Abnormal report information generation method and device of power distribution network, terminal equipment and storage medium

Abstract

The invention discloses a power distribution network exception report information generation method and device, terminal equipment and a storage medium, and the method can convert the power distribution network exception information into a corresponding semantic vector, thereby obtaining a corresponding semantic vector list. Calculating the similarity between each semantic vector in the semantic vector list and the index vector of each regulation document, and taking the regulation document with the highest similarity as a target regulation document which is most matched and most similar to the abnormal information of the power grid without manual lookup and comparison; the report content in the target regulation document can be automatically extracted, and finally the abnormal report information corresponding to the power grid abnormal information is automatically generated based on the report content and each abnormal text of the power grid abnormal information, so that automatic matching of the regulation document is realized, and the report content in the target regulation document can be automatically extracted. Errors caused by human factors are avoided, the reporting efficiency is improved, and the accuracy of abnormal reporting information is also improved.

Description

Technical Field

[0001] The present invention relates to the field of computer processing technology, and in particular to a method, device, terminal equipment and storage medium for generating abnormal reporting information of a distribution network. Background Art

[0002] In the modern power grid dispatching business, the rapid and accurate reporting of abnormal information of the power grid has become a key factor in ensuring the synchronization of power grid events and maintaining the smooth operation of the power grid. Abnormal situations may involve equipment failures, safety hazards, operating errors and other aspects. Timely and effective reporting can help organizations reduce potential risks and respond quickly. The procedure documents usually describe in detail the processing procedures and reporting requirements for various abnormal situations. By consulting the procedure documents, the dispatcher can accurately judge the level of the abnormality, the reporting unit, and the specific reporting requirements, etc., and obtain the abnormal reporting information corresponding to the current abnormal situation, so that when the abnormality is reported based on the abnormal reporting information in the future, it can be ensured that the abnormal situation and abnormal information can be accurately reported to the corresponding unit.

[0003] However, in the traditional abnormal information reporting process, it is usually relied on to manually review a large number of procedure documents to match the corresponding procedure documents, and to extract the required abnormal situation handling process and reporting requirement information based on personnel experience, such as the unit of the abnormal report, the time of the abnormal report or the method of the abnormal report. The traditional abnormal information reporting method relies on manual matching of procedure documents, which requires a lot of time and effort, resulting in a low generation rate of abnormal reporting information, and is prone to errors due to personal negligence or misunderstanding, affecting the accuracy of the abnormal reporting information content. Summary of the invention

[0004] The embodiments of the present invention provide a method, apparatus, terminal device and storage medium for generating abnormal reporting information of a distribution network, which can convert each abnormal text in the abnormal information of the power grid into a semantic vector, and perform similarity calculation with the index vector of the procedure document, thereby realizing automatic matching of the procedure document. This can effectively solve the problem in the prior art of relying on manual reference to a large number of procedure documents to match the corresponding procedure documents, resulting in a low generation rate and low accuracy of abnormal reporting information.

[0005] An embodiment of the present invention provides a method for generating abnormal reporting information of a distribution network, comprising:

[0006] Acquire power grid abnormality information input by a user; wherein the power grid abnormality information includes: a number of abnormal texts for describing abnormal phenomena;

[0007] The power grid abnormality information is converted into a semantic vector list, and the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document is calculated; wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities;

[0008] The procedure document with the highest similarity is used as the target procedure document, and then the corresponding reporting content in the target procedure document is extracted; wherein the reporting content includes: reporting unit, reporting time and reporting method;

[0009] According to each abnormal text and the report content, the abnormal report information corresponding to the abnormal information of the power grid is generated.

[0010] Preferably, the calculating the similarity between each procedure document and the semantic vector list according to each semantic vector in the semantic vector list and the index vector of each procedure document comprises:

[0011] For each procedure document, respectively, calculating the similarity between each semantic vector in the semantic vector list and the index vector of the procedure document to obtain the similarity between each semantic vector and the procedure document;

[0012] For each procedure document, the average similarity between the procedure document and the semantic vector list is obtained based on the similarity between each semantic vector in the semantic vector list and the procedure document, and the total number of semantic vectors in the semantic vector list; the average similarity is used as the similarity corresponding to the procedure document and the semantic vector list.

