Line loss report generation method and device of power equipment and electronic equipment

Abstract

The invention discloses a line loss report generation method and device of power equipment and electronic equipment. The method comprises the following steps: acquiring a line loss report generation demand of power equipment; performing semantic recognition on the line loss report generation demand to obtain a target semantic vector of the line loss report generation demand; based on the target semantic vector and a first knowledge base, a target report template is determined, and a plurality of report templates and semantic vectors corresponding to the report templates are stored in the first knowledge base; generating an initial line loss report of the power equipment based on the target report template; semantic recognition is conducted on the line loss report optimization requirement, an optimization target of the line loss report optimization requirement is obtained, and the optimization target comprises content expansion writing or content abbreviation; and based on the optimization target, optimizing the initial line loss report to obtain a target line loss report. According to the method and the device, the technical problem of inaccurate line loss report generation result of the power equipment in the prior art is solved.

Description

Technical Field

[0001] This application relates to the field of power systems, and more specifically, to a method, apparatus, and electronic device for generating line loss reports for power equipment. Background Technology

[0002] Line loss analysis reports for power equipment are crucial for assessing power grid operating efficiency and optimizing resource allocation. As power grids continue to expand and business data becomes increasingly complex, traditional manual methods of writing line loss analysis reports are no longer sufficient to meet the demands of efficient management, leading the industry to gradually shift towards automated report generation technologies.

[0003] Related technologies employ automated report generation methods based on fixed templates. These methods generate line loss analysis reports through pre-defined rules and simple data entry. For example, a report generation tool imports basic data into a fixed template to quickly generate a preliminary report framework. However, this template-based automated report generation method can only fill in data according to preset rules and cannot accurately understand the diverse and personalized line loss report requirements of users, making it difficult to flexibly generate report content that meets user expectations. Therefore, these technologies suffer from inaccurate line loss report generation results for power equipment.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for generating line loss reports for power equipment, in order to at least solve the technical problem of inaccurate line loss report generation results for power equipment in related technologies.

[0006] According to one aspect of the embodiments of this application, a method for generating line loss reports for power equipment is provided, comprising: obtaining line loss report generation requirements for power equipment; performing semantic recognition on the line loss report generation requirements to obtain a target semantic vector for the line loss report generation requirements, wherein the target semantic vector is used to describe the core content and key information of the line loss report generation requirements; determining a target report template based on the target semantic vector and a first knowledge base, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively; generating an initial line loss report for power equipment based on the target report template; performing semantic recognition on the line loss report optimization requirements to obtain optimization objectives for the line loss report optimization requirements, wherein the optimization objectives include content expansion or content abbreviation; and optimizing the initial line loss report based on the optimization objectives to obtain a target line loss report.

[0007] According to another aspect of the embodiments of this application, a line loss report generation device for power equipment is provided, comprising: a line loss report generation requirement acquisition module, used to acquire line loss report generation requirements for power equipment; a first semantic recognition module, used to perform semantic recognition on the line loss report generation requirements to obtain a target semantic vector of the line loss report generation requirements, wherein the target semantic vector is used to describe the core content and key information of the line loss report generation requirements; a target report template determination module, used to determine a target report template based on the target semantic vector and a first knowledge base, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively; an initial line loss report generation module, used to generate an initial line loss report for power equipment based on the target report template; a second semantic recognition module, used to perform semantic recognition on the line loss report optimization requirements to obtain optimization targets for the line loss report optimization requirements, wherein the optimization targets include content expansion or content abbreviation; and an optimization module, used to optimize the initial line loss report based on the optimization targets to obtain a target line loss report.

[0008] According to another aspect of the embodiments of this application, a non-volatile storage medium is provided, which stores a plurality of instructions adapted for a method for generating line loss reports for power equipment, any one of which is loaded by a processor.

[0009] According to another aspect of the embodiments of this application, an electronic device is provided, including: one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement any one of the following: a method for generating line loss reports for power equipment.

[0010] According to another aspect of the embodiments of this application, a computer program product is provided, which, when executed on a data processing device, is adapted to perform the steps of a method for generating line loss reports for power equipment.

[0011] In this embodiment, the following steps are taken: First, the line loss report generation requirement for power equipment is obtained. Then, semantic recognition is performed on the line loss report generation requirement to obtain a target semantic vector, which describes the core content and key information of the line loss report generation requirement. Based on the target semantic vector and a first knowledge base, a target report template is determined, whereby the first knowledge base stores multiple report templates and their corresponding semantic vectors. Based on the target report template, an initial line loss report for the power equipment is generated. Next, semantic recognition is performed on the line loss report optimization requirement to obtain the optimization objective, whereby the optimization objective includes content expansion or abbreviation. Based on the optimization objective, the initial line loss report is optimized to obtain the target line loss report. This achieves the technical effect of improving the accuracy of the generated line loss report, thereby solving the technical problem of inaccurate line loss report generation results in related technologies. Attached Figure Description

[0012] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0013] Figure 1 This is a flowchart of a method for generating line loss reports for power equipment according to an embodiment of this application;

[0014] Figure 2 This is a schematic diagram of an optional method for generating line loss reports for power equipment according to an embodiment of this application;

[0015] Figure 3 This is a schematic diagram of an optional power equipment line loss report generation device provided according to an embodiment of this application. Detailed Implementation

[0016] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0017] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0018] According to an embodiment of this application, a method embodiment for generating line loss reports for power equipment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] Figure 1 This is a flowchart of a method for generating line loss reports for power equipment according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0020] Step S102: Obtain the requirements for generating line loss reports for power equipment;

[0021] It's understandable that they want to obtain user requests for line loss reports on power equipment.

