Intelligent verification method and system for electric power work ticket content

Abstract

The invention provides an electric power work ticket content intelligent verification method and system, and the method comprises the steps: obtaining the target content information of an electric power work ticket, and determining the target graph structure data corresponding to the target content information according to the target content information of the electric power work ticket, finally, inputting the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; the dependency relationship verification model is obtained by training graph structure data corresponding to the historical power work ticket and a verification result corresponding to the graph structure data, and is used for obtaining a verification model of the dependency relationship among the entities in the target content information. According to the technical scheme, the dependency relationship between the entities in the power work ticket can be determined by using the dependency relationship verification model, and whether the power work ticket has logic errors, dependency relationship missing or dependency relationship conflicts or not is judged, so that a user can be efficiently assisted to determine the accuracy of the power work ticket based on the dependency relationship missing or the dependency relationship conflicts.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to an intelligent verification method and system for the content of an electric power work ticket. Background Art

[0002] A work ticket is a written instruction allowing work to be carried out on electrical equipment or power lines. It is also a written basis for clarifying safety responsibilities, briefing workers on safety, and ensuring the safety of workers. The management of work tickets is an important support for the development of power work, and the verification of work tickets is of great significance to the safety of people, property, etc.

[0003] In the existing technology, experienced power workers or safety managers are relied upon to carefully review every field and detail of the work ticket to ensure that it complies with relevant safety regulations, operating procedures and legal and regulatory requirements.

[0004] However, this method of the prior art is time-consuming, labor-intensive, inefficient, and has the technical problem of low accuracy. Summary of the invention

[0005] The present application provides a method and system for intelligently verifying the contents of an electric power work ticket, so as to solve the problems of inaccurate verification, low efficiency, time-consuming and labor-intensive verification of electric power work tickets in the prior art.

[0006] In a first aspect, an embodiment of the present application provides an intelligent verification method for the content of an electric power work ticket, comprising: Obtain target content information of the power work ticket; According to the target content information of the electric power work ticket, determining the target graph structure data corresponding to the target content information; The target graph structure data is input into a pre-trained dependency verification model to obtain a first verification result; the first verification result is used to indicate whether the power work ticket has a logical error, missing dependency or a dependency conflict; the dependency verification model is a verification model trained based on the graph structure data corresponding to historical power work tickets and the verification results corresponding to the graph structure data, and is used to obtain the dependency relationships between the entities in the target content information.

[0007] In one or more embodiments, the target graph structure data is the graph structure of the power work ticket; Correspondingly, the target graph structure data corresponding to the target content information is determined according to the target content information of the electric power work ticket, including: Performing structural processing on the target content information to obtain entities involved in the target content information and relationships between the entities; The entities are taken as nodes of the target graph structure data, and the relationships between the entities are taken as edges of the target graph structure data.

[0008] In one or more embodiments, after obtaining the target content information of the electric power work ticket, the method further includes: Inputting the target content information of the electric power work ticket into a pre-trained preliminary verification model to obtain a second verification result; the second verification result is used to identify key information in the electric power work ticket to determine whether the format or content of the electric power work ticket is wrong; Extracting features of the target content information of the electric power work ticket to obtain vector features of each part of the target content information; Determine the semantic similarity between each vector feature according to a preset similarity algorithm, wherein the semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information; The electric power work ticket is comprehensively evaluated according to the second verification result and the semantic similarity.

[0009] In one or more embodiments, after extracting features from the target content information of the electric work ticket to obtain vector features of each part of the target content information, the method further includes: The vector features are input into a pre-trained semantic understanding model to obtain a third verification result, which is used to express the rationality and consistency of the logical relationship and context dependency between the fields in the target content information. The semantic understanding model is constructed by an attention mechanism and trained based on the vector features corresponding to historical power work tickets and the verification results corresponding to historical power work tickets.

[0010] In one or more embodiments, the method further comprises: Determine the optimal solution set among multiple optimization objectives; During multiple runs of the target model, adjusting the parameters of the target model in real time to obtain different output solution sets, wherein the target model is any one of a dependency verification model, a semantic understanding model, and a preliminary verification model; Different output solution sets are evaluated according to the optimal solution set among the multiple optimization objectives to obtain an optimal target model.

[0011] In one or more embodiments, the method further comprises: Determine time series analysis information corresponding to a first electric power work ticket within a first preset time period before a current time node; The time series analysis information is input into a prediction model to obtain a prediction result, wherein the prediction result expresses the number and type of second power work tickets within a second preset time period after the current time node; the prediction model is trained based on historical power work tickets.

[0012] In one or more embodiments, the method further comprises: Inputting the target content information into an anomaly detection model to obtain abnormal data corresponding to the power work ticket, wherein the anomaly detection model is trained with the contents of multiple work tickets and is used to determine abnormal points in the target content information; An abnormal result of the target content information is determined according to the abnormal data.

[0013] In one or more embodiments, the method further comprises: According to the abnormal result, the target model is optimized to obtain an optimized target model.

[0014] In one or more embodiments, the method further comprises: The target verification results are encrypted by homomorphic encryption or differential privacy, and the target verification results include: a first verification result, a second verification result, and a third verification result.

[0015] In a second aspect, an embodiment of the present application provides an intelligent verification device for the content of an electric power work ticket, comprising: An acquisition module is used to obtain target content information of the power work ticket; A determination module, used to determine the target graph structure data corresponding to the target content information according to the target content information of the electric power work ticket; A processing module is used to input the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; the first verification result is used to indicate whether the power work ticket has a logical error, a missing dependency, or a dependency conflict; the dependency verification model is a verification model trained based on the graph structure data corresponding to historical power work tickets and the verification results corresponding to the graph structure data, and is used to obtain the dependency between the entities in the target content information.

[0016] In one or more embodiments, the target graph structure data is the graph structure of the power work ticket; Accordingly, the determining module is specifically used for: Performing structural processing on the target content information to obtain entities involved in the target content information and relationships between the entities; The entities are taken as nodes of the target graph structure data, and the relationships between the entities are taken as edges of the target graph structure data.