[0013] Preferably, the reported content also includes a reporting template;

[0014] The extracting the corresponding reporting content in the target procedure document includes:

[0015] Obtaining a preset prompt template for guiding the model to output preset content; wherein the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template;

[0016] Perform a word segmentation operation on each abnormal text in the power grid abnormal information, and generate a word segmentation list corresponding to the power grid abnormal information according to each word segmented;

[0017] The preset prompt template, the target procedure document and the word segmentation list are input into a preset document processing model, so that the document processing model extracts word features of each word in the word segmentation list, captures context information of each text paragraph in the target procedure document, and generates corresponding text features according to the context information corresponding to each text paragraph; extracts the text paragraph with the highest degree of relevance to the word segmentation list from the target procedure document according to each word feature and each text feature, and generates reporting content corresponding to the preset prompt template according to the preset prompt template and the text paragraph;

[0018] The generation of the preset document processing model includes:

[0019] Generate training samples according to the sample of the regulation document, the sample of the word segmentation list corresponding to the sample of the power grid abnormal information, and the sample of the preset prompt template; wherein the sample of the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template;

[0020] Taking each training sample and the actual reported content corresponding to each training sample as input and the predicted reported content of each training sample as output, the document processing model to be trained is iteratively trained until the model converges to generate a preset document processing model.

[0021] Preferably, in each iterative training, a training sample is input into the document processing model, so that the document processing model extracts the text paragraph with the highest correlation with the word segmentation list sample from the procedure document sample according to the word features and text features in the training sample, and generates the reported content prediction result corresponding to the preset prompt template sample according to the preset prompt template and the text paragraph;

[0022] The reported content prediction results are compared with the actual reported content results, and the network parameters of the document processing model are adjusted according to the comparison results.

[0023] Preferably, the document processing model comprises: an attention mechanism layer;

[0024] The extracting, from the target procedure document, the text paragraph with the highest degree of relevance to the word segmentation list according to each word feature and each text feature, comprises:

[0025] The attention mechanism layer is used to calculate the cosine similarity between each word feature and each text feature;

[0026] For each text feature, based on the activation function, the cosine similarity between the text feature and each word feature is converted into a probability distribution, and the corresponding attention weight is assigned to the text feature according to each probability distribution; wherein the attention weight is used to characterize the correlation between the text feature and the word list;

[0027] According to each attention weight, the text paragraph corresponding to the text feature with the highest attention weight is determined.

[0028] Preferably, the document processing model comprises: a fusion layer;

[0029] The generating, according to the preset prompt template and the text paragraph, the reporting content corresponding to the preset prompt template comprises:

[0030] The fusion layer is used to splice each first text vector corresponding to the preset prompt template and each second text vector corresponding to the text paragraph in the vector space to generate a plurality of spliced ​​third text vectors; wherein the first text vector corresponding to the preset prompt template includes: a unit text vector, a time text vector, a method text vector and a template text vector; the second text vector corresponding to the text paragraph includes: a reporting text vector, a reporting time text vector, a reporting method text vector and a reporting template text vector;

[0031] According to each of the spliced ​​third text vectors, report content corresponding to the preset prompt template is generated.

[0032] Preferably, the abnormal text includes: the abnormal location, abnormal time, abnormal device type and abnormal status; the abnormal status is used to characterize the device status when the abnormal situation occurs; the reporting template includes: a time filling part, a unit filling part, a device type filling part and an abnormal situation filling part;

[0033] The generating, according to each abnormal text and the reported content, abnormal reporting information corresponding to the abnormal information of the power grid includes:

[0034] Fill the abnormal text of the abnormal occurrence time, the reporting unit in the report content, the abnormal text of the abnormal device type and the abnormal text of the abnormal state into the time filling part, the unit filling part, the device type filling part and the abnormal situation filling part of the report template respectively to generate a filled report template;

[0035] Exception reporting information corresponding to the power grid exception information is generated according to the filled reporting template, the exception text of the location where the exception occurred, the reporting time in the reporting content, and the reporting method in the reporting content.

[0036] Based on the above method embodiments, the present invention provides corresponding device embodiments.

[0037] An embodiment of the present invention provides an abnormality reporting information generating device for a distribution network, comprising: an abnormality information acquiring module, a similarity calculating module, a reporting content extracting module and an abnormality reporting information generating module;

[0038] The abnormal information acquisition module is used to acquire the abnormal power grid information input by the user; wherein the abnormal power grid information includes: a number of abnormal texts for describing the abnormal phenomenon;

[0039] The similarity calculation module is used to convert the power grid abnormality information into a semantic vector list, and calculate the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document, wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities;

[0040] The reporting content extraction module is used to take the procedure document with the highest similarity as the target procedure document, and then extract the corresponding reporting content in the target procedure document; wherein the reporting content includes: reporting unit, reporting time and reporting method;

[0041] The abnormal reporting information generating module is used to generate abnormal reporting information corresponding to the abnormal information of the power grid according to each abnormal text and reporting content.

