Intelligent retrieval method and system for operation and maintenance service of station-city complex based on large model

Abstract

The invention relates to an intelligent retrieval method and system for station-city complex operation and maintenance services based on a large model. The method comprises the following steps: obtaining operation and maintenance query information which is input by a user and is related to station-city complex operation and maintenance; inputting the operation and maintenance query information into the operation and maintenance large model to query and obtain corresponding operation and maintenance response information; the operation and maintenance large model is obtained by training an original large model based on sample operation and maintenance data of the station city complex; the large model comprises a first model and a plurality of second models, and the training process comprises the steps of independently performing iterative training on each second model according to text data and annotation data of each business type, and stopping training until performance parameters of each second model meet preset conditions; and correspondingly inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, executing integrated training of the first model based on the feature set, and deploying the trained first model as an operation and maintenance large model.

Description

Technical Field

[0001] The disclosed embodiments relate to the technical field of intelligent traffic management, and in particular to a method and system for intelligent retrieval of station-city complex operation and maintenance services based on a large model. Background Art

[0002] At present, in large-scale transportation systems, there are many types of equipment in station-city complexes, and the operating status changes are complex. Conventional manual management methods and tools cannot efficiently process massive amounts of information. In the current operation and maintenance decision-making process, operation and maintenance personnel need to spend a lot of time acquiring and organizing data at work, lacking intelligent support. With the increasing complexity of equipment and systems, traditional operation and maintenance management methods have been unable to cope with increasingly complex operation and maintenance needs, especially with the continuous increase in operation and maintenance tasks, the requirements for response speed and decision-making accuracy are also constantly increasing. Traditional operation and maintenance methods usually rely on manual operation and experience judgment, resulting in long response times and error-prone operation and maintenance management results. Summary of the invention

[0003] In order to solve the above technical problems or at least partially solve the above technical problems, the embodiments of the present disclosure provide a method and system for intelligent retrieval of station-city complex operation and maintenance business based on a large model.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for intelligent retrieval of station-city complex operation and maintenance services based on a large model, the method comprising: Obtain operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user; The operation and maintenance query information is input into the operation and maintenance big model to query and obtain the corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on the sample operation and maintenance data of the station-city complex; the original big model includes a first model and a plurality of second models, and the sample operation and maintenance data includes text data and annotation data of a plurality of different business types related to operation and maintenance; the total number of the plurality of second models is the same as the total number of the plurality of different business types; And wherein, the training process of the original large model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, and performing integrated training of the first model based on the feature set, so as to deploy the first model after training as the operation and maintenance large model.

[0005] In one embodiment, the plurality of second models are pre-trained large models, and the training process of the original large model specifically includes: Fine-tune each of the pre-trained large models independently according to the text data of each business type and at least part or all of the annotated data, until the training is stopped when the performance parameters of each of the pre-trained large models meet the preset conditions; Input the text data of each business type into each pre-trained large model after training, form a feature set according to the output of each pre-trained large model, perform integrated training of the first model based on the feature set, end the training when the performance parameters of the first model meet the preset conditions, and deploy the first model after the training as the operation and maintenance large model.

[0006] In one embodiment, the training process of each second model includes: Identify and determine multiple entity keywords and the relationships between the multiple entity keywords from the text data of the sample operation and maintenance data, and construct a knowledge graph of graph-based text index based on the multiple entity keywords and the relationships between the multiple entity keywords; The retrieval enhanced generation model adopts a two-layer retrieval mechanism to retrieve and extract query keywords in the text data of each business type, queries target keywords similar to the query keywords from the knowledge graph, inputs the query keywords and the target keywords together into the second model to obtain generated content; and updates the second model based on the difference between the generated content and the annotated content.

[0007] In one embodiment, the method further comprises: A query user interface is displayed, wherein the query user interface includes a search box, and the search box is used to receive the operation and maintenance query information input by a user, wherein the operation and maintenance query information at least includes keywords, categories or dates related to operation and maintenance.

[0008] In one embodiment, the query user interface includes a first control, and the method further includes: In response to the triggering operation of the first control, common questions about the operation and maintenance of the station-city complex and their answers are displayed, or the user's historical query records are obtained, and questions about the operation and maintenance of the station-city complex related to the historical query records and their answers are displayed.

[0009] In one embodiment, the query user interface includes a second control, and the method further includes: In response to a triggering operation on the second control, displaying the user's historical operation and maintenance query information and the historical operation and maintenance response information given by the operation and maintenance big model; The filtering information input by the user is obtained, and based on the filtering information, target operation and maintenance query information and / or target operation and maintenance response information are filtered from the historical operation and maintenance query information and the historical operation and maintenance response information.

