101 results about "Industry standard" patented technology

The invention relates to the technical field of artificial intelligence, and discloses a contract generation system and method based on multiple agents, and the system comprises a user interaction module which is used for receiving a contract demand inputted by a user and displaying a generation result; the knowledge base module is used for storing and retrieving legal terms, industry specifications and contract templates in a distributed manner; the agent cluster module comprises a demand analysis agent, a law expert agent, an industry expert agent, a risk assessment agent, a language optimization agent and a coordination control agent, the agents interact through a standardized communication protocol, and a coordination strategy is dynamically adjusted based on contract complexity calculated in real time; and full-process automation of contract generation is realized. Through cooperative work of multiple agents, the efficiency and quality of contract generation are improved, the law compliance and industry suitability of the contract are ensured, the risk in the contract generation process is reduced, and the user experience is improved.

The invention relates to the technical field of translation, and discloses a translation calibration method based on big data, which can accurately identify semantic deviation in a multi-meaning scene, for example, accurately judge that'cell 'should be translated into'cell' instead of'house 'in a biological text. The system can detect the language field mismatching problem of spoken expression in the official file, and provides term correction suggestions conforming to industry standards. Aiming at technical terms with multiple meanings in legal clauses, the calibration process can be combined with context window analysis to select an optimal translation scheme, the omission ratio of manual review is remarkably reduced, continuous optimization and cross-language knowledge sharing of a translation calibration model are realized, and the accuracy of translation calibration is improved. The problems that a traditional system is lagged in updating and low-resource language performance is insufficient are solved.

The invention belongs to the field of digital twin technology and emergency management systems, and particularly relates to an emergency drill evaluation system and method based on a digital twin airport oil depot, and the system comprises a data acquisition and management module, an emergency drill scheme management module, an oil depot twin environment construction module, a simulation deduction module, and a comprehensive analysis and evaluation module. According to the emergency drill evaluation system and method based on the digital twin airport oil depot, through a multi-dimensional evaluation index system constructed by a comprehensive analysis and evaluation module, and in combination with an analytic hierarchy process and a fuzzy comprehensive evaluation method, a drill effect is quantitatively evaluated from multiple aspects of emergency response speed, resource utilization efficiency, loss control degree and the like; various problems in the drilling process can be deeply analyzed, the generated assessment report not only comprises detailed problem analysis, but also provides targeted improvement suggestions, and meanwhile, historical data and industry standards can be compared, so that assessment is more scientific and accurate, and powerful support is provided for emergency management decision making.

A computing platform performs automated due diligence by acquiring project scope information and documents, standardizing all data across all document types into a uniform format optimized for analysis and subjecting the data to repeated querying by large language model agents to triangulate it against a body of human content expertise coded into the platform. The platform compares variables with a project proforma, verifies permits by parsing permit data and cross checking a permitting authority portal, evaluates design completeness by detecting standards of readiness status, and measures the bankability of project agreements compared to industry standards. The platform computes environmental, social, and governance metrics, assesses counterparty risk and community sentiment using searches of public sources, and validates a reported capital stack by analyzing documents and applying industry standards thereto. The platform calculates risk metrics, applies weights to produce two scores, and generates a report that includes remediation steps.

The invention relates to the technical field of project management, in particular to an intelligent project management method and system based on home decoration construction, and the method comprises the steps: collecting the initial demand data of a home decoration construction project; splitting the home decoration construction project into a plurality of different construction subtasks; on the basis of a preset intelligent algorithm, historical project data and a preset industry standard are called, needed artificial resource data are matched for all the subtasks on the basis of a preset matching rule, and a construction progress plan is generated in combination with the dependency relationship between the subtasks; executing a home decoration construction project based on the construction progress plan, and collecting real-time progress data of the home decoration construction project; comparing the real-time progress data with the construction progress plan, and outputting a construction deviation report; judging whether the project meets an early warning condition or not; and when an early warning condition is met, positioning an early warning position and sending an early warning notice to the project management end. The method has the effect of remarkably reducing the construction shutdown rate, the rework rate and the construction period deviation.