[0042] Based on the above method embodiments, the present invention provides corresponding terminal device embodiments.

[0043] Another embodiment of the present invention 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, and when the processor executes the computer program, the method for generating abnormal reporting information of a distribution network described in the above-mentioned embodiment of the invention is implemented.

[0044] Based on the above method embodiments, the present invention provides corresponding storage medium item embodiments.

[0045] Another embodiment of the present invention provides a storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method for generating abnormal reporting information of a distribution network described in the above-mentioned embodiment of the invention.

[0046] The following beneficial effects are achieved by implementing the present invention:

[0047] The embodiment of the present invention provides a method, apparatus, terminal device and storage medium for generating abnormal reporting information of a distribution network. According to the abnormal power grid information input by the user, the present invention can convert each abnormal text in the abnormal power grid information into a corresponding semantic vector, thereby obtaining a semantic vector list corresponding to the abnormal power grid information. Based on each semantic vector in the semantic vector list and the index vector of each procedure document, the similarity between the semantic vector list and each procedure document can be calculated. Based on the procedure document with the highest similarity, a target procedure document that is most similar to the text of the abnormal power grid information can be obtained without manual review and comparison. Then, the reporting content in the target procedure document can be automatically extracted. Finally, based on the reporting unit, reporting time and reporting method in the reporting content, and each abnormal text, the abnormal reporting information corresponding to the abnormal power grid information is automatically generated. Compared with the prior art, the present invention can convert each abnormal text in the power grid abnormal information into a semantic vector, and perform similarity calculation with the index vector of the procedure document, thereby realizing automatic matching of the procedure document. In addition, the present invention improves the reporting efficiency by automatically extracting the reporting content in the target procedure document. Since errors caused by human factors are avoided, the quality and accuracy of the abnormal reporting information are further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of a method for generating abnormal reporting information of a distribution network provided by an embodiment of the present invention.

[0049] Figure 2 It is a structural diagram of a device for generating abnormal reporting information of a distribution network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] like Figure 1 FIG. 1 is a flow chart of a method for generating abnormality reporting information of a distribution network provided by an embodiment of the present invention. The method for generating abnormality reporting information of a distribution network comprises:

[0052] Step S1: obtaining power grid abnormality information input by a user; wherein the power grid abnormality information includes: a number of abnormal texts for describing abnormal phenomena;

[0053] Step S2: converting the power grid abnormality information into a semantic vector list, and calculating the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document, wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities;

[0054] Step S3: taking the procedure document with the highest similarity as the target procedure document, and then extracting the corresponding reporting content in the target procedure document; wherein the reporting content includes: reporting unit, reporting time and reporting method;

[0055] Step S4: Generate abnormal reporting information corresponding to the power grid abnormal information according to each abnormal text and reporting content.

[0056] For step S1, in a preferred embodiment, the user can input abnormal text for detailing the abnormal situation occurring in the power grid, such as the location where the abnormality occurred, the time when the abnormality occurred, the type of abnormal equipment, and the abnormal status. The abnormal status is used to characterize the status of the equipment when the abnormal situation occurs, such as whether the equipment is running, whether the operating parameters are abnormal (such as whether the current and voltage are beyond the normal range), whether the equipment has physical damage or fault indication, etc. The abnormal status describes the specific performance or condition of the equipment under abnormal circumstances. Based on the abnormal status, the nature and severity of the abnormal event can be more comprehensively understood, thereby matching the corresponding procedure documents.

[0057] For step S2, in a preferred embodiment, the present invention can convert the power grid anomaly information into a semantic vector list containing a plurality of semantic vectors before matching the power grid anomaly information with the procedure document, and the semantic vectors in the list correspond to the abnormal text.

[0058] Therefore, by converting power grid anomaly information into semantic vectors, the vector similarity calculation can be used to accurately match the regulation documents, which can capture the semantic similarity between texts rather than just the literal similarity, thereby improving the matching accuracy of the documents.