[0010] In one embodiment, the method further comprises: Collecting historical operation and maintenance query information input by the user and historical operation and maintenance response information given by the operation and maintenance big model, and obtaining updated text data from the text data of the multiple different business types; Key feature information is extracted based on the historical operation and maintenance query information, the historical operation and maintenance response information, and the updated text data, and each second model is updated based on the key feature information, thereby updating the first model.

[0011] In a second aspect, the embodiment of the present disclosure provides a large model-based intelligent retrieval system for station-city complex operation and maintenance business, including: The data acquisition module is used to obtain the operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user; A search and query module is used to input the operation and maintenance query information into the operation and maintenance big model to query and obtain corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on the sample operation and maintenance data of the station-city complex; the original big model includes a first model and a plurality of second models, the sample operation and maintenance data includes text data and annotation data of a plurality of different business types related to operation and maintenance; the total number of the plurality of second models is the same as the total number of the plurality of different business types; And wherein, the training process of the original large model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, and performing integrated training of the first model based on the feature set, so as to deploy the first model after training as the operation and maintenance large model.

[0012] In a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned embodiments of the intelligent retrieval method for station-city complex operation and maintenance business based on a large model.

[0013] In a fourth aspect, an embodiment of the present disclosure provides an electronic device, including: Processor; and Memory for storing computer programs; Wherein, the processor is configured to execute the intelligent retrieval method for station-city complex operation and maintenance business based on a large model of any of the above-mentioned embodiments by executing the computer program.

[0014] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages: The embodiment of the present disclosure provides a method and system for intelligent retrieval of station-city complex operation and maintenance business based on a big model, which obtains operation and maintenance query information related to the operation and maintenance of the station-city complex input by a user, and inputs the operation and maintenance query information into the operation and maintenance big model to query and obtain corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on sample operation and maintenance data of the station-city complex; the original big model includes a first model and multiple second models, and the sample operation and maintenance data includes text data and annotation data of multiple different business types related to operation and maintenance; the total number of the multiple second models is the same as the total number of types of the multiple different business types; and wherein the training process of the original big model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; the text data of each business type in the sample operation and maintenance data is input into each trained second model, a feature set is formed according to the output of each second model, and integrated training of the first model is performed based on the feature set, so that the first model after training is deployed as the operation and maintenance big model. The solution of this embodiment can be used as an intelligent operation and maintenance business knowledge retrieval system for the station-city complex, helping operation and maintenance personnel to search for various issues and corresponding solutions directly from the system, helping to improve the efficiency and accuracy of obtaining operation and maintenance response information such as fault diagnosis results, and assisting in improving the work efficiency and accuracy of staff, so that the operation and maintenance personnel of the station-city integrated three-dimensional network can quickly and accurately find relevant solutions when faced with problems, thereby shortening, for example, fault recovery time and improving problem handling efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1This is a flow chart of an intelligent retrieval method for station-city complex operation and maintenance business based on a large model according to an embodiment of the present disclosure; Figure 2 A schematic diagram of the training process of the operation and maintenance big model of the embodiment of the present disclosure; Figure 3 This is a flow chart of a training method for a large operation and maintenance model according to an embodiment of the present disclosure; Figure 4 It is a schematic diagram of an intelligent retrieval system for station-city complex operation and maintenance business based on a large model according to an embodiment of the present disclosure; Figure 5 Schematic diagram of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0018] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0019] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0020] It should be understood that, in the following, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0021] The large models involved in operation and maintenance management in related technologies lack domain specificity. Models that are well trained on general data are not applicable to the operation and maintenance business of various station-city complexes. In addition, performance is highly dependent on the large models used. If these large models fail to cover the complexity of certain queries or knowledge in specific fields, the accuracy of operation and maintenance query results will be insufficient.

[0022] Figure 1The flowchart of a method for intelligent retrieval of station-city complex operation and maintenance business based on a large model according to an embodiment of the present disclosure is shown in FIG. The method can be executed by an electronic device such as a computer or a mobile terminal, and specifically may include the following steps: Step S101: Obtain operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user.

[0023] Exemplarily, the operation and maintenance query information may be an inquiry text sentence related to operation and maintenance, “how to adjust the air conditioning system in the site to maintain optimal energy efficiency and comfort during hot summer weather”.