[0017] In one or more embodiments, after obtaining the target content information of the electric power work ticket, the processing module is further used to: Inputting the target content information of the electric power work ticket into a pre-trained preliminary verification model to obtain a second verification result; the second verification result is used to identify key information in the electric power work ticket to determine whether the format or content of the electric power work ticket is wrong; Extracting features of the target content information of the electric power work ticket to obtain vector features of each part of the target content information; Determine the semantic similarity between each vector feature according to a preset similarity algorithm, wherein the semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information; The electric power work ticket is comprehensively evaluated according to the second verification result and the semantic similarity.

[0018] In one or more embodiments, after extracting features from the target content information of the electric work ticket to obtain vector features of each part of the target content information, the processing module is further used to: The vector features are input into a pre-trained semantic understanding model to obtain a third verification result, which is used to express the rationality and consistency of the logical relationship and context dependency between the fields in the target content information. The semantic understanding model is constructed by an attention mechanism and trained based on the vector features corresponding to historical power work tickets and the verification results corresponding to historical power work tickets.

[0019] In one or more embodiments, the processing module is further configured to: Determine the optimal solution set among multiple optimization objectives; During multiple runs of the target model, adjusting the parameters of the target model in real time to obtain different output solution sets, wherein the target model is any one of a dependency verification model, a semantic understanding model, and a preliminary verification model; Different output solution sets are evaluated according to the optimal solution set among the multiple optimization objectives to obtain an optimal target model.

[0020] In one or more embodiments, the processing module is further configured to: Determine time series analysis information corresponding to a first electric power work ticket within a first preset time period before a current time node; The time series analysis information is input into a prediction model to obtain a prediction result, wherein the prediction result expresses the number and type of second power work tickets within a second preset time period after the current time node; the prediction model is trained based on historical power work tickets.

[0021] In one or more embodiments, the processing module is further configured to: Inputting the target content information into an anomaly detection model to obtain abnormal data corresponding to the power work ticket, wherein the anomaly detection model is trained with the contents of multiple work tickets and is used to determine abnormal points in the target content information; An abnormal result of the target content information is determined according to the abnormal data.

[0022] In one or more embodiments, the processing module is further configured to: According to the abnormal result, the target model is optimized to obtain an optimized target model.

[0023] In one or more embodiments, the processing module is further configured to: The target verification results are encrypted by homomorphic encryption or differential privacy, and the target verification results include: a first verification result, a second verification result, and a third verification result.

[0024] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method as described in the first aspect or any one of the above methods.

[0025] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect or any one of the above methods. In a fifth aspect, the present application provides a computer program, wherein the computer program product comprises a computer program, wherein the computer program is stored in a computer-readable storage medium, and at least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, the method described in the first aspect or any one of the methods can be implemented.

[0026] In a sixth aspect, the present application provides an intelligent verification system for the contents of an electric power work ticket, comprising: an electronic device and a first device; The electronic device obtains the electric power work ticket from the first device, and executes the electric power work ticket content intelligent verification method described in any one of the first aspect and various embodiments above.

[0027] The present application provides an intelligent verification method and system for the content of an electric work ticket. The method obtains the target content information of the electric work ticket, and then determines the target graph structure data corresponding to the target content information based on the target content information of the electric work ticket, and finally inputs the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; the first verification result is used to indicate whether there is a logical error, missing dependency, or dependency conflict in the electric work ticket; the dependency verification model is a verification model trained based on the graph structure data corresponding to the historical electric work ticket and the verification results corresponding to the graph structure data, and is used to obtain the dependency between each entity in the target content information. In this technical solution, the dependency verification model can be used to determine the dependency between each entity in the electric work ticket, and to judge whether there is a logical error, missing dependency, or dependency conflict in the electric work ticket, so as to efficiently assist the user to determine the accuracy of the electric work ticket based on this. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0029] Figure 1 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 1 ; Figure 2 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 2 ; Figure 3 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 3 ; Figure 4 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 4 ; Figure 5 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 5 ; Figure 6 A schematic diagram of the structure of an intelligent verification system for electric power work ticket contents provided in an embodiment of the present application; Figure 7 A schematic diagram of the structure of an intelligent verification device for the contents of an electric power work ticket provided in an embodiment of the present application; Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0030] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] Before introducing the embodiments of the present application, the application background of the embodiments of the present application is first explained: In the electrical field, a work ticket is a written instruction that allows work to be carried out on electrical equipment and power lines. It is also a written basis for clarifying safety responsibilities, briefing workers on safety, and ensuring the safety of workers. An operation ticket is a written order for electrical operation personnel on duty to operate. It is an important basis for preventing misoperation and ensuring personal safety, power grid safety, and equipment safety.

[0033] The management of two tickets, work tickets and operation tickets, is an important support for the implementation of various professional work. In actual operation and maintenance work, various task orders, arrival records, safety cards, etc. are also used to cooperate with the two tickets.

[0034] In the prior art, two-ticket management applications, work tickets and operation tickets, involve too much manual filling in content and are generally based on manual verification, that is, relying on experienced power workers or safety managers to carefully review every field and detail of the work ticket to ensure that it complies with relevant safety regulations, operating procedures, and legal and regulatory requirements.

[0035] However, this method is time-consuming, labor-intensive, and inefficient, which increases the workload of front-line personnel; on-site construction monitoring is insufficient, and violations cannot be identified in a timely manner; and smart devices are not used fully, failing to achieve real-time supervision and comparison effects.

[0036] In response to the technical problems existing in the prior art, the inventors of the present application have the following idea: after obtaining the electric work ticket, the content of the electronic work ticket can be first identified, and the content of the work ticket can be regarded as a graph structure, in which nodes represent entities in the text (such as equipment, operations, safety measures, etc.), and edges represent the relationships between them (such as dependencies, conflicts, etc.); then, based on the trained model, it can be determined whether there are logical errors, missing dependencies or dependency conflicts. The training of this model can be based on the construction of a graph structure based on historical electronic tickets, and the training can be based on the results corresponding to the historical electronic tickets to obtain a model for actual verification.

[0037] Wherein, matters not elaborated in detail are disclosed by the following embodiments.

[0038] The technical solution of the present application is described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0039] It is worth noting that the application fields of the disclosed methods, devices, electronic devices, systems and storage media are not limited.

[0040] Among them, the executor of this application is an electronic device, which can be a terminal device, etc.