The invention is suitable for the technical field of industrial automation and artificial intelligence, and provides an instrument construction intelligent recommendation system and method based on LangGraph, LLM and RAG, the system comprises a user interaction module, a LangGraph task control module, a hybrid intelligent classification module, an enhanced RAG module, an LLM recommendation generation module and a comprehensive evaluation system; the method comprises the steps that an instrument list and an instruction uploaded by a user are obtained, a task execution graph is constructed through LangGraph, subtasks are decomposed, after instrument list data are analyzed, a multi-level mixed strategy is adopted to classify instruments, industry standards and historical documents are retrieved through enhanced RAG based on a classification result, and construction recommendation texts are generated in combination with LLM. And finally evaluating and optimizing through a comprehensive evaluation module. According to the method, automation and precision of instrument construction recommendation are realized, the engineering efficiency and compliance are improved, the labor cost is reduced, and the method has a good industrial popularization prospect.

The invention discloses a contract auditing method and device based on an operation and maintenance large model, equipment and a medium, and relates to the field of artificial intelligence, and the method comprises the steps: integrating a historical purchase contract, a target law regulation, a preset industry standard and preset related data to generate historical contract auditing data, performing data marking on the historical contract auditing data to generate historical label data; inputting the historical contract auditing data and the historical label data into the initial operation and maintenance large model for training to obtain a to-be-tested operation and maintenance large model, and dividing the historical contract auditing data into a verification set and a test set according to a preset division mode, performing performance test operation on the to-be-tested operation and maintenance large model according to the verification set and the test set to obtain a target operation and maintenance large model; and obtaining a to-be-audited purchase contract, extracting target key information by using the target operation and maintenance large model, identifying a target risk point based on the target key information, and quantitatively evaluating the target risk point according to a preset risk matrix to generate a purchase contract audit report. And the pre-auditing efficiency of the purchase contract is improved.

The invention discloses a rehabilitation assistive device assessment method based on multi-agent and RAG technology. The method comprises the following steps: constructing a multi-agent system which comprises a rehabilitation physician agent, an auxiliary tool engineer agent, a nurse agent and a user interaction agent; building a dynamic knowledge base based on an RAG technology, wherein the dynamic knowledge base comprises an auxiliary tool product database, a rehabilitation medicine knowledge base, a clinical case base and an industrial standard base; interacting with the user, and dynamically adjusting the question mode, the language style and the question sequence according to the user type and the answer content; retrieving in a dynamic knowledge base according to the obtained information, and respectively outputting professional evaluation suggestions; and according to the collaborative decision model, integrating professional evaluation suggestions of the rehabilitation physician intelligent agent, the auxiliary tool engineer intelligent agent and the nursing personnel intelligent agent to generate a multi-dimensional auxiliary tool evaluation report and recommendation scheme. According to the method, the efficiency of the evaluation process of the assistive device is higher, and the evaluation result is more accurate and effective.

The invention discloses an electronic manufacturing industry production line UPH intelligent prediction method, system, device and medium, and the method can learn a deep interaction relationship between a line body and a product through adversarial training through a generator of a collaborative filtering generative adversarial network, so that even if a new product lacks enough historical data, the production line can be predicted more accurately. The generator can also find a mode similar to other line bodies in a potential space and generate a relatively reasonable UPH, and for an old product with a theoretical UPH and an actual UPH, the relatively reasonable UPH can be generated only by correspondingly inputting a current line body feature vector and the theoretical UPH of the product in the corresponding line body; according to the method, the difficulty that in the initial production stage of a new product, due to lack of reference bases, UPH can be generally formulated only according to the same industry standard or the average yield per hour of the production line can be solved; and meanwhile, the problem of UPH updating hysteresis caused by personnel or equipment can be solved.