[0024] Step S102: input the operation and maintenance query information into the operation and maintenance big model to query and obtain corresponding operation and maintenance response information; wherein, the operation and maintenance big model is obtained by training the original big model based on sample operation and maintenance data of the station-city complex.

[0025] Exemplarily, the operation and maintenance response information may be an answer result given to the above-mentioned inquiry text language, such as an operation and maintenance disposal measure or a suggested answer text sentence, for example, "It is recommended that the optimal temperature setting of the air conditioner is 26° and the operating time is set to X". Among them, the original large model may include a first model and multiple second models, which may specifically be machine learning models such as support vector machines, convolutional neural networks CNN, long short-term memory networks (LSTM, Long Short-Term Memory), Transformer models, etc. The sample operation and maintenance data includes text data and annotation data of multiple different business types related to operation and maintenance.

[0026] Specifically, the station-city complex is a fusion of transportation and the city, combining transportation hubs with urban functions, covering transportation, offices, commerce, hotels, cultural facilities, etc. Its types can be divided into high-speed rail station complexes, subway station complexes, etc. Therefore, it involves a variety of business data such as structure (such as building structure, electromechanical equipment, etc.), environment, energy consumption, flow of people, and events. In order to make the operation and maintenance retrieval scheme of this embodiment more field-specific and able to answer users' questions about the various businesses of the station-city complex, it is necessary to prepare the station-city complex operation and maintenance related data in advance, such as through the five major operating state indicators of structure, environment, energy consumption, flow of people and state of affairs, to provide targeted decision support and fault response guidance for site operation and maintenance. The pre-prepared data may include data related to the five major operating state indicators, such as building facility data, detailed information on electromechanical equipment, equipment failure and maintenance records, environmental monitoring data, energy consumption, flow of people data, equipment operation specifications, etc., to ensure that the data is comprehensive and accurate.

[0027] After that, data cleaning is performed, i.e., deleting duplicates, removing outliers, and standardizing data. Then, the data is classified by business type, thereby forming text data of multiple different business types. The text data of each business type can be annotated based on the specific operation and maintenance tasks to be implemented, such as question-answer pairs, faults, and corresponding solution strategies. In this way, the actual operation and maintenance data of the site is used for the following independent training to ensure that the corresponding model can accurately understand and respond to specific problems in site operations.

[0028] Next, you can perform natural language processing on the text content to remove irrelevant characters such as HTML tags, special symbols, etc., split continuous text strings into separate vocabulary units, and remove common but meaningless words such as "and" and "is".

[0029] Finally, feature engineering is used to convert the text data into a numerical format that can be processed by the machine learning algorithm. In other embodiments, for text data of different business types, a suitable storage system can also be selected according to the data access frequency and the data volume to store the training data corresponding to each second model respectively.

[0030] The total number of the plurality of second models is the same as the total number of the plurality of different business types, that is, each second model has training data of its corresponding business type. For example, business type 1 corresponds to model 1, business type 2 corresponds to model 2, and so on. Figure 2 As shown, the training process of the original large model can include the following steps: Iteratively training each second model independently according to the text data and the annotated data of each business type until the performance parameter of each second model meets the preset conditions and the training is stopped, wherein the text data of each business type corresponds to one second model; The text data of each business type in the sample operation and maintenance data is input into each trained second model, a feature set is formed according to the output of each second model, and integrated training of the first model is performed based on the feature set, so that the first model after training is deployed as the large operation and maintenance model.

[0031] Exemplarily, the feature set may be a set of feature vectors corresponding to the output of each second model. The performance parameter of each second model satisfies a preset condition, specifically, the performance parameter of each second model may be greater than or equal to a preset value (such as the loss function value of the model is less than or equal to a preset loss value), and the preset value may be set according to specific needs. It is understandable that the preset values ​​corresponding to the performance parameters of each second model may be different, for example, model 1 corresponds to preset value 1, and model 2 corresponds to preset value 2, and preset value 2 may be different from preset value 1, because the operation and maintenance tasks implemented after each second model is trained are usually different, and the training data is different.

[0032] For each of the operational tasks implemented for second model training, here are some examples: 1. Solve environmental regulation problems Q&A scenario: The operation and maintenance personnel ask how to adjust the air conditioning system in the site to maintain optimal energy efficiency and comfort during hot summer weather.

[0033] Data training: The model can use historical weather data, passenger flow, and air conditioning system performance data to recommend optimal temperature settings and operating times.