[0022] Step S104: Semantic recognition is performed on the line loss report generation requirement to obtain the target semantic vector of the line loss report generation requirement. The target semantic vector is used to describe the core content and key information of the line loss report generation requirement.

[0023] It is understandable that semantic recognition and analysis of user requests for generating line loss reports for power equipment are performed to obtain target semantic vectors that describe the core content and key information of these requests. Obtaining these target semantic vectors through semantic recognition lays the data foundation for subsequent generation of target line loss reports for power equipment based on these vectors.

[0024] In one optional embodiment, semantic recognition is performed on the line loss report generation requirement to obtain a target semantic vector for the line loss report generation requirement. This includes: preprocessing the line loss report generation requirement to obtain a standard generation requirement; performing semantic recognition on the standard generation requirement to determine an initial semantic vector for the standard generation requirement; and fusing the initial semantic vector and a knowledge vector using a gating mechanism to obtain a target semantic vector. Here, the knowledge vector refers to the structured information of the line loss report generation standard for power equipment, and the gating mechanism is used to control the fusion process between the target semantic vector and the knowledge vector.

[0025] The target semantic vector for generating line loss reports can be understood as follows: First, the line loss report generation requirements are preprocessed, including removing special characters and stop words, and word segmentation, to obtain the standard generation requirements for power equipment. Second, semantic recognition is performed on the above standard generation requirements to obtain the initial semantic vector of the standard generation requirements. Next, the initial semantic vector and the knowledge vector are fused using a gating mechanism to obtain the target semantic vector of the standard generation requirements. The aforementioned knowledge vector refers to the structured information of the line loss report generation standards for power equipment, and the gating mechanism is used to control the fusion process between the target semantic vector and the knowledge vector. The fusion of knowledge vectors ensures that the generated line loss report not only meets the user's personalized needs but also deeply integrates professional knowledge from the power industry, including historical data, industry standards, and expert experience, thereby ensuring the comprehensiveness, professionalism, and depth of the report content and meeting the high standards required for line loss reports in power system operations.

[0026] Optionally, the Guangming Big Data Model can be used to perform semantic recognition on the line loss report generation request to obtain the target semantic vector. The user inputs the request for relevant chapters of the report (i.e., the line loss report generation request), and the Guangming Big Data Model performs semantic recognition on the natural language text of the input line loss report generation request, analyzes the core content and key information of the user's request, and transforms it into a semantic vector (i.e., the target semantic vector) that can be processed by a computer.

[0027] Optionally, the Guangming Big Data model leverages deep learning and natural language processing technologies to construct a dedicated knowledge graph covering power industry terminology and business rules. When receiving user requests, the model can intelligently parse ambiguous expressions. For example, it can automatically transform "analyze recent line loss anomalies" into "statistically analyze the line loss rate fluctuations over the past 30 days on a daily / weekly basis, label outliers, and associate them with the corresponding regional equipment status." Compared to keyword matching technology based on fixed templates, this model can capture implicit information in user requests. When generating line loss reports, it automatically associates historical similar case data and accurately locates core demands through semantic similarity calculations. This improves the match between report content and user expectations by more than 40%, effectively solving the problem of rework in line loss reports caused by misunderstandings of requirements.

[0028] Optionally, after a user submits a request to generate a line loss report through the interactive interface, the original text of the request is first preprocessed to remove special characters and stop words and standardize the text format. Then, the cleaned request text (i.e., the standard generated request) is transmitted to the Guangming Big Data Model processing module. An optimized Transformer architecture is used, adding a power cross-modal adaptation layer, and deep association modeling is achieved through a hybrid attention mechanism. This model primarily uses this architecture to process multimodal data from the power industry, enabling functions such as power knowledge memorization and understanding, and multimodal fusion analysis. The specific process includes: tagging nouns, verbs, etc., using a part-of-speech tagging model; clarifying grammatical relationships between words using dependency parsing; extracting key business requirements based on an attention mechanism, such as "comparative analysis of 10kV line loss rates over the past three years"; and finally, encoding semantic information into a high-dimensional semantic vector (i.e., the target semantic vector) to accurately identify user requirements and provide a data foundation for subsequent processing.

[0029] Optionally, for specialized terms in the power industry (such as "ultra-high voltage converter valve" and "bus load forecasting"), a domain-enhanced word segmentation algorithm can be used. This algorithm generates a dedicated vocabulary using a pre-trained power corpus, addressing the problem of inaccurate segmentation of power terms by the word segmenter. The aforementioned domain-enhanced word segmentation algorithm is a natural language processing technique specifically optimized for domain-specific corpora, primarily used to accurately segment continuous text into meaningful lexical units.

[0030] Optionally, the pre-trained power corpus is a large-scale collection of text data customized for the power industry to train deep learning models (especially natural language processing models). The corpus contains various types of text, including professional documents, technical reports, operation manuals, standards and specifications, expert discussions, and meeting minutes, aiming to enable the model to learn to understand and process the specific language, terminology, and business logic of the power industry.

[0031] Optionally, by combining the power industry knowledge graph, domain-customized tags (such as "equipment type", "failure mode" and "topology node") can be introduced at the annotation layer. Through weakly supervised learning, power entities in the text can be automatically annotated to solve the problem of insufficient coverage of industry entities.

[0032] Alternatively, by optimizing the Transformer architecture, employing stacked multi-head self-attention layers and feedforward neural networks, and using dynamic position encoding techniques (combining grid topology distance and text semantic distance), the ability of the Guangming large model to capture the spatial-temporal-semantic triple correlation in power data can be enhanced.