[0041] Figure 1 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 1 ,like Figure 1 As shown, the method may include the following steps: Step 11, obtaining target content information of the power work ticket; The target content information may refer to recording and managing various information in the maintenance and overhaul of power equipment, including equipment information, operation information, safety measures information, working conditions, etc.

[0042] Optionally, you can also: input the target content information into an anomaly detection model to obtain abnormal data corresponding to the power work ticket. The anomaly detection model is trained based on a large amount of work ticket content information and is used to determine the abnormal points existing in the target content information; based on the abnormal data, determine the abnormal results of the target content information.

[0043] In this embodiment, before inputting the target content information into the anomaly detection model, the data of the target content information needs to be preprocessed, including cleaning, standardizing or normalizing the data to remove noise and outliers to ensure the quality and consistency of the data.

[0044] Then, a subsample is randomly selected from the target content information, and the data is recursively segmented, that is, a feature is randomly selected in the subsample, and a segmentation point is randomly selected within the value range of the feature for segmentation. A tree node is constructed according to the segmentation result, and the above steps are recursively repeated for each child node to construct an isolation tree, thereby constructing an isolation forest.

[0045] Furthermore, the isolation forest is used to calculate the path length of each data point in the target content information in each isolated tree, the anomaly threshold is set according to the path length output by the model, and the anomaly score is calculated based on the path length. If the anomaly score is greater than the preset anomaly threshold, the data point is an abnormal data.

[0046] Therefore, the new power work ticket data is input into the anomaly detection model. The model will determine whether the data is abnormal based on the path length and give a corresponding abnormality degree score, thereby determining the abnormal result of the target content information, that is, judging whether the new work ticket data is abnormal.

[0047] It can be understood that by performing anomaly detection on the target content information through the anomaly detection model, abnormal data in the power work ticket can be discovered in time, thereby discovering potential safety hazards and avoiding safety accidents caused by erroneous or incomplete work ticket information. The anomaly detection model can effectively identify abnormal data points, thereby reducing the errors caused by human subjective judgment and improving the accuracy and reliability of the detection results.

[0048] In some embodiments, the distance between each data in the target content information and its K nearest neighbors can also be calculated by the K-Nearest Neighbors (KNN) method, and data points with larger distances are regarded as anomalies, thereby completing the anomaly detection process.

[0049] Step 12: According to the target content information of the electric power work ticket, determine the target graph structure data corresponding to the target content information.

[0050] Among them, graph structured data is a data representation method, in which data points (nodes) and the relationships (edges) between data points are represented as graphs. Converting the target content information of the power work ticket into graph structured data means representing each information item in the power work ticket as a node in the graph, and representing the relationship between them as an edge in the graph. Specifically, each information item in the power work ticket (such as equipment, operating steps, safety measures, etc.) is defined as a node in the graph; the edges between the nodes are defined based on the relationship between the information items (such as the sequence of operating steps, the association between safety measures and operating steps, etc.). In some embodiments, attributes can also be attached to the graph structured data, that is, relevant attribute information is attached to each node and edge, such as the type of node, the weight of the edge, etc.

[0051] For example, for an electricity work ticket, it contains target content information: equipment A needs maintenance, operation steps 1, 2, and 3 need to be performed in sequence, and safety measures X and Y need to be completed before operation step 1. This information can be converted into corresponding graph structure data: nodes: equipment A, operation step 1, operation step 2, operation step 3, safety measure X, safety measure Y; edges: operation step 1 to operation step 2, operation step 2 to operation step 3, safety measure X to operation step 1, safety measure Y to operation step 1, thereby constructing a graph structure data reflecting the content of the electricity work ticket for further analysis and processing.

[0052] It can be understood that by converting the content information of the power work ticket into graph structured data, it is possible to more effectively identify abnormal situations in the work ticket, such as the missing of certain key nodes or the unreasonable connection method of certain edges. On the other hand, through graph structured data analysis, bottlenecks and optimization points in the workflow can be discovered, thereby improving the workflow.

[0053] Step 13: Input the target graph structure data into a pre-trained dependency verification model to obtain a first verification result.

[0054] Among them, the first verification result is used to indicate whether there is a logical error, missing dependency or dependency conflict in the power work ticket; the dependency verification model is a verification model trained based on the graph structure data corresponding to the historical power work ticket and the verification results corresponding to the graph structure data, and is used to obtain the dependency relationship between each entity in the target content information.

[0055] In an embodiment of the present application, the dependency verification model can be a graph neural network (GNN), such as a graph convolutional network (GCN) or a graph attention network (GAT), etc. By iteratively updating the representation of the nodes, the model can capture the complex dependencies between the nodes. Specifically, the target graph structure data of the current power work ticket is input into the pre-trained dependency verification model, and the model will analyze the nodes and edges in the target graph structure data, check the dependencies between them, and output the first verification result.

[0056] The first verification result output by the dependency verification model is used to indicate whether there are logical errors, missing dependencies or dependency conflicts in the power work ticket. These problems have a corresponding relationship with the target graph structure data. Specifically, logical errors are manifested as logical inconsistencies or errors in the graph structure data, such as an operation step being executed without completing the predecessor step; missing dependencies are manifested as certain necessary dependencies are missing in the graph structure data, such as an operation step lacks necessary safety measures; dependency conflicts are manifested as the existence of conflicting dependencies in the graph structure data, such as a circular dependency between two operation steps.

[0057] It should be noted that the pre-training process of the dependency verification model includes: first, collecting a large number of historical power work tickets and their corresponding graph structure data. Then, verifying these graph structure data, marking logical errors, missing dependencies or dependency conflicts, etc. Finally, using these labeled data to train the model, by learning and parsing the relationships in the graph, the model can capture the complex dependencies between nodes, so that it can automatically identify and verify the dependencies in new graph structure data.

[0058] Therefore, after the model training is completed, the new power work ticket content is input into the model, and the model will verify the power work ticket based on the learned relationship to check whether there are logical errors, missing dependencies or conflicts, etc.

[0059] In some embodiments, in addition to graph neural networks, the dependency verification model can also be a node embedding algorithm, such as DeepWalk, Node2Vec, LINE, etc., which captures the relationship between nodes by embedding nodes into a low-dimensional vector space, and then uses these embedded vectors for further dependency analysis and error detection.