[0034] 2. Peak traffic management Question and Answer Scenario: How to adjust entrance and exit management during holidays or large-scale events to avoid crowding and improve safety.

[0035] Data training: Using crowd flow data from past holidays, accident records, and current safety protocols, the model is trained to provide real-time crowd control strategies and emergency evacuation recommendations.

[0036] 3. Energy consumption optimization Q&A scenario: Ask how to reduce nighttime energy consumption without compromising safety and basic operational needs.

[0037] Data training: Combining historical energy usage data, nighttime operating activities, and equipment energy efficiency standards, the model provides energy-saving operation recommendations, such as adjusting lighting and non-critical equipment usage.

[0038] 4. Fault response and maintenance Q&A scenario: The elevator stops running, and the operation and maintenance personnel need to quickly diagnose the problem and obtain maintenance instructions.

[0039] Data training: Use the elevator’s fault history, maintenance manuals, and success stories to train the model to provide fault analysis and repair steps.

[0040] 5. Safety monitoring and emergency response Question and answer scenario: Found a suspicious package, asked how to perform security checks and determine whether passengers need to be evacuated.

[0041] Data training: Using security protocols, historical security incident handling results and relevant laws and regulations, the training model provides immediate action guidance and communication strategies. The above is just an example and is not limited to this.

[0042] After training with different base models, i.e., multiple second models and corresponding text data according to the business type of the data, since the data involved in the station-city complex is large and complex, at least some or all of the text data of multiple different business types are associated with each other. Therefore, in this embodiment, the ensemble learning method is continued to be used for fusion training, that is, the outputs of these base models are input as new feature sets into a meta-model, i.e., the first model, whose task is to learn how to most effectively combine feature information from different base models. In the training phase, the performance of each base model is first evaluated on an independent validation set (obtained by dividing part of the data based on the sample operation and maintenance data). After the performance meets the requirements, the outputs of these trained base models are then used to train the meta-model. This process can optimize each model through cross-validation to ensure that each model makes full use of the data while avoiding overfitting. Finally, the trained meta-model is deployed in the actual operation and maintenance environment as a large operation and maintenance model.

[0043] This integrated model will be able to provide more accurate and reliable predictions than a single model, helping to improve the efficiency and accuracy of operation and maintenance response information such as fault diagnosis results. In addition, a model performance monitoring mechanism and a regular update process can be established to adapt to changes in the operation and maintenance environment and new data inputs. For details, please refer to the corresponding description below.

[0044] In this embodiment, the first model and the second model are both large language models (LLM), which can deeply understand and process natural language text, so as to achieve more accurate semantic analysis when processing operation and maintenance text data. Through the natural language processing (NLP) capability of the large model, it can automatically extract valuable feature information such as feature vectors from various unstructured text data, such as operation and maintenance fault reports, maintenance logs, operation manuals, and user feedback for model training, so that the trained operation and maintenance large model can finally improve the accuracy of the given operation and maintenance response information such as fault diagnosis results.

[0045] The above scheme of this embodiment can help operation and maintenance personnel to quickly and accurately locate and solve problems, reduce the response time for handling equipment failures, and can realize automatic classification of problems and accurate recommendation of solutions. Based on this large model, an intelligent operation and maintenance business knowledge retrieval system for the station-city complex can be established, which can centrally store and manage problems and solutions in the system, so that operation and maintenance personnel of the station-city integrated three-dimensional network can quickly and accurately find relevant solutions when facing problems, thereby shortening the fault recovery time and improving the efficiency of problem handling. In addition, it can also accumulate and preserve the knowledge and experience of the team and promote knowledge sharing among team members.

[0046] In one embodiment, continue to refer to Figure 2As shown, the multiple second models are pre-trained large models, and the training process of the original large model specifically includes: Fine-tune each of the pre-trained large models independently according to the text data of each business type and at least part or all of the annotated data, until the training is stopped when the performance parameters of each of the pre-trained large models meet the preset conditions; Input the text data of each business type into each pre-trained large model after training, form a feature set according to the output of each pre-trained large model, perform integrated training of the first model based on the feature set, end the training when the performance parameters of the first model meet the preset conditions, and deploy the first model after the training as the operation and maintenance large model.