[0059] By calculating the similarity between each procedural document and the semantic vector list, the most relevant procedural document can be quickly located, reducing the time for manual review and screening of procedural documents and improving the efficiency of generating abnormal reporting information. Specifically, when calculating the similarity between each procedural document and the semantic vector list, it specifically includes:

[0060] For each procedure document, respectively, calculating the similarity between each semantic vector in the semantic vector list and the index vector of the procedure document to obtain the similarity between each semantic vector and the procedure document;

[0061] For each procedure document, the average similarity between the procedure document and the semantic vector list is obtained based on the similarity between each semantic vector in the semantic vector list and the procedure document, as well as the total number of semantic vectors in the semantic vector list; the average similarity is used as the similarity corresponding to the procedure document and the semantic vector list. It can be understood that the average similarity can more comprehensively reflect the overall correlation between the procedure document and the power grid abnormality information, and is an important basis for the subsequent positioning of the procedure document.

[0062] Illustratively, in order to match the semantic vector of power grid anomaly information, the present invention also needs to construct an index vector for each procedure document, and the constructed index vector can represent the main content and features of the procedure document, which is the basis for subsequent similarity calculation.

[0063] A procedure document generally refers to a standard document that specifies, explains or guides a process, operation or behavior. In the field of power dispatching, the procedure document matched by the present invention usually contains specific reporting rules, processes, requirements, etc. when handling abnormal power grid conditions, and the procedure document is an important reference for dispatchers when handling abnormal power grid conditions. Schematically, the procedure document may contain the following contents:

[0064] Reporting rules: The procedure document will clearly specify which unit or units should be responsible for reporting different types of power grid anomalies, as well as the priority and urgency of the report.

[0065] Reporting process: Detailed description of the complete steps from abnormality discovery to final reporting, including information collection, analysis and judgment, decision making and reporting execution.

[0066] Reporting requirements: including reporting time limit, reporting method (such as written report, telephone notification, etc.), format and content requirements of reporting information, etc.

[0067] Exception handling guidance: In addition to reporting rules, the procedure documents can also include preliminary handling guidance for power grid abnormalities to help dispatchers take necessary emergency measures before reporting.

[0068] Relevant regulations and standards: Reference or include other regulations, standards or guidance documents related to power grid anomaly handling to ensure consistency and compliance of the reporting process.

[0069] The present invention can automatically match relevant procedure documents based on the abnormal situation input by the dispatcher, thereby obtaining requirements and rules for abnormal reporting, and automatically generate abnormal reporting information of the current abnormal situation based on the relevant procedure documents.

[0070] In schematic form, even if the reporting rules of the procedure document are updated, the present invention only needs to directly update the content of the procedure document in the knowledge base when the procedure document changes. There is no need for the dispatcher to repeat the learning manually. Instead, the similarity between each semantic vector in the power grid anomaly information and the index vector of the procedure document needs to be calculated to achieve automatic matching of the procedure document, thereby reducing the learning cost and reference time of the new procedure.

[0071] For step S3, in a preferred embodiment, the present invention can extract the procedure document with the highest similarity based on the similarity corresponding to each procedure document and the semantic vector list, and mark it as the target procedure document to be processed, thereby extracting the reporting content that can be used to generate abnormal reporting information based on the target procedure document, such as the reporting unit, reporting time, reporting method and reporting template.

[0072] Illustratively, the embodiment of the present invention can process the standard procedure document and extract the content related to abnormal reporting through a preset document processing model, and then:

[0073] Obtaining a preset prompt template for guiding the model to output preset content; wherein the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template;

[0074] Perform a word segmentation operation on each abnormal text in the power grid abnormal information, and generate a word segmentation list corresponding to the power grid abnormal information according to each word segmented;

[0075] The preset prompt template, the target procedure document and the word segmentation list are input into a preset document processing model, so that the document processing model extracts word features of each word in the word segmentation list, captures context information of each text paragraph in the target procedure document, and generates corresponding text features according to the context information corresponding to each text paragraph; extracts the text paragraph with the highest degree of relevance to the word segmentation list from the target procedure document according to each word feature and each text feature, and generates reporting content corresponding to the preset prompt template according to the preset prompt template and the text paragraph;

[0076] The generation of the preset document processing model includes:

[0077] Generate training samples according to the sample of the regulation document, the sample of the word segmentation list corresponding to the sample of the power grid abnormal information, and the sample of the preset prompt template; wherein the sample of the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template;

[0078] Taking each training sample and the actual reported content corresponding to each training sample as input and the predicted reported content of each training sample as output, the document processing model to be trained is iteratively trained until the model converges to generate a preset document processing model.