[0060] It can be understood that by inputting the target graph structure data into the pre-trained dependency verification model, logical errors, missing dependencies and dependency conflicts in the power work ticket can be automatically identified. This process is based on historical data and machine learning models, which can effectively learn and parse the complex relationships in the graph, thereby improving the efficiency and accuracy of the verification and ensuring the correctness and safety of the power work ticket.

[0061] The intelligent verification method for the content of an electric work ticket provided in the embodiment of the present application obtains the target content information of the electric work ticket, and then determines the target graph structure data corresponding to the target content information according to the target content information of the electric work ticket, and finally inputs the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; the first verification result is used to indicate whether there is a logical error, missing dependency, or dependency conflict in the electric work ticket; the dependency verification model is a verification model trained based on the graph structure data corresponding to the historical electric work ticket and the verification results corresponding to the graph structure data, and is used to obtain the dependency between each entity in the target content information. In this technical solution, the dependency verification model can be used to determine the dependency between each entity in the electric work ticket, and judge whether there is a logical error, missing dependency, or dependency conflict in the electric work ticket, so as to efficiently assist the user to determine the accuracy of the electric work ticket based on this.

[0062] Based on the above embodiment, the target graph structure data is the graph structure of the power work ticket; Figure 2 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 2 ,like Figure 2 As shown, step 12 may include the following steps: Step 21: Structural processing is performed on the target content information to obtain entities involved in the target content information and the relationships between the entities.

[0063] Among them, structured processing refers to converting unstructured or semi-structured data (such as text, tables, etc.) into a structured data form. In this embodiment, structured processing means extracting the information in the power work ticket and representing it in the form of a graph.

[0064] The entities involved in the target content information include equipment, tasks, personnel, time, location, etc. The relationships between the entities include various types, for example, a certain task is performed by a certain person, a certain equipment needs to be used in a certain task, etc.

[0065] It can be understood that by structuring the target content information, the information in the power work ticket can be better organized and managed to facilitate query and analysis. In some embodiments, algorithms can also be used for automated processing based on the obtained structured data, such as task scheduling, risk assessment, etc.

[0066] Step 22: Use the entities as nodes of the target graph structure data, and use the relationships between the entities as edges of the target graph structure data.

[0067] In this embodiment, each entity can be represented as a node in the graph, for example, a specific device can be a node, and a specific task can be another node. At the same time, each relationship can be represented as an edge in the graph and connect related nodes, for example, the edge between the task node and the personnel node can represent the "execution" relationship, and the edge between the task node and the device node can represent the "use" relationship, etc.

[0068] It can be understood that by converting entities and the relationships between entities into nodes and edges of target graph structure data, the organization of data is clearer and easier to understand and manage.

[0069] The intelligent verification method for the content of the electric work ticket provided in the embodiment of the present application obtains the entities involved in the target content information and the relationships between the entities by performing structured processing on the target content information, and then uses the entities as nodes of the target graph structure data, and uses the relationships between the entities as edges of the target graph structure data. This technical solution realizes the construction of the graph structure of the electric work ticket, which provides a basis for the subsequent accurate determination of the verification results.

[0070] Based on the above embodiment, the target graph structure data is the graph structure of the power work ticket; Figure 3 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 3 ,like Figure 3 As shown, after step 11, the following steps may be included: Step 31: input the target content information of the electric power work ticket into a pre-trained preliminary verification model to obtain a second verification result.

[0071] The second verification result is used to identify key information in the power work ticket to determine whether the format or content of the power work ticket is wrong.

[0072] In this embodiment, the pre-trained preliminary verification model may refer to a Transformer model, and therefore it is necessary to first convert the target content information of the electric power work ticket into a format that can be processed by the model, such as converting the text information into a word vector or an embedded representation, and then using the Transformer model to extract the features of the input data through a self-attention mechanism, capture the key information and relationships therein, and generate a second verification result, including identified key information, such as equipment name, task time, operation steps, etc., so as to further identify possible format or content errors in the electric power work ticket.

[0073] It is understandable that the Transformer model is usually pre-trained through large-scale data, and processes the input data based on the self-attention mechanism and deep learning capabilities. At the same time, it can also process the input data in parallel, thereby improving the speed and efficiency of verification.

[0074] Step 32: extract features from the target content information of the electric power work ticket to obtain vector features of each part of the target content information.

[0075] Among them, feature extraction of the target content information of the power work ticket can be carried out by using natural language processing (NLP) technology, such as word embeddings, term frequency-inverse document frequency (TF-IDF), bidirectional encoder representations (Bidirectional Encoder Representations from Transformers, BERT), etc., to convert the text into a vector representation.

[0076] In this embodiment, the sentence vector representation function of BERT is used to extract features of the target content information of the power work ticket (such as work tasks, safety measures, etc.). The result of the feature extraction is usually represented in the form of a vector, that is, a vector feature. At the same time, multiple feature extractions are performed on the target content information of the power work ticket, and finally a comprehensive vector feature, that is, a high-dimensional vector, is obtained, which represents all the key information of this power work ticket.

[0077] Furthermore, the method also includes: inputting the vector features into a pre-trained semantic understanding model to obtain a third verification result, the third verification result is used to express the rationality and consistency of the logical relationship and context dependency between the various fields in the target content information, the semantic understanding model is constructed by an attention mechanism and trained based on the vector features corresponding to the historical power work tickets and the verification results corresponding to the historical power work tickets.

[0078] In this embodiment, the pre-trained semantic understanding model may refer to a model customized for the content of power work tickets based on the Transformer structure, such as BERT, Generative Pre-trained Transformer (GPT), etc., which contains more layers and larger embedding dimensions to capture more complex semantic information and contextual relationships. Specifically, the use of the self-attention mechanism (Self-Attention) to capture important information and long-distance dependencies in the input data (vector features) can enhance the model's attention to key information, and by calculating the attention weight between each word and other words and dynamically assigning weights to different parts, the model can focus on the key parts in the input sequence.

[0079] At the same time, the model is trained based on a large amount of historical power work ticket data and its corresponding vector features and verification results. By learning these historical data, the model can capture common patterns and logical relationships in power work tickets. Specifically, the model is trained using a professional corpus containing a large amount of power work ticket text. Through the two stages of pre-training and fine-tuning, the model can accurately understand the professional terminology, work processes and safety regulations of the power industry.