[0047] Exemplarily, each second model can be a pre-trained large model that has been pre-trained. For example, different base models, i.e., second models, are used to perform independent training according to the business type of the text data, and then the model is fine-tuned for text data of different business types such as structure, environment, energy consumption, flow of people, and events. After that, the feature set composed of the output of each pre-trained large model is integrated and learned by combining the ensemble learning method, and the first model is trained based on the feature set. By fine-tuning the pre-trained model and training the first model based on the feature set composed of the output of each pre-trained large model, the trained operation and maintenance large model can further improve the accuracy of operation and maintenance response information such as fault diagnosis results, while improving the efficiency of model training and reducing training costs.

[0048] Based on any one of the above embodiments, in one embodiment, in combination with reference Figure 3 As shown in , the training process of each second model may include the following steps: Step S301: identifying and determining a plurality of entity keywords and the relationships between the plurality of entity keywords from the text data of the sample operation and maintenance data, and constructing a knowledge graph of a graph-based text index based on the plurality of entity keywords and the relationships between the plurality of entity keywords; Step S302: The retrieval enhancement generation model uses a two-layer retrieval mechanism to retrieve and extract query keywords in the text data of each business type, queries target keywords similar to the query keywords from the knowledge graph, and inputs the query keywords and the target keywords into the second model together to obtain generated content; and updates the second model based on the difference between the generated content and the annotated content.

[0049] For example, entity recognition technology can be used to extract entities such as structured data objects such as electromechanical equipment and human traffic from the text data of sample operation and maintenance data. Entity keywords can be words that represent attributes such as the name, location, and energy consumption of electromechanical equipment. The relationship between multiple entity keywords can be the relationship between objects indicated by different entities, such as the relationship between electromechanical equipment (such as elevators) and human traffic. After that, a knowledge graph with graph-based text index can be established.

[0050] The two-layer retrieval mechanism extracts global keywords and local keywords from the text data input into the model during training as query keywords, so as to obtain query keywords in the text data from both high-level and low-level aspects. This can capture complex interdependencies, and at the same time combine the previously constructed knowledge graph to obtain similar keywords as an aid, and input them into the second model together to obtain generated content, and then update the second model. In this way, the final trained operation and maintenance large model can further improve the accuracy of operation and maintenance response information such as fault diagnosis results.

[0051] The accurate retrieval capability of the retrieval enhancement technology (RAG) used in the prior art can improve the natural language generation capability, but the traditional RAG method performs poorly when facing complex or uncommon queries and relies on a predefined document set. For example, the existing RAG technology relies on vector similarity and re-ranking algorithms to retrieve the most similar questions and answers, which is not accurate enough when facing complex queries or new types of queries involved in the operation and maintenance of the station-city complex, resulting in the accuracy of the output results of the final trained operation and maintenance large model to be further improved. In this regard, a new method is used in this embodiment to improve the text indexing and retrieval process using a graph structure, extract entities and their relationships from text data through a large model, and construct a knowledge graph of graph-based text indexing; in addition, the combination of a two-layer retrieval mechanism can accurately index specific details locally, while globally and widely indexing and retrieving related topics and concepts, so as to ensure that the text generated by the trained operation and maintenance large model is both accurate and information-rich, and can further improve the accuracy of operation and maintenance response information such as fault diagnosis results.

[0052] In one embodiment, the method further includes: displaying a query user interface, the query user interface including a search box, the search box being used to receive the operation and maintenance query information input by a user, the operation and maintenance query information including at least operation and maintenance related keywords, categories or dates.

[0053] For example, this embodiment can provide a search and query solution that is easy for users to operate, provide an intelligent search box in a prominent position in the query user interface, and provide a powerful search engine that allows users to conduct detailed searches by keywords, categories, or dates to quickly find relevant troubleshooting solutions and best practice guidelines. In some embodiments, the search algorithm can be optimized to ensure that the most relevant documents, solutions, and other response information are returned even when there are errors in the input or incomplete keywords.

[0054] In some embodiments, knowledge content classification can also be implemented: the knowledge base content is clearly classified according to logic and function, such as structural maintenance, environmental monitoring, etc., and each classification is equipped with an intuitive icon and a brief description. Users can use filters to filter required content based on content type (such as guidelines, case studies, operating procedures), relevance, or last updated time. In some embodiments, a multi-level navigation menu can also be provided to allow users to drill down into specific subcategories, making it easier for users to accurately find knowledge items in specific fields.

[0055] In one embodiment, the query user interface includes a first control such as a virtual button, and the method further includes: in response to a trigger operation on the first control such as a mouse click or a touch click, displaying common questions about the operation and maintenance of the station-city complex and their answers, or obtaining the user's historical query records, and displaying questions about the operation and maintenance of the station-city complex related to the historical query records and their answers.