[0079] Indicatively, in each iterative training, a training sample is input into the document processing model, so that the document processing model extracts the text paragraph with the highest degree of relevance to the word segmentation list sample from the procedure document sample according to the word features and text features in the training sample, and generates a reported content prediction result corresponding to the preset prompt template sample according to the preset prompt template and the text paragraph;

[0080] The reported content prediction results are compared with the actual reported content results, and the network parameters of the document processing model are adjusted according to the comparison results.

[0081] Specifically, when processing documents through a trained document processing model, it is first necessary to obtain a preset prompt template for guiding the model to output preset content, and input the preset prompt template Prompt into the model so that the model summarizes and outputs the reporting unit, reporting time, reporting method and reporting template based on the template. The preset prompt template can make the model more targeted and accurate when extracting and generating information.

[0082] Schematically, the document processing model of the present invention includes: an input layer, a feature extraction layer, an attention mechanism layer, a fusion layer, and an output layer;

[0083] The input layer is used to receive the abnormal information, target procedure documents and preset prompt templates input by the user, and convert the text information into a format that can be processed by the model, such as converting the text into a word vector. This ensures that input data in different forms can be processed uniformly, providing a basis for subsequent feature extraction and model operation.

[0084] The feature extraction layer is used to extract the features of each word in the word segmentation list, such as word frequency, part of speech, semantic vector, etc. It is also possible to use structures such as recurrent neural networks (RNN), long short-term memory networks (LSTM) or Transformer to capture the context information of each text paragraph in the target procedure document and generate corresponding text features. The feature extraction layer can extract rich text features, providing important information input for the subsequent attention mechanism layer.

[0085] The attention mechanism layer is used to calculate the cosine similarity between each word feature and each text feature;

[0086] For each text feature, based on the activation function, the cosine similarity between the text feature and each word feature is converted into a probability distribution, and the corresponding attention weight is assigned to the text feature according to each probability distribution; wherein the attention weight is used to characterize the correlation between the text feature and the word list;

[0087] According to each attention weight, the text paragraph corresponding to the text feature with the highest attention weight is determined.

[0088] The attention mechanism layer of the present invention can calculate the correlation or similarity between each word feature and each text feature, and then extract the text paragraph with the highest correlation with the word segmentation list from the target procedure document. The attention mechanism layer can extract the text paragraph with the highest correlation with the word segmentation list (i.e., the abnormal information input by the user), thereby ensuring that the information processed subsequently is accurate and targeted. That is, the attention mechanism layer can focus on the text paragraph related to the abnormal information, reduce the interference of noise information, and improve the pertinence and accuracy of the model.

[0089] The preset prompt template defines the basic structure and requirements of the reported content, and the fusion layer can be used to splice the first text vectors corresponding to the preset prompt template and the second text vectors corresponding to the text paragraph in the vector space to generate a plurality of spliced ​​third text vectors; wherein the first text vector corresponding to the preset prompt template includes: a unit text vector, a time text vector, a method text vector and a template text vector; the second text vector corresponding to the text paragraph includes: a reporting text vector, a reporting time text vector, a reporting method text vector and a reporting template text vector;

[0090] According to each of the spliced ​​third text vectors, report content corresponding to the preset prompt template is generated.

[0091] The output layer is used to output the generated report content in a standardized format for subsequent processing or display.

[0092] The fusion layer of the present invention can combine the structural requirements of the preset prompt template with the specific content of the text paragraph, and generate a spliced ​​third text vector by vector splicing. These vectors contain both the basic structure and requirements in the preset prompt template and the text content most relevant to the abnormal information extracted from the target procedure document. Therefore, the fusion layer ensures that the generated report content is both in compliance with the specification and can accurately reflect the abnormal information input by the user.

[0093] Therefore, by processing the target procedure document through the preset document processing model, the content related to the power grid abnormal information can be automatically extracted, so that the corresponding abnormal reporting information can be generated, thereby improving the efficiency and accuracy of abnormal information reporting.

[0094] For step S4, in a preferred embodiment, the abnormal text includes: the location where the abnormality occurred, the time when the abnormality occurred, the abnormal device type and the abnormal state; the abnormal state is used to characterize the device state when the abnormal situation occurs; the reporting template includes: a time filling part, a unit filling part, a device type filling part and an abnormal situation filling part;

[0095] The generating, according to each abnormal text and the reported content, abnormal reporting information corresponding to the abnormal information of the power grid includes:

[0096] Fill the abnormal text of the abnormal occurrence time, the reporting unit in the report content, the abnormal text of the abnormal device type and the abnormal text of the abnormal state into the time filling part, the unit filling part, the device type filling part and the abnormal situation filling part of the report template respectively to generate a filled report template;

[0097] Exception reporting information corresponding to the power grid exception information is generated according to the filled reporting template, the exception text of the location where the exception occurred, the reporting time in the reporting content, and the reporting method in the reporting content.