[0033] Optionally, the knowledge vector can include not only structured information about power equipment line loss report generation standards, but also structured information from power regulations, equipment manuals, etc. Encoding structured knowledge such as power regulations and equipment manuals into trainable knowledge vectors and dynamically fusing them into the model's hidden layer representation during semantic extraction (i.e., fusing the initial semantic vector and knowledge vector using a gating mechanism to obtain the target semantic vector) enhances the memorization and reasoning capabilities of professional knowledge.

[0034] Step S106: Based on the target semantic vector and the first knowledge base, determine the target report template, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively;

[0035] It is understandable that, based on the target semantic vector generated according to the standard, the target report template is determined from a first knowledge base that stores multiple report templates and their corresponding semantic vectors. Through precise template matching, the problem of inaccurate template selection can be effectively solved, thereby improving the accuracy of the generated power equipment line loss reports and user satisfaction.

[0036] In one optional embodiment, determining a target report template based on a target semantic vector and a first knowledge base includes: determining multiple similarity results based on the target semantic vector and semantic vectors corresponding to multiple report templates, wherein the multiple similarity results correspond one-to-one with the multiple report templates, and the similarity results are used to describe the degree of similarity between the semantic vectors corresponding to the report templates in the first knowledge base and the target semantic vector; selecting multiple initial report templates from the first knowledge base based on the multiple similarity results, a preset similarity threshold, and a preset number of templates; and integrating the multiple initial report templates to obtain a target report template, wherein the integration process includes at least one of the following: merging duplicate chapters, sorting chapters, and supplementing chapters.

[0037] It is understood that a similarity analysis is performed on the target semantic vector of the standard generation requirement and the semantic vectors corresponding to multiple report templates in the first knowledge base, respectively, to obtain multiple similarity results corresponding one-to-one with the multiple report templates. These similarity results describe the degree of similarity between the semantic vectors corresponding to the report templates in the first knowledge base and the target semantic vector. From these multiple similarity results, a first set of report templates with similarities greater than a preset threshold is determined. The report templates in this first set are then sorted from highest to lowest similarity, and the report templates with the highest similarity ranking are selected from the first set according to a preset number of templates, serving as multiple initial report templates. These multiple initial report templates are then integrated, including merging duplicate chapters, sorting chapters, and supplementing chapters, to obtain the target report template. This template integration process, especially the chapter sorting and supplementation, makes the generated line loss report more systematic and well-organized, not only facilitating user understanding and review but also meeting the stringent requirements of the power industry for the content structure of line loss reports, thus improving the professionalism and accuracy of the line loss reports.

[0038] Optionally, merging duplicate chapters is to avoid redundancy in the generated line loss report and improve its conciseness; chapter sorting refers to rearranging the chapter order according to the priority of business logic to ensure the logicality and readability of the line loss report content; chapter supplementation refers to automatically adding essential chapters required by industry standards, such as line loss analysis methods and data source descriptions, to improve the structure of the line loss report.

[0039] Optionally, the above business logic may include, but is not limited to, the importance and relevance of chapter content, the level and depth of analysis, industry standards and norms, and the user's cognitive process. First, determine which chapters are most crucial for the overall understanding of the line loss report; these chapters will be placed in more prominent or earlier positions. Second, start with a high-level overview and gradually delve into specific content analysis and detailed discussions, helping readers understand the line loss status of power equipment from a macro to micro perspective. Next, for certain chapters, according to power industry standards, they must appear in specific positions within the line loss report; for example, the summary is usually placed at the beginning of the line loss report for quick comprehension of the core content. Finally, based on the user's logical thinking process when reading the line loss report, arrange the chapters in an easy-to-understand and follow order; for example, present statistical results first, then conduct causal analysis, and finally provide optimization suggestions. This allows for a gradual understanding of the problem and the acquisition of solutions.

[0040] Optionally, the target report template can be obtained as follows: Construct a knowledge base related to line loss reports (i.e., the first knowledge base), which stores multiple report templates and their corresponding semantic vectors. Based on the target semantic vector obtained from semantic recognition, the Guangming Big Data Model searches the first knowledge base, recalls multiple initial report templates with high matching degree to user needs, and integrates the recalled initial report templates to form a preliminary report framework (i.e., the target report template).

[0041] Optionally, the Guangming Big Data Model processing module matches the target semantic vector of the user's needs with the semantic vectors of 300+ report templates stored in the first knowledge base. During the retrieval process, an improved cosine similarity algorithm can be used, introducing TF-IDF (Term Frequency-Inverse Document Frequency) weighting factors to enhance keyword influence. The similarity Sim(A,B) between the target semantic vector and the semantic vector of any template in the first knowledge base can be determined as follows:

[0042]

[0043] Where A represents the target semantic vector, and B represents the semantic vector of any template, A i Let B represent the word embedding vector of the i-th word in A. i This represents the word embedding vector of the i-th word in B. This represents the TF-IDF weight of the i-th word in A. Let represent the TF-IDF weight of the i-th word in B, where n represents the total number of words.

[0044] Optionally, TF-IDF is used to evaluate the importance of a word to a document in a document set. TF-IDF is calculated based on two components: Term Frequency (TF) and Inverse Document Frequency (IDF). Term Frequency refers to the frequency with which a word appears in a document, reflecting its importance within that document. If a word appears frequently in a document, its contribution to the document is likely significant; conversely, if a word appears infrequently, its contribution is likely small. Inverse Document Frequency measures the general importance of a word, i.e., how rare a word is in the document set. If a word appears in many documents, its IDF value will be low, indicating that the word is likely common and has little distinguishing effect on documents; conversely, if a word appears only in a few documents, its IDF value will be high, indicating that the word plays a significant role in distinguishing these documents.