[0056] For example, in this embodiment, FAQ can be displayed: a FAQ section is provided in the user interface to centrally display common questions and answers about the operation and maintenance of the station-city complex, helping users to quickly find solutions to common questions. In some embodiments, FAQ questions and answers related to historical query records such as browsing and searching of users can be recommended and displayed to provide answers to problems that users may encounter in advance.

[0057] In one embodiment, the query user interface includes a second control such as a virtual button, and the method further includes: in response to a trigger operation on the second control such as a mouse click or a touch click, displaying the user's historical operation and maintenance query information and the historical operation and maintenance response information given by the operation and maintenance big model; obtaining the filtering information input by the user, and filtering the target operation and maintenance query information and / or target operation and maintenance response information from the historical operation and maintenance query information and the historical operation and maintenance response information based on the filtering information.

[0058] In this embodiment, the user's query history can be displayed: an easily accessible interface section is designed in the user interface to specifically display the user's past operation and maintenance query information and the response given by the system, such as providing a list view, in which each entry contains the query time, query question, and the answer provided by the system. In some embodiments, users can search and filter history records by filtering information by conditions such as date, question keywords, or answer type, and users can also export their query history records, which may be in PDF or Excel format. Automatic cleanup options can also be provided to allow users to set the length of time to retain history records, etc.

[0059] In one embodiment, the method may further include the following steps: Collecting historical operation and maintenance query information input by the user and historical operation and maintenance response information given by the operation and maintenance big model, and obtaining updated text data from the text data of the multiple different business types; Key feature information is extracted based on the historical operation and maintenance query information, the historical operation and maintenance response information, and the updated text data, and each second model is updated based on the key feature information, thereby updating the first model.

[0060] Exemplarily, the text data of multiple different business types can be documents such as operation manuals, performance parameters of new equipment, etc., and the updated text data is the updated operation manuals, performance parameters of new equipment, etc. In this embodiment, an incremental learning strategy can be adopted. First, the operation and maintenance query data during the user's use and the historical operation and maintenance response information given by the large model, i.e., feedback records, are collected. Combined with the updated operation manuals, performance parameters of new equipment, etc., these new text data are classified according to different business types, and then the model is updated and trained in a targeted manner. The key features such as feature vectors are extracted from these classified new text data, and then the second model and the first model are updated using these key features. The process of updating and training can refer to the above-mentioned embodiment. At this time, only the training data is different, and the knowledge base, i.e., the knowledge graph, can be updated according to the new text data and feedback during the updating and training process. For example, if a new type of fault or a new maintenance strategy is detected, the relevant content and model should be updated in a timely manner to reflect the latest knowledge and user needs, and measures should be taken to avoid forgetting old knowledge, such as retaining old knowledge through experience playback. Finally, the performance of the model can be evaluated regularly to ensure that new knowledge is properly integrated, trained and updated, and the overall quality of the operation and maintenance knowledge retrieval system and the accuracy of the response results, i.e., the operation and maintenance response information, can be improved by updating and adjusting the model parameters.

[0061] The above solution in this embodiment can also bring the following beneficial effects: (1)Improve problem-solving efficiency and accuracy Updating the training model based on the accumulated and updated problem-solving solutions can help operation and maintenance personnel quickly locate and solve problems, reduce fault response time, and realize automatic classification of problems and accurate recommendation of solutions, thereby improving problem-solving efficiency and accuracy.

[0062] (2) Improving decision-making quality The data in the knowledge retrieval system can support data analysis and reporting, provide decision support for management, and through the analysis of historical data, predict potential problems and trends and formulate response strategies in advance.

[0063] (3) Reduce operation and maintenance costs By improving problem-solving efficiency and accuracy and reducing duplication of labor, the labor cost of operation and maintenance can be reduced, helping to reduce business interruption and recovery costs caused by improper problem solving.

[0064] It should be noted that, although the steps of the method in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc. In addition, it is also easy to understand that these steps may be, for example, executed synchronously or asynchronously in multiple modules / processes / threads.

[0065] like Figure 4 As shown, the embodiment of the present disclosure provides a large model-based intelligent retrieval system for station-city complex operation and maintenance business, including: The data acquisition module 401 is used to acquire operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user; A search and query module 402 is used to input the operation and maintenance query information into the operation and maintenance big model to query and obtain corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on the sample operation and maintenance data of the station-city complex; the original big model includes a first model and a plurality of second models, and the sample operation and maintenance data includes text data and annotation data of a plurality of different business types related to operation and maintenance; the total number of the plurality of second models is the same as the total number of the plurality of different business types; And wherein, the training process of the original large model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, and performing integrated training of the first model based on the feature set, so as to deploy the first model after training as the operation and maintenance large model.