[0098] Specifically, the following is a specific example for illustrating how to generate abnormal report information corresponding to the power grid abnormal information according to the abnormal text and the report content:

[0099] Abnormal text example: Abnormal location: XX Road Substation, XX District, XX City; Abnormal time: 14:30, May 15, 2023; Abnormal equipment type: transformer; Abnormal status: overheating, the temperature exceeds the set threshold;

[0100] Example of reporting content (extracted from the target procedure document): Reporting unit: Operation and Maintenance Department of XX Power Company; Reporting time: 14:35, May 15, 2023; Reporting method: telephone report + system entry;

[0101] Report template: Time filling part: [time]; Unit filling part: [unit]; Equipment type filling part: [equipment type]; Abnormal situation filling part: [abnormal situation];

[0102] Indicatively, the template may also include other fixed texts, such as "Dear leader, the following abnormal situation of the power grid is reported:" etc.;

[0103] When generating a filled report template, the time of the abnormality "May 15, 2023 14:30" can be filled in the time filling part; the reporting unit "XX Power Company Operation and Maintenance Department" can be filled in the unit filling part; the abnormal equipment type "transformer" can be filled in the equipment type filling part; the abnormal state "overheating, temperature exceeds the set threshold" can be filled in the abnormal situation filling part, and the filled report template can be finally generated as follows:

[0104] Dear leader, we now report the following abnormal situation of the power grid:

[0105] Time: 14:30, May 15, 2023;

[0106] Unit: Operation and Maintenance Department of XX Power Company;

[0107] Equipment type: Transformer;

[0108] Abnormal situation: overheating, the temperature exceeds the set threshold.

[0109] Furthermore, the process of generating the final report information based on the above filled report template and other information is as follows:

[0110] On the basis of the filled reporting template, add the location where the abnormality occurred "XX Road Substation, XX District, XX City"; add the reporting time extracted from the reporting content "May 15, 2023 14:35"; indicatively, although this time is close to the time when the abnormality occurred, this is the actual reporting time; add the reporting method "telephone report + system entry".

[0111] The final generated exception reporting information is as follows:

[0112] Dear leader, we now report the following abnormal situation of the power grid:

[0113] Time: 14:30, May 15, 2023;

[0114] Location: Substation, XX Road, XX District, XX City;

[0115] Unit: Operation and Maintenance Department of XX Power Company;

[0116] Reporting time: 14:35, May 15, 2023;

[0117] Reporting method: telephone report + system entry;

[0118] Equipment type: Transformer;

[0119] Abnormal situation: overheating, the temperature exceeds the set threshold.

[0120] Therefore, the filling method of the present invention is based on the direct extraction of the abnormal text and the reported content, which reduces the possibility of human input errors. At the same time, the accuracy of the information is ensured by a clear filling correspondence (such as filling the abnormal occurrence time into the time filling part). By using the reporting template extracted from the procedure document, the format and content of the abnormal reporting information are ensured to be consistent and standardized with the procedure document, reducing misunderstandings or delays caused by inconsistent formats or incomplete information.

[0121] By automating the filling method and generation process, the preparation time for reporting information can be greatly shortened. At the same time, the standardized format also simplifies the processing flow of the recipient, improves work efficiency, and ensures that the abnormal situation and abnormal information can be accurately reported to the corresponding unit when the abnormal report is made based on the abnormal report information, which helps to share information and work together between different departments and teams, thereby improving the overall ability to deal with power grid abnormalities.

[0122] In a preferred embodiment, the present invention can adopt RAG technology, that is, retrieval enhanced generation technology. Through efficient indexing and vectorization of procedure documents, it can quickly locate the applicable procedure documents when the user inputs an abnormal situation. By using RAG technology to retrieve relevant procedure documents in real time, it can automatically match abnormal events with corresponding processing specifications to ensure timely and accurate information.

[0123] In a preferred embodiment, the document processing model of the present invention is a large language model (LLM). By combining the natural language processing capabilities of the LLM, the present invention can deeply understand the context of the exception information input by the user, and accurately extract key information from the retrieved target procedure document, so that the generation of reported information is more targeted and accurate, and effectively ensures the consistency and integrity of the exception reporting information.