[0045] Optionally, a similarity threshold (i.e., a preset similarity threshold) is set to 0.7, and the top 5 highly matched initial report templates (i.e., the preset number of templates) are integrated. For the above 5 initial report templates, duplicate chapters can be merged using NLP (Natural Language Processing) text alignment technology (i.e., duplicate chapter merging), the content order can be rearranged based on business logic priority (i.e., chapter sorting), and essential chapters required by industry standards can be automatically supplemented (i.e., chapter supplementation). Finally, a preliminary report framework including modules such as table of contents, summary, line loss data statistics, cause analysis, and optimization suggestions can be constructed.

[0046] Step S108: Based on the target report template, generate an initial line loss report for the power equipment;

[0047] It is understandable that data is populated into the target report template to generate an initial line loss report for power equipment.

[0048] In one optional embodiment, generating an initial line loss report for power equipment based on a target report template includes: determining a data acquisition instruction based on the target report template; acquiring target data from a database based on the data acquisition instruction; and filling the target data into the target report template to obtain the initial line loss report.

[0049] Understandably, based on the target report template, the data type to be populated is determined, and a data retrieval instruction is generated. According to the data retrieval instruction, the target data to be populated is retrieved from the database. The target data is then populated into the target report template to obtain the initial line loss report. By automating data retrieval and population, the problems of low data integration efficiency, frequent human errors, and inconsistent line loss report standards during the line loss report generation process are avoided, achieving an intelligent upgrade in line loss report generation and improving the accuracy of the generated line loss reports.

[0050] Optionally, the corresponding line loss data, equipment operation data, and other business data (i.e., target data) can be retrieved from the data platform's business space according to the data requirements of each chapter of the report via the data platform API (Application Programming Interface). The retrieved target data is then populated into the preliminary report framework to generate a complete unit line loss business analysis report (i.e., the initial line loss report). The data platform API generates data retrieval instructions based on the data requirements of each chapter in the preliminary report framework and sends them to the power system business database. The database filters and returns the corresponding business data (i.e., target data) according to the instructions. After receiving the data, the data platform interaction module transmits it to the user interaction module, populates it into the preliminary report framework, and generates the initial line loss report.

[0051] Optionally, based on the API interface of the data platform, data source response can be achieved in seconds. By establishing indexes for 10 types of core business data, such as equipment ledgers, metering data, and load curves, the system automatically matches the optimal data retrieval path according to demand characteristics. For example, when generating a transformer area line loss report, the system retrieves electricity consumption data, production system equipment parameters, and GIS (Geographic Information System) geographic information in parallel. Through data cleaning algorithms, missing values ​​and outliers are automatically processed, improving integration efficiency by 80% compared to manual configuration. This reduces the generation time of a regular line loss report from 4 hours to less than 15 minutes, meeting the timeliness requirements of real-time power grid monitoring and decision support.

[0052] Step S110: Perform semantic recognition on the line loss report optimization requirements to obtain the optimization objectives of the line loss report optimization requirements, wherein the optimization objectives include content expansion or content abbreviation;

[0053] Understandably, if it's necessary to optimize the generated initial line loss report, the first step is to obtain the user's optimization requirements for the initial line loss report, and then perform semantic recognition on these requirements to determine the optimization goals, including content expansion or abbreviation. By semantically recognizing the optimization requirements and optimizing the initial line loss report, the accuracy of the generated target line loss report and user satisfaction can be further improved.

[0054] Step S112: Based on the optimization objective, optimize the initial line loss report to obtain the target line loss report.

[0055] It is understandable that the initial line loss report is optimized according to the optimization objectives to obtain the target line loss report for power equipment. By personalizing the initial line loss report, the information needs under different scenarios and preferences can be met, thereby improving the accuracy of the target line loss report and user satisfaction.

[0056] In one optional embodiment, when the optimization objective includes content expansion, the initial line loss report is optimized based on the optimization objective to obtain a target line loss report. This includes: obtaining expanded content of the initial line loss report from a second knowledge base, wherein the content of the second knowledge base includes at least: historical line loss cases, comparative analysis of power equipment parameters, and expert recommendations; the expanded content includes at least: historical line loss cases of equipment similar to the power equipment, comparative analysis of power equipment parameters, and expert recommendations for the power equipment; and expanding the initial line loss report based on the expanded content to obtain the target line loss report.

[0057] Understandably, when the optimization objective includes content expansion, the initial line loss report is optimized using the following method. Based on the line loss report optimization requirements, expanded content for the initial line loss report is obtained from a second knowledge base. This second knowledge base includes at least: historical line loss cases of similar power equipment, comparative analysis of power equipment parameters, and expert recommendations for the power equipment. The initial line loss report is then expanded based on this expanded content to obtain the target line loss report for the power equipment. By expanding the content of the initial line loss report, the problems of its limited content and lack of depth are addressed, thereby meeting the high requirements for the depth and breadth of information in the target line loss report.

[0058] In one optional embodiment, when the optimization objective includes content abbreviation, the initial line loss report is optimized based on the optimization objective to obtain a target line loss report, including: performing redundancy analysis on the initial line loss report to obtain abbreviated content of the initial line loss report; and abbreviating the content of the initial line loss report based on the abbreviated content to obtain the target line loss report.

[0059] Understandably, when the optimization objective includes content abbreviation, the initial line loss report is optimized as follows: Redundancy analysis is performed on the initial line loss report to determine the abbreviations that need to be removed. Based on these abbreviations, the initial line loss report is abbreviated to obtain the target line loss report for the power equipment. By removing redundant information and focusing on key content, the abbreviation function can improve the readability and conciseness of the target line loss report without reducing its accuracy, thereby enhancing user satisfaction.