[0066] In one embodiment, at least part or all of the text data of multiple different business types are associated with each other.

[0067] In one embodiment, the plurality of second models are pre-trained large models, and the training process of the original large model specifically includes: Fine-tune each of the pre-trained large models independently according to the text data of each business type and at least part or all of the annotated data, until the training is stopped when the performance parameters of each of the pre-trained large models meet the preset conditions; Input the text data of each business type into each pre-trained large model after training, form a feature set according to the output of each pre-trained large model, perform integrated training of the first model based on the feature set, end the training when the performance parameters of the first model meet the preset conditions, and deploy the first model after the training as the operation and maintenance large model.

[0068] In one embodiment, the training process of each second model includes: Identify and determine multiple entity keywords and the relationships between the multiple entity keywords from the text data of the sample operation and maintenance data, and construct a knowledge graph of graph-based text index based on the multiple entity keywords and the relationships between the multiple entity keywords; The retrieval enhanced generation model adopts a two-layer retrieval mechanism to retrieve and extract query keywords in the text data of each business type, queries target keywords similar to the query keywords from the knowledge graph, inputs the query keywords and the target keywords together into the second model to obtain generated content; and updates the second model based on the difference between the generated content and the annotated content.

[0069] In one embodiment, the system also includes a display module for: displaying a query user interface, the query user interface including a search box, the search box for receiving the operation and maintenance query information input by a user, the operation and maintenance query information at least including operation and maintenance related keywords, categories or dates.

[0070] In one embodiment, the query user interface includes a first control, and the display module of the system can also be used to: in response to a trigger operation on the first control, display common questions and answers about the operation and maintenance of the station-city complex, or obtain the user's historical query records and display questions and answers about the operation and maintenance of the station-city complex related to the historical query records.

[0071] In one embodiment, the query user interface includes a second control, and the system may further include a filtering module, which is used to: display the user's historical operation and maintenance query information and the historical operation and maintenance response information given by the operation and maintenance big model in response to a trigger operation on the second control; obtain the filtering information input by the user, and filter the target operation and maintenance query information and / or target operation and maintenance response information from the historical operation and maintenance query information and the historical operation and maintenance response information based on the filtering information.

[0072] In one embodiment, the system may also include an update module, which is used to: collect historical operation and maintenance query information input by users and historical operation and maintenance response information given by the operation and maintenance large model, and obtain updated text data from the text data of the multiple different business types; extract key feature information based on the historical operation and maintenance query information, the historical operation and maintenance response information and the updated text data, and update each second model based on the key feature information, thereby updating the first model.

[0073] Regarding the system in the above embodiment, the specific manner in which each module performs operations and the corresponding technical effects brought about have been described in detail in the embodiment of the method, and will not be elaborated here.

[0074] It should be noted that although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the present disclosure, the features and functions of two or more modules or units described above can be concretized in one module or unit. Conversely, the features and functions of a module or unit described above can be further divided into multiple modules or units for concretization. The components displayed as modules or units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. Those of ordinary skill in the art can understand and implement it without paying creative work.

[0075] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for intelligent retrieval of station-city complex operation and maintenance business based on a large model according to any of the above embodiments is implemented.

[0076] Exemplarily, the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0077] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0078] The present disclosure also provides an electronic device, including a processor and a memory, wherein the memory is used to store a computer program, wherein the processor is configured to execute the large model-based intelligent retrieval method for station-city complex operation and maintenance business of any of the above embodiments by executing the computer program.

[0079] Refer to the following Figure 5 The electronic device 600 according to this embodiment of the present invention is described. Figure 5 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0080] like Figure 5 As shown, the electronic device 600 is in the form of a general computing device. The components of the electronic device 600 may include but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0081] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present invention described in the above method embodiment section of this specification. For example, the processing unit 610 can perform the following steps: Figure 1The steps of the method shown in .

[0082] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0083] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0084] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0085] The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 650. In addition, the electronic device 600 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0086] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the method steps of the above-mentioned embodiments according to the implementation methods of the present disclosure.