The invention belongs to the technical field of data mining, and discloses a standard necessary patent prediction method based on a large language model and a related device. The method comprises the following steps: acquiring a feature group of a patent text to be predicted; inputting the feature group of the patent text to be predicted into a pre-trained large language model for prediction, and obtaining a standard necessary patent prediction result based on the large language model; wherein the pre-trained large language model is obtained by training a general large language model based on a plurality of standard necessary patents. According to the model, a domain knowledge enhanced large language model and a dynamic time sequence analysis framework are fused, and patent domain pre-training and multi-source heterogeneous data fusion are introduced, so that the patent text understanding precision of the model is improved; and meanwhile, a dynamic prediction mechanism is constructed in combination with industrial standard evolution data and technology development trend data, so that prospective pre-judgment of standard necessary patents is realized, and powerful support is provided for enterprise patent strategy formulation, technology research and development decision and market competition.

The invention discloses a method for connecting longitudinal steel bars of fabricated concrete prefabricated components, which comprises the following steps of: preparing a rectangular metal frame body with connecting holes at two ends, and respectively connecting the two ends of the frame body structure with the longitudinal steel bars of the two prefabricated components to be assembled by using a straight thread connection method; therefore, the problems of centering deviation and difficulty in connection of the longitudinal steel bars of the fabricated concrete prefabricated component are solved, and meanwhile, the requirements of the existing industrial standard of steel bar mechanical connection technical regulations JGJ 107 in China on a steel bar mechanical connection joint can be met. According to the method, the problem of unreliable connection caused by centering deviation of the longitudinal steel bars of the two prefabricated parts is solved by preparing the frame body matched with the centering deviation value of the steel bars, the connection method is simple in step, needed materials are simple and easy to obtain, the installation efficiency of longitudinal steel bar connection of the fabricated concrete prefabricated parts can be improved, and the purposes of simple construction operation and low cost are achieved. The installation quality is guaranteed, and the stress performance of the connector is improved.

The invention discloses an intelligent object searching and cleaning method based on SLAM and RRT technologies, belongs to the technical field of intelligent robots, and aims at solving the problems that an existing product is single in function, poor in environmental adaptability, low in intelligence and lack of industrial standards. The robot comprises an object searching module, a cleaning module, a control module, an interaction module and an industrial standard adaptation module, the object searching module achieves differentiated grabbing through multi-sensor fusion, SLAM + improved RRT navigation and optimized YOLOv5s recognition in cooperation with a replaceable mechanical arm, the cleaning module comprises a replaceable cleaning assembly and a material self-adaption and stain detection unit, a robot body is designed in an IP67 waterproof and foldable mode, the robot body is adapted to a wet / narrow space, and the robot body is provided with an intelligent control module. The interaction module achieves real-time control through APP dual connection and H.265 low-delay video transmission, indexes are standardized through modularization, and a valuables payment mechanism is established. Cleaning and object searching are integrated, the object searching efficiency is improved, the coverage rate of cleaning blind areas is high, the robot is suitable for family and commercial scenes, industry standardization is promoted, and the robot has high practicability and market value.

The invention discloses a method for improving an industrial safety education training effect based on an AI large model. The method comprises the following steps: analyzing, vectorizing and storing latest industrial safety related documents to obtain a safety knowledge base; the related documents comprise industrial standard specifications, laws and regulations and enterprise specifications related to industrial safety; the knowledge base comprises a plurality of sub knowledge bases with different themes and different types; matching an associated target sub-knowledge base from the safety knowledge base based on the training plan, so as to generate an employee training task based on the target sub-knowledge base; based on the training task, utilizing an AI large model to generate a training problem in the training process, and using the training problem for employee training to obtain a training result of the employee; and based on the training result, using an AI large model to generate test questions corresponding to different employees, and examining the employees based on the test questions to obtain an examination result. The training effect of employee safety education can be improved, and potential safety hazards are reduced.

The invention discloses a test case generation method and device, a medium and a product, and relates to the technical field of automatic testing, and the test case generation method comprises the steps: obtaining a Schema file of a financial industry specification standard for generating a test case; under the condition that the Schema file is a first version Schema file, analyzing the first version Schema file to obtain first target field information; under the condition that the Schema file is a second version Schema file, comparing an old version Schema file with the second version Schema file, determining difference contents, and analyzing the difference contents to obtain first target field information; and generating a test case according to the first target field information. According to the invention, automatic generation of the corresponding test case after the financial industry specification standard is changed is realized.