[0060] Optionally, after the initial line loss report is generated, the user can select the paragraph to be adjusted and choose the expansion or abbreviation function, while inputting the adjustment requirements (i.e., line loss report optimization requirements). After receiving the user's requirements, the Guangming Big Data Model performs secondary semantic analysis on the selected initial line loss report paragraphs, and combines relevant knowledge and data from the second knowledge base to optimize and adjust the paragraph content, thereby enabling the expansion or abbreviation function.

[0061] Optionally, users can submit optimization requests (i.e., line loss report optimization requests) through a combination of paragraph selection, function button clicks, and text box input on the interactive interface. Upon receiving the optimization instruction, the Guangming Big Data Model processing module initiates secondary semantic analysis, combining the industry standard expression library, excellent case library, and acquired business data from the second knowledge base to optimize the content. In the expansion scenario, for the "Line Loss Anomaly Cause Analysis" paragraph, the model can automatically supplement historical fault data of similar equipment, comparative analysis of power equipment parameters, and expert experience suggestions; in the abbreviation scenario, redundancy analysis can be performed through semantic compression technology to remove redundant descriptions while maintaining text coherence. After grammatical validation and logical checks, the optimized template line loss report content is updated in real time to the report display interface, and a version comparison function is provided to facilitate users to review modification records.

[0062] Optionally, line loss reports can be expanded or abbreviated. In abbreviated mode, key information extraction algorithms can automatically identify conclusive statements in the report, compressing a lengthy analysis report into a core summary. Expanded mode, based on the power business knowledge base (i.e., the second knowledge base), supplements key conclusions with multi-dimensional data, such as automatically adding comparisons with the same period, horizontal comparisons with similar transformer areas, and analysis of the impact of meteorological factors for abnormal line loss rates. Users can also perform in-depth optimization of specific chapters through a visual editing interface, saving modified versions in real time and providing version comparison functionality, so that line loss reports can meet the concise reporting needs of upward reporting as well as adapt to in-depth analysis scenarios for technical discussions.

[0063] Through the above steps S102 to S112, the goal is to obtain the target template of the line loss report by semantically recognizing the line loss report generation requirements, thereby generating the initial line loss report of the power equipment, and optimizing the initial line loss report according to the line loss report optimization requirements, so as to finally obtain the target line loss report of the power equipment. This achieves the technical effect of improving the accuracy of the line loss report generation results of the power equipment, and solves the technical problem of inaccurate line loss report generation results of the power equipment in related technologies.

[0064] Based on the above embodiments and optional embodiments, this application proposes an optional implementation method for generating line loss reports for power equipment. This optional implementation method can be understood as an intelligent generation and semantic-driven content optimization method for line loss reports based on the Guangming Big Data Model, used to solve the problems of limited semantic understanding ability, low data integration efficiency, and lack of content optimization function in related technologies. Limited semantic understanding ability means that data is filled in only according to preset rules, which cannot accurately understand the diverse and personalized line loss report requirements of users, and it is difficult to flexibly generate target line loss reports that meet user expectations. Low data integration efficiency means that when obtaining the business data required for line loss reports, data is manually exported and searched, which makes it difficult to achieve fast and accurate data retrieval, resulting in low efficiency in line loss report generation. Lack of content optimization function means that the generated line loss reports do not have the ability to be subsequently optimized, and cannot be expanded or abbreviated according to user needs, making it difficult to meet the usage needs in different scenarios.

[0065] The steps of the intelligent generation of line loss reports and semantic-driven content optimization method based on the Guangming Big Data Model include:

[0066] Step S1, User Request Semantic Recognition. The user inputs the request for generating relevant chapters of the report (i.e., the request for generating a line loss report). The Guangming Big Data Model performs semantic recognition on the natural language text of the input line loss report generation request, analyzes the core content and key information of the user's request, and transforms it into a semantic vector (i.e., the target semantic vector) that can be processed by a computer.

[0067] When a user submits a request to generate a line loss report through the interactive interface, the original text of the request is first preprocessed to remove special characters and stop words and standardize the text format. Then, the cleaned request text (i.e., the standard generated request) is transmitted to the Guangming Big Data Model processing module. An optimized Transformer architecture is used, adding a power cross-modal adaptation layer, and deep correlation modeling is achieved through a hybrid attention mechanism. This model primarily uses this architecture to process multimodal data from the power industry, enabling functions such as power knowledge memorization and understanding, and multimodal fusion analysis. The specific process includes: tagging nouns, verbs, etc., using a part-of-speech tagging model, and using dependency parsing to clarify the grammatical relationships between words; extracting key business requirements based on an attention mechanism, such as "comparative analysis of 10kV line loss rates over the past three years"; and finally, encoding semantic information into a high-dimensional semantic vector (i.e., the target semantic vector) to accurately identify user requirements and provide a data foundation for subsequent processing.

[0068] Step S11, word segmentation and annotation layer.

[0069] Step S111, word segmentation technology. For specialized terms in the power industry (such as "ultra-high voltage converter valve" and "bus load forecasting"), a domain-enhanced word segmentation algorithm is employed. This algorithm generates a dedicated vocabulary using a pre-trained power corpus, addressing the problem of inaccurate segmentation of power terms by the traditional word segmenter. This domain-enhanced word segmentation algorithm is a natural language processing technique specifically optimized for domain-specific corpora, primarily used to accurately segment continuous text into meaningful lexical units.

[0070] Step S112, Enhanced Labeling. By combining the power industry knowledge graph, domain-customized tags (such as "equipment type," "failure mode," and "topology node") are introduced into the labeling layer. Weakly supervised learning is used to automatically label power entities in the text, addressing the issue of insufficient coverage of industry entities.

[0071] Step S12, semantic extraction layer.

[0072] Step S121, Basic Architecture. Optimize the Transformer architecture by using stacked multi-head self-attention layers and feedforward neural networks. Enhance the Guangming large model's ability to capture the spatial-temporal-semantic triple correlations in power data through dynamic position encoding technology (combining grid topology distance and text semantic distance).