[0124] Therefore, the embodiment of the present invention realizes rapid retrieval and intelligent analysis of information through the combination of RAG technology and LLM, forms an integrated solution, and optimizes the information processing flow.

[0125] like Figure 2 As shown, based on the above-mentioned various embodiments of the method for generating abnormal reporting information of the distribution network, the present invention provides a corresponding device embodiment;

[0126] An embodiment of the present invention provides an abnormality reporting information generating device for a distribution network, comprising: an abnormality information acquiring module, a similarity calculating module, a reporting content extracting module and an abnormality reporting information generating module;

[0127] The abnormal information acquisition module is used to acquire the abnormal power grid information input by the user; wherein the abnormal power grid information includes: a number of abnormal texts for describing the abnormal phenomenon;

[0128] The similarity calculation module is used to convert the power grid abnormality information into a semantic vector list, and calculate the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document, wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities;

[0129] The reporting content extraction module is used to take the procedure document with the highest similarity as the target procedure document, and then extract the corresponding reporting content in the target procedure document; wherein the reporting content includes: reporting unit, reporting time and reporting method;

[0130] The abnormal reporting information generating module is used to generate abnormal reporting information corresponding to the abnormal information of the power grid according to each abnormal text and reporting content.

[0131] It should be noted that the device embodiments described above are merely schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without paying creative labor.

[0132] Those skilled in the art can clearly understand that, for the sake of convenience and simplicity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0133] Based on the above-mentioned various embodiments of the method for generating abnormal reporting information of the distribution network, the present invention provides corresponding embodiments of terminal equipment items.

[0134] An embodiment of the present invention 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, wherein when the processor executes the computer program, it implements a method for generating abnormal reporting information of a distribution network as described in any method embodiment of the present invention.

[0135] The terminal device may be a computing terminal device such as a desktop computer, a notebook, a palm computer, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0136] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0137] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device or other volatile solid-state storage device.

[0138] Based on the above-mentioned various embodiments of the method for generating abnormal reporting information of the distribution network, the present invention provides a corresponding storage medium item embodiment.

[0139] An embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for generating abnormal reporting information of a distribution network as described in any method embodiment of the present invention.

[0140] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0141] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for generating abnormal reporting information of a distribution network, characterized in that: include: Acquire power grid abnormality information input by a user; wherein the power grid abnormality information includes: a number of abnormal texts for describing abnormal phenomena; The power grid abnormality information is converted into a semantic vector list, and the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document is calculated; wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities; The procedure document with the highest similarity is used as the target procedure document, and then the corresponding reporting content in the target procedure document is extracted; wherein the reporting content includes: reporting unit, reporting time and reporting method; According to each abnormal text and the report content, the abnormal report information corresponding to the abnormal information of the power grid is generated.

2. A method for generating abnormal reporting information of a distribution network according to claim 1, characterized in that: The calculating the similarity between each procedure document and the semantic vector list according to each semantic vector in the semantic vector list and the index vector of each procedure document comprises: For each procedure document, respectively, calculating the similarity between each semantic vector in the semantic vector list and the index vector of the procedure document to obtain the similarity between each semantic vector and the procedure document; For each procedure document, the average similarity between the procedure document and the semantic vector list is obtained based on the similarity between each semantic vector in the semantic vector list and the procedure document, and the total number of semantic vectors in the semantic vector list; the average similarity is used as the similarity corresponding to the procedure document and the semantic vector list.

3. A method for generating abnormal reporting information of a distribution network as claimed in claim 2, characterized in that: The reported content also includes a reporting template; The extracting the corresponding reporting content in the target procedure document includes: Obtaining a preset prompt template for guiding the model to output preset content; wherein the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template; Perform a word segmentation operation on each abnormal text in the power grid abnormal information, and generate a word segmentation list corresponding to the power grid abnormal information according to each word segmented; The preset prompt template, the target procedure document and the word segmentation list are input into a preset document processing model, so that the document processing model extracts word features of each word in the word segmentation list, captures context information of each text paragraph in the target procedure document, and generates corresponding text features according to the context information corresponding to each text paragraph; extracts the text paragraph with the highest degree of relevance to the word segmentation list from the target procedure document according to each word feature and each text feature, and generates reporting content corresponding to the preset prompt template according to the preset prompt template and the text paragraph; The generation of the preset document processing model includes: Generate training samples according to the sample of the regulation document, the sample of the word segmentation list corresponding to the sample of the power grid abnormal information, and the sample of the preset prompt template; wherein the sample of the preset prompt template includes: reporting unit, reporting time, reporting method and reporting template; Taking each training sample and the actual reported content corresponding to each training sample as input and the predicted reported content of each training sample as output, the document processing model to be trained is iteratively trained until the model converges to generate a preset document processing model.