[0080] For example, assuming that the target content information contained in the power work ticket includes tasks, equipment, time, and personnel, the feature extraction process converts this information into vector features, and then inputs it into the semantic understanding model. The model may check the following logical relationships and context dependencies, such as whether "task" is reasonably associated with "equipment"; whether "time" is within a reasonable working time range; whether "personnel" is qualified to perform "tasks", etc. If the model finds that the time is not within a reasonable range or the personnel are not qualified to perform the task, these errors will be marked in the third verification result.

[0081] It can be understood that through this semantic understanding model, key information such as equipment name, operating instructions, safety measures, etc. can be automatically identified to ensure the accuracy and completeness of this information in the work ticket. At the same time, through deep learning technologies such as self-attention mechanism, the model can capture the logical relationship and context dependency between the various fields in the power work ticket, so as to judge the consistency and rationality between them, and then can efficiently and accurately identify logical errors and inconsistencies in power work tickets, detect potential problems such as missing key information, irregular filling methods or potential safety hazards, etc., improve the accuracy and efficiency of verification, enhance safety and compliance, and by automatically prompting these problems so that staff can modify and improve the work ticket in time, it provides important support for the efficient operation and maintenance of the power system.

[0082] Step 33: Determine the semantic similarity between each vector feature according to a preset similarity algorithm. The semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information.

[0083] In this embodiment, the preset similarity algorithm can be a measurement method such as cosine similarity or Euclidean distance, which is used to verify the consistency between fields, detect duplicate or similar content, detect duplicate or similar content, etc. during the verification process. Specifically, by calculating the semantic similarity between different fields in the work ticket, it can be determined whether they are semantically consistent, such as whether the description in the safety measures field matches the description of the work task, and whether there is a semantic conflict or inconsistency. At the same time, in a large power system, there may be multiple work tickets involving similar operations or equipment. By calculating the semantic similarity between these work tickets, it can be detected whether there is duplicate or similar content, thereby avoiding redundancy and confusion. In addition, the power work ticket usually follows a certain template and format. By calculating the semantic similarity between the work ticket and the standard template, it can assist in determining whether the power work ticket meets the specified template and format requirements.

[0084] It should be noted that cosine similarity measures the cosine value of the angle between two vectors and determines the semantic similarity between each vector feature based on the cosine value of the angle, while Euclidean distance measures the straight-line distance between two vectors. The smaller the value, the more similar the two vectors are.

[0085] It can be understood that by determining the semantic similarity between each vector feature, on the basis of the preliminary verification model, it is possible to further verify the consistency between each field in the work ticket, detect duplicate or similar content, and assist in template matching, thereby ensuring the accuracy and completeness of the work ticket content through in-depth verification.

[0086] Step 34: Comprehensively evaluate the electric power work ticket based on the second verification result and the semantic similarity.

[0087] It is understandable that by performing a preliminary check on the target content information of the power work ticket through the preliminary check model and obtaining the second check result, it is possible to identify obvious errors in the format or content, ensure the elimination of basic errors, and improve the accuracy of subsequent processing. By calculating the semantic similarity between each vector feature through a preset similarity algorithm, potential content duplication, similarity or inconsistency problems can be detected. Therefore, based on the second check result and semantic similarity, a comprehensive evaluation of the power work ticket can be performed, which can combine format errors, content errors and semantic similarity to provide a comprehensive evaluation result.

[0088] Furthermore, for the above implementation process, homomorphic encryption or differential privacy can be used to encrypt the target verification results, which include: a first verification result, a second verification result, and a third verification result.

[0089] In this embodiment, in order to protect the privacy of data, homomorphic encryption or differential privacy can be used. Specifically, homomorphic encryption refers to the use of fully homomorphic encryption or semi-homomorphic encryption to protect the security of sensitive data, allowing calculations to be performed on encrypted data without decrypting the data, thereby protecting data privacy. For example, the first, second, and third verification results can be directly used for verification in an encrypted state without exposing the original data. Differential privacy refers to protecting the privacy of individual data by adding noise, and by setting a reasonable privacy budget to balance the relationship between privacy protection and data utility, so that even if the verification results are obtained, it is difficult to infer the original data.

[0090] It is understandable that by adopting homomorphic encryption or differential privacy technology, it is possible to prevent the power work ticket from being illegally accessed or tampered with during transmission and storage, ensuring the confidentiality and integrity of the data, while reducing the risk of human error and improving the overall security of the system.

[0091] The intelligent verification method for the content of the electric work ticket provided in the embodiment of the present application obtains a second verification result by inputting the target content information of the electric work ticket into a pre-trained preliminary verification model; the second verification result is used to identify the key information in the electric work ticket to determine the format or content errors in the electric work ticket, and to extract features of the target content information of the electric work ticket to obtain the vector features of each part of the target content information, and then determine the semantic similarity between each vector feature according to a preset similarity algorithm, and the semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information, and finally, the electric work ticket is comprehensively evaluated according to the second verification result and the semantic similarity. In this technical solution, the work ticket is comprehensively evaluated in combination with the results of the preliminary verification and the semantic similarity, and the user can be given modification suggestions or risk warnings.

[0092] Based on the above embodiment, the target graph structure data is the graph structure of the power work ticket; Figure 4 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 4 ,like Figure 4 As shown, the method may also include the following steps: Step 41: Determine the optimal solution set among multiple optimization objectives.

[0093] In this embodiment, for the dependency verification model, the optimization goal may be to minimize dependency conflicts, maximize task parallelism, minimize operation time, etc. The optimal solution set is to find a balance point between these optimization goals so that the dependency is best handled and conflicts and delays are reduced.

[0094] For the semantic understanding model, optimization goals may include improving the accuracy of semantic parsing, reducing misunderstandings and ambiguities, and increasing processing speed. The optimal solution is to find a balance between these optimization goals so that semantic understanding is both accurate and efficient.

[0095] For the preliminary verification model, the optimization goals may include improving the verification accuracy, reducing false positives and false negatives, and increasing the verification speed. The optimal solution set is to find a balance between these optimization goals to make the preliminary verification more accurate.

[0096] Furthermore, the optimal solution set between these objectives can be found by adopting the non-dominated sorting genetic algorithm II (NSGA-II), which can balance multiple objectives through fast non-dominated sorting, crowding distance calculation and elite retention strategy.