[0073] Step S122, Domain Enhancement Mechanism. Structured knowledge such as power regulations and equipment manuals is encoded into trainable knowledge vectors. During semantic extraction, these vectors are dynamically fused into the hidden layer representation of the model through a gating mechanism (i.e., the initial semantic vector and the knowledge vector are fused using a gating mechanism to obtain the target semantic vector), thereby improving the memory and reasoning ability of professional knowledge.

[0074] Step S2: First Knowledge Base Template Recall and Integration. A knowledge base related to line loss reports (i.e., the first knowledge base) is constructed, storing various report templates and their corresponding semantic vectors. Based on the target semantic vectors obtained from semantic recognition, the Guangming Big Data Model searches the first knowledge base, recalling multiple initial report templates with high matching degrees to user needs. These recalled initial report templates are then integrated to form a preliminary report framework (i.e., the target report template).

[0075] The Guangming Big Data Model processing module matches the target semantic vector of the user's needs with the semantic vectors of over 300 report templates stored in the first knowledge base. During the retrieval process, an improved cosine similarity algorithm is used, and a TF-IDF weight factor is introduced to enhance the influence of keywords. The method for determining the similarity Sim(A,B) between the target semantic vector and the semantic vector of any template in the first knowledge base is the same as in the above embodiment, and will not be repeated here.

[0076] The similarity threshold (i.e., the preset similarity threshold) is set to 0.7, and the top 5 highly matching initial report templates (i.e., the preset number of templates) are integrated. Duplicate chapters are merged using NLP text alignment technology (i.e., duplicate chapter merging), the content order is rearranged based on business logic priority (i.e., chapter sorting), and essential chapters required by industry standards are automatically added (i.e., chapter supplementation). Finally, a preliminary report framework is constructed, including modules such as table of contents, summary, line loss data statistics, cause analysis, and optimization suggestions.

[0077] Step S3, Business Data Acquisition. Using the data platform API, based on the data requirements of each chapter of the report, locate and acquire the corresponding line loss data, equipment operation data, and other business data (i.e., target data) in the data platform's business space. Fill the acquired target data into the preliminary report framework to generate a complete unit line loss business analysis report (i.e., the initial line loss report).

[0078] Based on the data requirements of each chapter in the preliminary report framework, the data platform API generates data retrieval instructions and sends them to the power system business database. The database filters and returns the corresponding business data (i.e., target data) according to the instructions. After receiving the data, the data platform interaction module transmits it to the user interaction module, populates it into the preliminary report framework, and generates the initial line loss report.

[0079] Step S4: Report Content Optimization. After the initial line loss report is generated, the user can select the paragraph to be adjusted and choose the expansion or abbreviation function, while inputting the adjustment requirements (i.e., line loss report optimization requirements). After receiving the user's requirements, the Guangming Big Data Model performs secondary semantic analysis on the selected initial line loss report paragraphs, and combines relevant knowledge and data from the second knowledge base to optimize and adjust the paragraph content, thus enabling the expansion or abbreviation function.

[0080] Users submit optimization requests (i.e., line loss report optimization requests) through a combination of paragraph selection, function button clicks, and text box input on the interactive interface. Upon receiving the optimization instruction, the Guangming Big Data Model processing module initiates secondary semantic analysis, combining industry standard expression libraries, excellent case libraries, and acquired business data from the second knowledge base to optimize the content. In the expansion scenario, for the "Line Loss Anomaly Cause Analysis" paragraph, the model automatically supplements historical fault data of similar equipment, comparative analysis of power equipment parameters, and expert experience suggestions; in the abbreviation scenario, semantic compression technology is used to perform redundancy analysis, removing redundant descriptions while maintaining text coherence. The optimized template line loss report content is updated in real time to the report display interface after syntax validation and logical checks, and a version comparison function is provided to facilitate users to review modification records.

[0081] By utilizing the above-mentioned intelligent generation of line loss reports and semantic-driven content optimization methods based on the Guangming big data model, it is possible to accurately understand user needs, efficiently integrate business data, and flexibly optimize report content.

[0082] Accurately understanding user needs refers to the fact that Guangming's big data model, relying on deep learning and natural language processing technologies, constructs a dedicated knowledge graph covering power industry terminology and business rules. When receiving user requests, the model can intelligently parse ambiguous expressions. For example, "analyze recent line loss anomalies" is automatically transformed into "statistically analyze the line loss rate fluctuations over the past 30 days on a daily / weekly basis, mark outliers, and associate them with the corresponding regional equipment status." Compared to keyword matching technology based on fixed templates, this model can capture implicit information in user needs. When generating line loss reports, it automatically associates historical similar case data and accurately locates the core requirements through semantic similarity calculation. This improves the match between report content and user expectations by more than 40%, effectively solving the problem of rework in line loss reports caused by misunderstandings of requirements.

[0083] Efficient integration of business data refers to the ability to achieve sub-second response times from data sources through the API interfaces of the data platform. By establishing indexes for 10 types of core business data, such as equipment ledgers, metering data, and load curves, the system automatically matches the optimal data retrieval path based on demand characteristics. For example, when generating a transformer area line loss report, the system retrieves electricity consumption data, production system equipment parameters, and GIS system geographic information in parallel. Through data cleaning algorithms, it automatically processes missing and outlier values, improving integration efficiency by 80% compared to manual configuration. This reduces the generation time of a routine line loss report from 4 hours to less than 15 minutes, meeting the timeliness requirements of real-time power grid monitoring and decision support.

