power grid dispatching operation security verification method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the power grid dispatching operation security verification method of any embodiment of the present invention.
[0143] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.
[0144] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.
[0145] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting various parts of the terminal device via various interfaces and lines.
[0146] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the power grid dispatch operation security verification method described in any of the above-described method embodiments of the present invention.
[0147] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.
[0148] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A method for verifying the safety of power grid dispatching and operation, characterized in that, include: Obtain the scheduling execution text and construct a knowledge graph based on the scheduling execution text; The user scheduling operation sequence to be verified is decomposed into a set of scheduling operations, and a pre-trained deep learning model is used to construct word vectors based on the set of scheduling operations. A similarity search is performed between the word vectors and the entities and relationships in the knowledge graph, and a query set is constructed based on the entities and relationships whose similarity meets the preset conditions. Based on the entities and relationships in the query set, determine the standard operation sequence that meets the operation procedure; The user scheduling operation sequence to be verified is compared with the standard operation sequence. If the similarity exceeds a preset threshold, the user scheduling operation sequence to be verified is deemed to have passed the verification.
2. The safety verification method for power grid dispatching operation as described in claim 1, characterized in that, The step of constructing a knowledge graph based on the scheduled execution text includes: Extract entities and relationships from the scheduling and running text, and construct a knowledge graph based on the entities and relationships in the scheduling and running text.
3. The safety verification method for power grid dispatching operation as described in claim 1, characterized in that, The step of decomposing the user scheduling operation sequence to be verified into a set of scheduling operations, and constructing word vectors based on the set of scheduling operations using a pre-trained deep learning model, includes: The sequence of user scheduling operations to be verified is decomposed into a set of scheduling operations using a large language model; A pre-trained deep learning model is used to extract word segments from the scheduling operation set, and a word vector set is constructed based on the word segments.
4. The safety verification method for power grid dispatching operation as described in claim 1, characterized in that, The step of determining the standard operation sequence that satisfies the operation procedure based on the entities and relations in the query set includes: Based on the entities and relationships in the query set, determine the operation sequence in the knowledge graph; The operation sequences are integrated to form a standard operation sequence that meets the operation procedures.
5. The safety verification method for power grid dispatching operation as described in claim 1, characterized in that, Before comparing the user scheduling operation sequence to be verified with the standard operation sequence, and determining that the user scheduling operation sequence to be verified has passed verification when the similarity exceeds a preset threshold, the method further includes: Input parameters are extracted from the standard operating sequence and substituted into a pre-built power grid operation model. If the constraints in the power grid operation model are met, the physical feasibility of the standard operating sequence is determined.
6. The safety verification method for power grid dispatching operation as described in claim 5, characterized in that, The construction of the power grid operation model includes: Construct an objective function based on the total number of units and the deviation adjustment cost of the units, with the goal of minimizing the total system adjustment in multiple stages; The power grid operation model is constructed based on the objective function and constraints, wherein the constraints include system operation constraints, new energy unit operation constraints, and power system stability and power flow constraints.
7. A safety verification device for power grid dispatching and operation, characterized in that, include: The knowledge graph construction module is used to obtain the scheduling and execution text and construct a knowledge graph based on the scheduling and execution text. The word vector construction module is used to decompose the user scheduling operation sequence to be verified into a set of scheduling operations, and to construct word vectors based on the set of scheduling operations using a pre-trained deep learning model. The query set construction module is used to perform similarity search between the word vectors and entities and relations in the knowledge graph, and construct the query set based on entities and relations whose similarity meets preset conditions. The standard operation sequence determination module is used to determine the standard operation sequence that satisfies the operation procedure based on the entities and relationships in the query set. The operation sequence verification module is used to compare the similarity of the user scheduling operation sequence to be verified with the standard operation sequence, and to determine that the user scheduling operation sequence to be verified has passed the verification when the similarity exceeds a preset threshold.
8. The safety verification device for power grid dispatching operation as described in claim 7, characterized in that, The step of constructing a knowledge graph based on the scheduled execution text includes: Extract entities and relationships from the scheduling and running text, and construct a knowledge graph based on the entities and relationships in the scheduling and running text.
9. A terminal device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the security verification method for power grid dispatch operation as described in any one of claims 1-6.
10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the security verification method for power grid dispatch operation as described in any one of claims 1-6.