[0097] It can be understood that by determining the optimal solution set among multiple optimization objectives, the model can adapt to complex and changing environments and improve the adaptability and flexibility of the system.

[0098] Step 42: During multiple runs of the target model, the parameters of the target model are adjusted in real time to obtain different output solution sets; Among them, the target model is any one of the dependency verification model, the semantic understanding model, and the preliminary verification model.

[0099] In this embodiment, during the operation of the algorithm, it is necessary to collect output results and related performance indicators (such as accuracy, efficiency, error rate, etc.), and use these feedbacks to adjust model parameters and adjust algorithm parameters in real time, such as population size, crossover probability, mutation probability, etc., to optimize algorithm performance.

[0100] It is understandable that by adjusting parameters in real time, the model can better adapt to different input data and operating environments, improve the accuracy and reliability of the output results, and be able to adapt to different scenarios and changing conditions, improve the robustness and adaptability of the model, and reduce dependence on specific parameter settings.

[0101] Step 43: Evaluate different output solution sets according to the optimal solution sets among multiple optimization objectives to obtain the optimal target model.

[0102] In this embodiment, for multi-objective optimization, it is impossible to improve one objective without deteriorating other objectives. These solution sets represent the best trade-offs between various objectives. Therefore, different output solution sets are evaluated. The evaluation criteria are based on multiple optimization objectives. By comparing the performance of these solution sets on various objectives, it is determined which solution sets belong to the optimal solution set as the optimal objective model, that is, the solution that best meets the current needs is selected as the final solution for power work ticket verification.

[0103] It can be understood that by determining the optimal solution set between multiple optimization objectives, a balance point can be found between different models so that the system achieves the best performance as a whole, rather than just the optimization of a single model. By optimizing the model, a data-driven approach can be used for decision-making, ensuring that the system is more intelligent and efficient when handling complex tasks.

[0104] In the above implementation, the multi-objective optimization method will comprehensively consider the output results of these different target models, and find the optimal verification strategy or parameter configuration by balancing various optimization goals (such as improving verification accuracy, reducing false alarm rate, speeding up processing speed, etc.). In this way, the entire verification system can verify the content of the power work ticket more comprehensively and accurately, thereby improving the safety production level of the power industry.

[0105] The intelligent verification method for the content of the electric work ticket provided in the embodiment of the present application determines the optimal solution set between multiple optimization targets for the target model, where the target model is any one of the dependency verification model, the semantic understanding model, and the preliminary verification model. Then, during the multiple runs of the target model, the parameters of the target model are adjusted in real time to obtain different output solution sets. Finally, different output solution sets are evaluated according to the optimal solution set between multiple optimization targets to obtain the optimal target model. In this technical solution, the verification of the dependency verification model, the semantic understanding model, and the preliminary verification model is realized by using the target optimization method, so that more accurate verification results can be obtained later.

[0106] Based on the above embodiment, the target graph structure data is the graph structure of the power work ticket; Figure 5 Schematic diagram of the process of the intelligent verification method of the power work ticket content provided in the embodiment of the present application Figure 5 ,like Figure 5 As shown, the method may also include the following steps: Step 51: Determine the time series analysis information corresponding to the first power work ticket within a first preset time period before the current time node.

[0107] The current time node refers to the current time point of the system, that is, the moment for analysis and decision-making.

[0108] The first preset duration refers to a fixed time length traced back from the current time node. For example, this period of time can be the past 24 hours, a week, a month, etc. The specific duration is set according to business needs.

[0109] The first power work ticket refers to a specific power work ticket within a preset time period.

[0110] Time series analysis information may refer to information obtained by analyzing the time series data of the power work ticket within a preset time period. This information may include the submission time, approval time, execution time, completion time, etc. of the work ticket, as well as the intervals and trends between these time points.

[0111] In this embodiment, the autoregressive integrated moving average model (ARIMA) can be used to perform time series analysis on the historical data of work tickets to predict the number and type of future work tickets. Specifically, it is necessary to collect historical data of power work tickets, including the number, type, processing time, etc.; then, based on these data, time series analysis is performed to build a prediction model, so as to estimate the parameters of the model through historical data, such as the difference order, autoregressive coefficient, and moving average coefficient; finally, the constructed model is used to predict the number and type of future work tickets, etc., to provide a reference for the allocation and scheduling of power work tickets.

[0112] It is understandable that through time series analysis, bottlenecks and delays in the processing of power work tickets can be identified, the workflow can be optimized, and overall work efficiency can be improved. At the same time, time series analysis can help predict future workload and resource requirements, prepare in advance, and avoid waste of resources and work backlogs.

[0113] Step 52: Input the time series analysis information into the prediction model to obtain the prediction result.

[0114] Among them, the prediction result expresses the number and type of second power work tickets within a second preset time period after the current time node; the prediction model is trained based on historical power work tickets.

[0115] It can be understood that by using historical data and time series analysis information, the predictive model is able to identify patterns and trends in the data, thereby improving the accuracy of the prediction. Based on the prediction results, potential risks and problems can be identified, and preventive measures can be taken in advance to reduce the occurrence of failures and accidents and improve the safety and reliability of the system.

[0116] Optionally, the target model may be optimized according to the abnormal result to obtain an optimized target model.

[0117] It should be noted that abnormal results refer to abnormal situations or results that deviate from expectations during the operation of the model. These abnormalities may be caused by data noise, improper model parameter settings, changes in the external environment, etc. By analyzing the abnormal results, the root causes of the abnormalities can be found, such as checking data quality, model parameters, model structure, etc., so that the target model can be optimized and adjusted according to the diagnosis results, including retraining the model, adjusting model parameters, improving model structure, introducing new features, etc.

[0118] It is understandable that after the optimization process, the model is validated to ensure that its performance is improved, abnormal results are reduced, its performance and stability are improved, it can better cope with abnormal situations and provide more accurate and reliable results.

[0119] The intelligent verification method for the content of the electric power work ticket provided in the embodiment of the present application determines the time series analysis information corresponding to the first electric power work ticket within the first preset time before the current time node, and then inputs the time series analysis information into the prediction model to obtain the prediction result, wherein the prediction result expresses the number and type of the second electric power work ticket within the second preset time after the current time node; the prediction model is obtained by training based on historical electric power work tickets. In this technical solution, the prediction of electronic tickets that may appear in the future is realized, providing a reference for the allocation and scheduling of electric power work tickets.