[0084] Flexible report content optimization refers to providing functions for expanding and abbreviating line loss reports. In abbreviated mode, a key information extraction algorithm automatically identifies conclusive statements in the report, compressing a lengthy analysis report into a core summary. Expanded mode, based on the power business knowledge base (i.e., a second knowledge base), supplements key conclusions with multi-dimensional data, such as automatically adding comparisons with the same period last year, horizontal comparisons with similar transformer areas, and analyses of the impact of meteorological factors for abnormal line loss rates. Users can also perform in-depth optimization of specific chapters through a visual editing interface, saving modified versions in real time and providing version comparison functionality. This allows line loss reports to meet both the concise reporting needs for upward reporting and the in-depth analysis scenarios required for technical discussions.

[0085] Figure 2 This is a structural diagram of an optional method for generating line loss reports for power equipment according to an embodiment of this application, as shown below. Figure 2The diagram illustrates the process of generating a power equipment line loss report. First, the user submits a line loss report generation request, which is then input into the line loss report generation agent. The Guangming Big Data model within this agent performs semantic recognition on the request, obtaining a target semantic vector and determining the report type, time, and unit required by the user. A similarity analysis is then performed between this target semantic vector and the semantic vectors corresponding to report templates in the first knowledge base. Multiple initial report templates are selected from the first knowledge base and integrated to obtain the target report template. Based on the data requirements of the obtained target report template, the target data is retrieved by calling the data platform API and populated into the target report template, resulting in the user's initial line loss report. If the initial line loss report needs to be optimized and rewritten, the optimization requirements of the line loss report are semantically identified to determine the data required for optimization. The optimization and rewriting of the initial line loss report is then performed by calling the data platform API. The rewritten line loss report is updated to the report display interface in real time, and a version comparison function is provided to facilitate users to review the modification records. The initial line loss report and the rewritten line loss report are then integrated to obtain the user's target line loss report.

[0086] The above-mentioned optional implementation methods achieve at least the following effects: by semantically recognizing the requirements for generating line loss reports, a target semantic vector is obtained, laying a data foundation for the subsequent generation of target line loss reports for power equipment based on the target semantic vector; the fusion of knowledge vectors enables the generated line loss reports to not only meet the personalized needs of users, but also deeply integrate professional knowledge of the power industry, including historical data, industry standards, and expert experience, thereby ensuring the comprehensiveness, professionalism, and depth of the report content and meeting the high standards required for line loss reports in power system operation; by semantically recognizing the requirements for optimizing line loss reports and optimizing the initial line loss reports, the accuracy of the generated target line loss reports and user satisfaction can be further improved.

[0087] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0088] This embodiment also provides a line loss report generation device for power equipment, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0089] According to an embodiment of this application, an apparatus embodiment for implementing a method for generating line loss reports for power equipment is also provided. Figure 3 This is a schematic diagram of a line loss report generation device for power equipment according to an embodiment of this application, such as... Figure 3 As shown, the above-mentioned power equipment line loss report generation device includes a line loss report generation requirement acquisition module 302, a first semantic recognition module 304, a target report template determination module 306, an initial line loss report generation module 308, a second semantic recognition module 310, and an optimization module 312. The device will be described below.

[0090] The line loss report generation requirement acquisition module 302 is used to acquire the line loss report generation requirements of power equipment.

[0091] The first semantic recognition module 304 is connected to the line loss report generation requirement acquisition module 302. It is used to perform semantic recognition on the line loss report generation requirement to obtain the target semantic vector of the line loss report generation requirement. The target semantic vector is used to describe the core content and key information of the line loss report generation requirement.

[0092] The target report template determination module 306 is connected to the first semantic recognition module 304 and is used to determine the target report template based on the target semantic vector and the first knowledge base. The first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively.

[0093] The initial line loss report generation module 308 is connected to the target report template determination module 306 and is used to generate an initial line loss report for power equipment based on the target report template.

[0094] The second semantic recognition module 310 is connected to the initial line loss report generation module 308 and is used to perform semantic recognition on the line loss report optimization requirements to obtain the optimization objectives of the line loss report optimization requirements. The optimization objectives include content expansion or content abbreviation.

[0095] The optimization module 312, connected to the second semantic recognition module 310, is used to optimize the initial line loss report based on the optimization target to obtain the target line loss report.

[0096] This application provides a power equipment line loss report generation device. By setting a line loss report generation requirement acquisition module 302, a first semantic recognition module 304, a target report template determination module 306, an initial line loss report generation module 308, a second semantic recognition module 310, and an optimization module 312, it can achieve the goal of obtaining the target template of the line loss report by performing semantic recognition on the line loss report generation requirement, thereby generating the initial line loss report of the power equipment, and optimizing the initial line loss report according to the line loss report optimization requirement, and finally obtaining the target line loss report of the power equipment. This achieves the technical effect of improving the accuracy of the line loss report generation result of the power equipment, and thus solves the technical problem of inaccurate line loss report generation results of the power equipment in related technologies.

[0097] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0098] It should be noted that the above-mentioned line loss report generation requirement acquisition module 302, first semantic recognition module 304, target report template determination module 306, initial line loss report generation module 308, second semantic recognition module 310, and optimization module 312 correspond to steps S102 to S112 in the embodiments. The instances and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0099] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0100] The aforementioned power equipment line loss report generation device may also include a processor and a memory. The line loss report generation requirement acquisition module 302, the first semantic recognition module 304, the target report template determination module 306, the initial line loss report generation module 308, the second semantic recognition module 310, the optimization module 312, etc. are all stored in the memory as program units, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0101] The processor contains a core that retrieves the corresponding program unit from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0102] This application provides a non-volatile storage medium storing a program that, when executed by a processor, implements a method for generating line loss reports for power equipment.