[0087] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0088] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligent retrieval of station-city complex operation and maintenance services based on a large model, characterized in that: The method includes: Obtain operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user; The operation and maintenance query information is input into the operation and maintenance big model to query and obtain the corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on the sample operation and maintenance data of the station-city complex; the original big model includes a first model and a plurality of second models, and the sample operation and maintenance data includes text data and annotation data of a plurality of different business types related to operation and maintenance; the total number of the plurality of second models is the same as the total number of the plurality of different business types; And wherein, the training process of the original large model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, and performing integrated training of the first model based on the feature set, so as to deploy the first model after training as the operation and maintenance large model.

2. The method according to claim 1, characterized in that The plurality of second models are pre-trained large models, and the training process of the original large model specifically includes: Fine-tune each of the pre-trained large models independently according to the text data of each business type and at least part or all of the annotated data, until the training is stopped when the performance parameters of each of the pre-trained large models meet the preset conditions; Input the text data of each business type into each pre-trained large model after training, form a feature set according to the output of each pre-trained large model, perform integrated training of the first model based on the feature set, end the training when the performance parameters of the first model meet the preset conditions, and deploy the first model after the training as the operation and maintenance large model.

3. The method according to claim 1 or 2, characterized in that: The training process of each second model includes: Identify and determine multiple entity keywords and the relationships between the multiple entity keywords from the text data of the sample operation and maintenance data, and construct a knowledge graph of graph-based text index based on the multiple entity keywords and the relationships between the multiple entity keywords; The retrieval enhanced generation model adopts a two-layer retrieval mechanism to retrieve and extract query keywords in the text data of each business type, queries target keywords similar to the query keywords from the knowledge graph, inputs the query keywords and the target keywords together into the second model to obtain generated content; and updates the second model based on the difference between the generated content and the annotated content.

4. The method according to claim 1 or 2, characterized in that: The method further includes: A query user interface is displayed, wherein the query user interface includes a search box, and the search box is used to receive the operation and maintenance query information input by a user, wherein the operation and maintenance query information at least includes keywords, categories or dates related to operation and maintenance.

5. The method according to claim 4, characterized in that The query user interface includes a first control, and the method further includes: In response to the triggering operation of the first control, common questions about the operation and maintenance of the station-city complex and their answers are displayed, or the user's historical query records are obtained, and questions about the operation and maintenance of the station-city complex related to the historical query records and their answers are displayed.

6. The method according to claim 4, characterized in that The query user interface includes a second control, and the method further includes: In response to a triggering operation on the second control, displaying the user's historical operation and maintenance query information and the historical operation and maintenance response information given by the operation and maintenance big model; The filtering information input by the user is obtained, and based on the filtering information, target operation and maintenance query information and / or target operation and maintenance response information are filtered from the historical operation and maintenance query information and the historical operation and maintenance response information.

7. The method according to claim 3, characterized in that The method further includes: Collecting historical operation and maintenance query information input by the user and historical operation and maintenance response information given by the operation and maintenance big model, and obtaining updated text data from the text data of the multiple different business types; Key feature information is extracted based on the historical operation and maintenance query information, the historical operation and maintenance response information, and the updated text data, and each second model is updated based on the key feature information, thereby updating the first model.

8. An intelligent retrieval system for station-city complex operation and maintenance business based on a large model, characterized in that: include: The data acquisition module is used to obtain the operation and maintenance query information related to the operation and maintenance of the station-city complex input by the user; A search and query module is used to input the operation and maintenance query information into the operation and maintenance big model to query and obtain corresponding operation and maintenance response information; wherein the operation and maintenance big model is obtained by training the original big model based on the sample operation and maintenance data of the station-city complex; the original big model includes a first model and a plurality of second models, the sample operation and maintenance data includes text data and annotation data of a plurality of different business types related to operation and maintenance; the total number of the plurality of second models is the same as the total number of the plurality of different business types; And wherein, the training process of the original large model includes: iteratively training each second model independently according to the text data and annotation data of each business type, until the training is stopped when the performance parameters of each second model meet the preset conditions, wherein the text data of each business type corresponds to a second model; inputting the text data of each business type in the sample operation and maintenance data into each trained second model, forming a feature set according to the output of each second model, and performing integrated training of the first model based on the feature set, so as to deploy the first model after training as the operation and maintenance large model.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the large-model-based intelligent retrieval method for station-city complex operation and maintenance business as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: processor; as well as Memory for storing computer programs; Wherein, the processor is configured to execute the intelligent retrieval method for station-city complex operation and maintenance business based on a large model as described in any one of claims 1 to 7 by executing the computer program.

Images