4. A method for generating abnormal reporting information of a distribution network as claimed in claim 3, characterized in that: In each iterative training, a training sample is input into the document processing model, so that the document processing model extracts the text paragraph with the highest degree of relevance to the word segmentation list sample from the procedure document sample according to the word features and text features in the training sample, and generates a reported content prediction result corresponding to the preset prompt template sample according to the preset prompt template and the text paragraph; The reported content prediction results are compared with the actual reported content results, and the network parameters of the document processing model are adjusted according to the comparison results.

5. A method for generating abnormal reporting information of a distribution network as claimed in claim 4, characterized in that: The document processing model includes: an attention mechanism layer; The extracting, from the target procedure document, the text paragraph with the highest degree of relevance to the word segmentation list according to each word feature and each text feature, comprises: The attention mechanism layer is used to calculate the cosine similarity between each word feature and each text feature; For each text feature, based on the activation function, the cosine similarity between the text feature and each word feature is converted into a probability distribution, and the corresponding attention weight is assigned to the text feature according to each probability distribution; wherein the attention weight is used to characterize the correlation between the text feature and the word list; According to each attention weight, the text paragraph corresponding to the text feature with the highest attention weight is determined.

6. A method for generating abnormal reporting information of a distribution network as claimed in claim 5, characterized in that: The document processing model includes: a fusion layer; The generating, according to the preset prompt template and the text paragraph, the reporting content corresponding to the preset prompt template comprises: The fusion layer is used to splice each first text vector corresponding to the preset prompt template and each second text vector corresponding to the text paragraph in the vector space to generate a plurality of spliced ​​third text vectors; wherein the first text vector corresponding to the preset prompt template includes: a unit text vector, a time text vector, a method text vector and a template text vector; the second text vector corresponding to the text paragraph includes: a reporting text vector, a reporting time text vector, a reporting method text vector and a reporting template text vector; According to each of the spliced ​​third text vectors, report content corresponding to the preset prompt template is generated.

7. A method for generating abnormal reporting information of a distribution network as claimed in claim 3, characterized in that: The abnormal text includes: the location where the abnormality occurred, the time when the abnormality occurred, the abnormal device type and the abnormal state; the abnormal state is used to characterize the device state when the abnormal situation occurs; the reporting template includes: a time filling part, a unit filling part, a device type filling part and an abnormal situation filling part; The generating, according to each abnormal text and the reported content, abnormal reporting information corresponding to the abnormal information of the power grid includes: Fill the abnormal text of the abnormal occurrence time, the reporting unit in the report content, the abnormal text of the abnormal device type and the abnormal text of the abnormal state into the time filling part, the unit filling part, the device type filling part and the abnormal situation filling part of the report template respectively to generate a filled report template; Exception reporting information corresponding to the power grid exception information is generated according to the filled reporting template, the exception text of the location where the exception occurred, the reporting time in the reporting content, and the reporting method in the reporting content.

8. A device for generating abnormality reporting information of a distribution network, characterized in that: include: Abnormal information acquisition module, similarity calculation module, reporting content extraction module and abnormal reporting information generation module; The abnormal information acquisition module is used to acquire the abnormal power grid information input by the user; wherein the abnormal power grid information includes: a number of abnormal texts for describing the abnormal phenomenon; The similarity calculation module is used to convert the power grid abnormality information into a semantic vector list, and calculate the similarity between each semantic vector in the semantic vector list and the index vector of each procedure document, wherein each semantic vector in the semantic vector list corresponds to an abnormal text; and the procedure document is used to characterize the corresponding reporting rules when handling power grid abnormalities; The reporting content extraction module is used to take the procedure document with the highest similarity as the target procedure document, and then extract the corresponding reporting content in the target procedure document; wherein the reporting content includes: reporting unit, reporting time and reporting method; The abnormal reporting information generating module is used to generate abnormal reporting information corresponding to the abnormal information of the power grid according to each abnormal text and reporting content.

9. A terminal device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements a method for generating abnormal reporting information of a distribution network as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the method for generating abnormal reporting information of a distribution network as described in any one of claims 1 to 7.

Images