[0120] Optionally, based on the above embodiments, a reinforcement learning architecture can also be introduced, that is, defining the state space (such as the content of the current electronic work ticket, historical verification results, etc.) and the action space (such as adjusting the verification strategy, modifying parameters, etc.).

[0121] Among them, a policy network can be trained based on the Q-learning algorithm to enable it to select the optimal action based on the current state. Through continuous trial and error and feedback, the policy network can gradually learn the optimal verification strategy. During the verification process, the agent can obtain the effect and existing problems of the current strategy through the real-time feedback mechanism. Based on these feedbacks, the agent can dynamically adjust its behavior strategy to better adapt to the needs of the verification task.

[0122] For different types of verification tasks, multiple strategies can be designed and integrated and switched according to actual needs during the verification process. This multi-strategy integration and switching method can further improve the flexibility and accuracy of verification.

[0123] In complex verification scenarios, a single agent may not be able to handle all tasks. In this case, a multi-agent collaborative reinforcement learning algorithm can be introduced. Through the collaboration and competition between multiple agents, a more comprehensive and accurate verification of the content of the electronic work ticket can be achieved.

[0124] Exemplarily, the agent system allows each agent to learn and make decisions independently in a distributed environment, and can realize information sharing and strategy coordination through communication and collaboration mechanisms. In a multi-agent system, agents can complete tasks together through a collaborative mechanism, or improve their learning efficiency and performance through a competitive mechanism. This combination of collaboration and competition can stimulate the learning potential of agents and promote the continuous optimization of verification tasks. In the reinforcement learning algorithm of multi-agent collaboration, the strategies of each agent need to be optimized collaboratively to achieve the optimal performance of the overall system. This can be achieved by designing a reasonable reward function and collaboration mechanism to ensure the accuracy and efficiency of the verification task. Technologies such as deep learning can be used to realize knowledge sharing and migration between agents. By sharing knowledge bases or model parameters, the learning ability and generalization ability of agents can be improved, thereby achieving better processing of complex verification tasks. Then the above-mentioned dependency verification model, preliminary verification model, semantic understanding model, and multi-objective optimization can all work as independent agents, thereby forming an agent system to achieve more comprehensive and efficient verification.

[0125] Based on the dependency verification model, preliminary verification model, semantic understanding model, and the combination of multi-objective optimization and reinforcement learning, an efficient and accurate intelligent verification method for the content of power work tickets can be constructed. This method can automatically identify errors and abnormal content in work tickets, improve verification efficiency and accuracy, and provide strong guarantees for the company's safe production.

[0126] In the above embodiment, Figure 6 A schematic diagram of the structure of the intelligent verification system for the contents of an electric power work ticket provided in an embodiment of the present application is shown in FIG. Figure 6 As shown, the system includes: an electronic device 61 and a first device 62; The electronic device 61 obtains the power work ticket from the first device 62 and executes the intelligent verification method of the power work ticket content of any one of the above method embodiments.

[0127] The intelligent verification system for the contents of electric work tickets provided in the embodiment of the present application, the principles and technical effects of its technical solution are as shown in the above embodiment and will not be repeated here.

[0128] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0129] Figure 7 This is a schematic diagram of the structure of the intelligent verification device for the content of the power work ticket provided in the embodiment of the present application. Figure 7 As shown, the device comprises: An acquisition module 71 is used to acquire target content information of an electric power work ticket; A determination module 72, for determining target graph structure data corresponding to the target content information according to the target content information of the electric power work ticket; Processing module 73 is used to input the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; the first verification result is used to indicate whether there is a logical error, missing dependency or dependency conflict in the power work ticket; the dependency verification model is a verification model trained based on the graph structure data corresponding to the historical power work ticket and the verification results corresponding to the graph structure data, and is used to obtain the dependency relationship between each entity in the target content information.

[0130] In one or more embodiments, the target graph structure data is a graph structure of an electric power work ticket; Accordingly, the determination module 72 is specifically configured to: Performing structural processing on the target content information to obtain entities involved in the target content information and the relationships between the entities; The entities are taken as nodes of the target graph structure data, and the relationships between the entities are taken as edges of the target graph structure data.

[0131] In one or more embodiments, after obtaining the target content information of the electric power work ticket, the processing module 73 is further used to: Inputting the target content information of the electric power work ticket into a pre-trained preliminary verification model to obtain a second verification result; the second verification result is used to identify key information in the electric power work ticket to determine whether there is a format or content error in the electric power work ticket; Extract features of the target content information of the power work ticket to obtain vector features of each part of the target content information; According to a preset similarity algorithm, the semantic similarity between each vector feature is determined. The semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information; Based on the second verification result and semantic similarity, a comprehensive evaluation is conducted on the power work ticket.

[0132] In one or more embodiments, after extracting features from the target content information of the electric power work ticket to obtain vector features of each part of the target content information, the processing module 73 is further used to: The vector features are input into a pre-trained semantic understanding model to obtain a third verification result, which is used to express the rationality and consistency of the logical relationship and context dependency between the various fields in the target content information. The semantic understanding model is constructed by an attention mechanism and trained based on the vector features corresponding to the historical power work tickets and the verification results corresponding to the historical power work tickets.

[0133] In one or more embodiments, the processing module 73 is further configured to: Determine the optimal solution set among multiple optimization objectives; During multiple runs of the target model, the parameters of the target model are adjusted in real time to obtain different output solution sets, where the target model is any one of the dependency verification model, the semantic understanding model, and the preliminary verification model; Different output solution sets are evaluated according to the optimal solution set among multiple optimization objectives to obtain the optimal target model.

[0134] In one or more embodiments, the processing module 73 is further configured to: Determine time series analysis information corresponding to a first electric power work ticket within a first preset time period before a current time node; The time series analysis information is input into the prediction model to obtain a prediction result, which expresses the number and type of second power work tickets within a second preset time period after the current time node; the prediction model is trained based on historical power work tickets.