A novel, simple, and low-cost oscillator solution that meets new industry standards and addresses the growing issue of GNSS vulnerability is described. The solution identifies the technologies and products that can provide protection against GNSS anomalies and outages. The solution is based on a simple low-cost oscillator with holdover capability using a self-disciplined circuit that supports future IoT devices and network.

This invention discloses a rapid assessment and precise repair method for the degree of pulverization in ballastless track bed concrete. By preparing concrete reference specimens of different strength grades, a correlation curve between pull-out strength and overall strength is established. Combined with probe diameter selection, surface grinding, and adhesive bonding techniques, a non-destructive and rapid detection of the pulverized layer is achieved. Specifically, the method includes: a preliminary preparation stage where specimens are prepared and a pull-out strength-strength model is established; an evaluation stage where layer-by-layer pull-out testing combined with precise hollow drilling quantitatively determines the depth and extent of pulverization; and a repair stage where the pulverized layer is replaced with a high-toughness material and a drainage slope is installed. This invention solves the problems of low efficiency, significant damage, and inaccurate repair associated with existing methods. Through parametric design and a modular system, it significantly improves detection accuracy and repair effectiveness. It is applicable to the maintenance of concrete structures such as high-speed railway track base plates, meets industry standards, and has the advantages of simple operation, low cost, and strong adaptability.

According to the personalized recommendation system and method for the digital media art CAD design, user operation behaviors are captured, periodic laws are mined in combination with time sequence analysis, user portraits are updated, color distribution and geometric morphology characteristics are extracted through OpenCV, a knowledge graph is generated in combination with a BERT + ResNet multi-mode learning model, and the personalized recommendation system and method for the digital media art CAD design are obtained. A relation network between nodes is expanded, a design trend is analyzed through a crawler technology and social network data, a reinforcement learning optimization strategy is adopted, a knowledge graph is updated, a logic dependency relationship between CAD design parameters is analyzed, finite element analysis results and industry standards are integrated, an updated feasibility graph is generated, and a design scheme is encoded into a gene sequence; generating a candidate scheme through crossover variation, calculating the style deviation degree between the recommendation scheme and the existing works, tracking the adoption rate and the modification frequency, and evaluating the recommendation strategy effect through an A / B test.

Card edge connectors that are reliable and robust against impacts by external forces while having narrower widths by industry standards. A connector includes a housing having a body with a slot elongated longitudinally. The housing has a pair of towers extending vertically from opposite ends of the longitudinal slot. Each tower has a slot elongated vertically, and is narrower than the body in a transverse direction. The longitudinal slot and the vertical slots are configured to receive edges of an electronic card. The connector has a pair of reinforcing members. Each reinforcing member is at least partially disposed on a tower and configured to support a joint between the tower and the body of the housing. Such a configuration can increase robustness of the connector at the joints and therefore reduce the risks of deforming or cracking the towers when the connector is impacted by an external force.

The embodiment of the invention provides a full-process intelligent software detection method and system based on a large language model, and belongs to the technical field of software testing. Comprising the following steps: based on a large language model learning industry standard and a historical document, carrying out comprehensive scanning and consistency checking on tested data; automatically extracting demand features and constructing a knowledge graph in the process of learning industry standards and historical documents by the large language model; analyzing the demand document through the natural language processing capability of the large language model; performing security testing on the software application API and the host according to the test case; identifying software code vulnerabilities according to a traditional scanning tool and a large model reasoning capability; and generating a standard test report based on the repair suggestion and the detection report. According to the full-process intelligent software detection method, a linked and synergistic detection system is built through a large language model, all links of software detection are comprehensively enabled, and full-process detection of software is achieved.