[0103] This application provides an electronic device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: obtaining a line loss report generation requirement for power equipment; performing semantic recognition on the line loss report generation requirement to obtain a target semantic vector, wherein the target semantic vector describes the core content and key information of the line loss report generation requirement; determining a target report template based on the target semantic vector and a first knowledge base, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to each report template; generating an initial line loss report for power equipment based on the target report template; performing semantic recognition on the line loss report optimization requirement to obtain an optimization objective, wherein the optimization objective includes content expansion or content abbreviation; and optimizing the initial line loss report based on the optimization objective to obtain a target line loss report. The device in this document can be a server, PC, etc.

[0104] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: obtaining a line loss report generation requirement for power equipment; performing semantic recognition on the line loss report generation requirement to obtain a target semantic vector for the line loss report generation requirement, wherein the target semantic vector is used to describe the core content and key information of the line loss report generation requirement; determining a target report template based on the target semantic vector and a first knowledge base, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively; generating an initial line loss report for power equipment based on the target report template; performing semantic recognition on the line loss report optimization requirement to obtain an optimization objective for the line loss report optimization requirement, wherein the optimization objective includes content expansion or content abbreviation; optimizing the initial line loss report based on the optimization objective to obtain a target line loss report.

[0105] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0107] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0108] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0109] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0110] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0111] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0112] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0113] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0114] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for generating line loss reports for power equipment, characterized in that, include: Requires the generation of line loss reports for power equipment; Semantic recognition is performed on the line loss report generation requirement to obtain the target semantic vector of the line loss report generation requirement, wherein the target semantic vector is used to describe the core content and key information of the line loss report generation requirement; Based on the target semantic vector and the first knowledge base, a target report template is determined, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively; Based on the target report template, an initial line loss report for the power equipment is generated; Semantic recognition is performed on the line loss report optimization requirements to obtain the optimization objectives of the line loss report optimization requirements, wherein the optimization objectives include content expansion or content abbreviation; Based on the optimization objective, the initial line loss report is optimized to obtain the target line loss report.

2. The method according to claim 1, characterized in that, The semantic recognition of the line loss report generation requirement to obtain the target semantic vector of the line loss report generation requirement includes: The line loss report generation requirements are preprocessed to obtain standard generation requirements; The standard generation requirements are semantically identified to determine the initial semantic vector of the standard generation requirements; The initial semantic vector and the knowledge vector are fused using a gating mechanism to obtain the target semantic vector. The knowledge vector refers to the structured information of the power equipment line loss report generation standard. The gating mechanism is used to control the fusion process between the target semantic vector and the knowledge vector.

3. The method according to claim 1, characterized in that, The step of determining the target report template based on the target semantic vector and the first knowledge base includes: Based on the target semantic vector and the semantic vectors corresponding to the multiple report templates, multiple similarity results are determined, wherein the multiple similarity results correspond one-to-one with the multiple report templates, and the similarity results are used to describe the degree of similarity between the semantic vectors corresponding to the report templates in the first knowledge base and the target semantic vector; Based on the multiple similarity results, a preset similarity threshold and a preset number of templates are used to select multiple initial report templates from the first knowledge base; The multiple initial report templates are integrated to obtain the target report template, wherein the integration process includes at least one of the following: merging duplicate chapters, sorting chapters, and supplementing chapters.

4. The method according to claim 1, characterized in that, The process of generating an initial line loss report for the power equipment based on the target report template includes: Based on the target report template, determine the data acquisition instructions; Based on the data acquisition instruction, the target data is retrieved from the database; The target data is filled into the target report template to obtain the initial line loss report.

5. The method according to claim 1, characterized in that, When the optimization objective includes content expansion, the optimization of the initial line loss report based on the optimization objective to obtain the target line loss report includes: The expanded content of the initial line loss report is obtained from the second knowledge base, wherein the content of the second knowledge base includes at least: historical line loss cases, comparative analysis of power equipment parameters, and expert advice; the expanded content includes at least: historical line loss cases of equipment similar to the power equipment, comparative analysis of the power equipment parameters, and expert advice on the power equipment. Based on the expanded content, the initial line loss report is expanded to obtain the target line loss report.

6. The method according to any one of claims 1 to 5, characterized in that, When the optimization objective includes content abbreviations, optimizing the initial line loss report based on the optimization objective to obtain the target line loss report includes: Redundancy analysis is performed on the initial line loss report to obtain the abbreviated content of the initial line loss report; Based on the abbreviated content, the initial line loss report is abbreviated to obtain the target line loss report.

7. A device for generating line loss reports for power equipment, characterized in that, include: The line loss report generation requirement acquisition module is used to acquire the line loss report generation requirements of power equipment; The first semantic recognition module is used to perform semantic recognition on the line loss report generation requirement to obtain the target semantic vector of the line loss report generation requirement, wherein the target semantic vector is used to describe the core content and key information of the line loss report generation requirement; The target report template determination module is used to determine the target report template based on the target semantic vector and the first knowledge base, wherein the first knowledge base stores multiple report templates and semantic vectors corresponding to the multiple report templates respectively; An initial line loss report generation module is used to generate an initial line loss report for the power equipment based on the target report template. The second semantic recognition module is used to perform semantic recognition on the line loss report optimization requirements to obtain the optimization objectives of the line loss report optimization requirements, wherein the optimization objectives include content expansion or content abbreviation; An optimization module is used to optimize the initial line loss report based on the optimization target to obtain a target line loss report.

8. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores multiple instructions, which are adapted to be loaded by a processor and executed by the line loss report generation method for power equipment according to any one of claims 1 to 6.

9. An electronic device, characterized in that, include: One or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the line loss report generation method for power equipment according to any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the line loss report generation method for power equipment as described in any one of claims 1 to 6.

Images