[0135] In one or more embodiments, the processing module 73 is further configured to: Input the target content information into the anomaly detection model to obtain the abnormal data corresponding to the power work ticket. The anomaly detection model is trained by the contents of multiple work tickets and is used to determine the abnormal points in the target content information; Based on the abnormal data, the abnormal results of the target content information are determined.

[0136] In one or more embodiments, the processing module 73 is further configured to: According to the abnormal results, the target model is optimized to obtain the optimized target model.

[0137] In one or more embodiments, the processing module 73 is further configured to: The target verification results are encrypted using homomorphic encryption or differential privacy, and the target verification results include: a first verification result, a second verification result, and a third verification result.

[0138] The device provided in the embodiments of the present application can be used to execute the method in any of the above embodiments. Its implementation principle and technical effects are similar and will not be repeated here.

[0139] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. In addition, all or part of these modules can be integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0140] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 8 As shown, the electronic device may include: a processor 81, a memory 82, and computer program instructions stored in the memory 82 and executable on the processor 81. When the processor 81 executes the computer program instructions, the method provided in any of the aforementioned embodiments is implemented.

[0141] Optionally, the above-mentioned components of the electronic device may be connected via a system bus.

[0142] The memory 82 may be a separate storage unit or a storage unit integrated in the processor 81. The number of the processor 81 may be one or more.

[0143] It should be understood that the processor 81 can be a central processing unit (CPU), or other general-purpose processors 81, digital signal processors 81 (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor 81 can be a microprocessor 81 or the processor 81 can also be any conventional processor 81, etc. The steps of the method disclosed in the present application can be directly embodied as being executed by the hardware processor 81, or can be executed by a combination of hardware and software modules in the processor 81.

[0144] The system bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The memory 82 may include a random access memory 82 (RAM), and may also include a non-volatile memory 82 (NVM), such as at least one disk storage 82.

[0145] All or part of the steps of the above-mentioned method embodiments can be completed by hardware related to program instructions. The above-mentioned program can be stored in a readable memory 82. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the above-mentioned memory 82 (storage medium) includes: read-only memory 82 (ROM), RAM, flash memory 82, hard disk, solid state hard disk, magnetic tape, floppy disk, optical disc and any combination thereof.

[0146] The electronic device provided in the embodiment of the present application can be used to execute the method provided in any of the above method embodiments. Its implementation principles and technical effects are similar and will not be repeated here.

[0147] An embodiment of the present application provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the computer executes the above method.

[0148] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0149] Optionally, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0150] An embodiment of the present application also provides a computer program product, which includes a computer program. The computer program is stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, the above method can be implemented.

[0151] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An intelligent verification method for the content of an electric power work ticket, characterized in that: include: Obtain target content information of the power work ticket; According to the target content information of the electric power work ticket, determining the target graph structure data corresponding to the target content information; Inputting the target graph structure data into a pre-trained dependency verification model to obtain a first verification result; The first verification result is used to indicate whether the electric power work ticket has a logical error, a dependency missing, or a dependency conflict; The dependency verification model is a verification model trained based on graph structure data corresponding to historical power work orders and verification results corresponding to the graph structure data, and is used to obtain the dependency relationships between entities in the target content information.

2. The method according to claim 1, characterized in that The target graph structure data is the graph structure of the electric power work ticket; Correspondingly, the target graph structure data corresponding to the target content information is determined according to the target content information of the electric power work ticket, including: Performing structural processing on the target content information to obtain entities involved in the target content information and relationships between the entities; The entities are taken as nodes of the target graph structure data, and the relationships between the entities are taken as edges of the target graph structure data.

3. The method according to claim 1 or 2, characterized in that: After obtaining the target content information of the electric power work ticket, the method further includes: Inputting the target content information of the electric power work ticket into a pre-trained preliminary verification model to obtain a second verification result; the second verification result is used to identify key information in the electric power work ticket to determine whether the format or content of the electric power work ticket is wrong; Extracting features of the target content information of the electric power work ticket to obtain vector features of each part of the target content information; Determine the semantic similarity between each vector feature according to a preset similarity algorithm, wherein the semantic similarity is used to express the consistency, repeated content, and similar content between each field in the target content information; The electric power work ticket is comprehensively evaluated according to the second verification result and the semantic similarity.

4. The method according to claim 3, characterized in that After extracting features from the target content information of the electric power work ticket to obtain vector features of each part of the target content information, the method further includes: The vector features are input into a pre-trained semantic understanding model to obtain a third verification result, which is used to express the rationality and consistency of the logical relationship and context dependency between the fields in the target content information. The semantic understanding model is constructed by an attention mechanism and trained based on the vector features corresponding to historical power work tickets and the verification results corresponding to historical power work tickets.

5. The method according to claim 4, characterized in that The method further comprises: Determine the optimal solution set among multiple optimization objectives; During multiple runs of the target model, adjusting the parameters of the target model in real time to obtain different output solution sets, wherein the target model is any one of a dependency verification model, a semantic understanding model, and a preliminary verification model; Different output solution sets are evaluated according to the optimal solution set among the multiple optimization objectives to obtain an optimal target model.

6. The method according to claim 4, characterized in that The method further comprises: Determine time series analysis information corresponding to a first electric power work ticket within a first preset time period before a current time node; The time series analysis information is input into a prediction model to obtain a prediction result, wherein the prediction result expresses the number and type of second power work tickets within a second preset time period after the current time node; the prediction model is trained based on historical power work tickets.

7. The method according to claim 5, characterized in that The method further comprises: Inputting the target content information into an anomaly detection model to obtain abnormal data corresponding to the power work ticket, wherein the anomaly detection model is trained with the contents of multiple work tickets and is used to determine abnormal points in the target content information; An abnormal result of the target content information is determined according to the abnormal data.

8. The method according to claim 7, characterized in that The method further comprises: According to the abnormal result, the target model is optimized to obtain an optimized target model.

9. The method according to claim 8, characterized in that The method further comprises: The target verification results are encrypted by homomorphic encryption or differential privacy, and the target verification results include: a first verification result, a second verification result, and a third verification result.

10. An intelligent verification system for the contents of an electric power work ticket, characterized in that: include: An electronic device and a first device; The electronic device obtains the electric power work ticket from the first device and executes the intelligent verification method for the content of the electric power work ticket as described in any one of claims 1-9.

Images