The invention provides an intelligent design and data management system for a pipeline support hanger, and relates to the technical field of industrial automatic design and data management, and the system comprises a standard database module, a size calculation module, a weight calculation module, a drawing generation module and an intelligent auxiliary module. The design period of the traditional pipeline support hanger needing 30 days is shortened, the efficiency is improved, the human input is reduced, and the cost is reduced. The size calculation module adopts a formula to realize automatic generation of the blanking size, the weight calculation module realizes batch calculation through a three-dimensional volume integral formula, compared with traditional manual calculation, the efficiency is improved, the problems of time consumption and error proneness of manual calculation are thoroughly solved, the precision and consistency are guaranteed through full-process automatic design, personal errors are eliminated, and the production efficiency is improved. The system records industry standards through the standard database module and updates the industry standards in real time, and standard reference errors are avoided.

A system and method for validating aviation regulatory compliance is disclosed. The system employs a method using AI to check aviation document compliance with regulations. It parses documents locally, compares semantically via hybrid AI, identifies gaps, offers remediation options from original texts, updates scores, and generates reports, ensuring efficiency and privacy.

The invention relates to the technical field of power system automation, and discloses a transformer substation protection information check report automatic generation method, system, equipment and medium, and the method comprises the steps: obtaining a fault recording file of a protection device, analyzing a fault report file, extracting protection motion information, generating a complete protection configuration matrix table based on a transformer substation configuration file, and storing the complete protection configuration matrix table. By analyzing pressing plate control words in a constant value list, identifying the on-off state of each protection function, marking the input protection in a table, intelligently matching the fault recording fluctuation condition in the acceptance period with a protection configuration matrix, automatically identifying the executed, non-executed and abnormally executed protection items, and adopting a natural language processing technology, the protection function of the protection function is automatically identified. And converting the technical data into text description conforming to industry specifications, and finally generating a formatted check report containing a complete protection configuration table. The completeness and accuracy of protection information checking are ensured, and the problem of low efficiency of a traditional manual checking mode is fundamentally solved.

A mirror plate defect repairing method for a large hydroelectric generator set, comprising the following steps: S1. performing flaw detection on the original defect part of the mirror plate to determine the defect position; S2. cleaning the defect; S3. performing flaw detection inspection on the cleaned area; S4. performing welding repair by using laser cladding technology; S5. polishing the welding repaired part to be flat and performing dye penetrant inspection; S6. after no defect is found in the flaw detection, performing heat treatment on the cladding area by using laser to adjust the hardness of the cladding area; and S7. performing honing processing on the working surface of the mirror plate. Through the method, the parallelism and roughness quality of the mirror plate after the repairing and finishing can meet the industry standards, and the repaired mirror plate can meet the safe and stable operation requirements. The service life of the mirror plate is prolonged, and compared with directly replacing the mirror plate, the method has great advantages, the repairing period is shorter, the repairing cost is lower, and great economic benefits are brought to early power generation of the generator set.

This invention discloses a gas pipe fitting with a self-sealing connection structure, relating to the field of gas pipe fittings. Addressing the problem of poor installation and gas leakage in existing small-diameter gas pipes during threaded connections due to the lack of industry standards for the use of necessary auxiliary sealing materials such as PTFE tape and sealant, the invention proposes the following solution: It includes a first pipe, with a second pipe at one end. The first pipe has a first hole along its axial direction, and the second pipe has a second hole along its axial direction. Sealing grooves are formed around the inner walls of the first and second pipes at their adjacent ends, within the first and second holes. This invention features a novel structure and effectively solves the problem of poor installation and gas leakage in existing small-diameter gas pipes during threaded connections due to the lack of industry standards for the use of necessary auxiliary sealing materials such as PTFE tape and sealant.

The utility model discloses a rapid demolding structure for an automobile mold, and belongs to the field of automobile mold production. The rapid demolding structure comprises an automobile bottom mold, a mold making cavity is formed in the upper surface of the automobile bottom mold, concave grooves are formed in the two sides, close to the mold making cavity, of the automobile bottom mold, and side fixing plates are fixedly installed on the surfaces, close to the concave grooves, of the mold making cavity; by means of secondary ejection of the jacking plate and the ejector pin column, in the rapid demolding process, the product quality is prone to being affected by demolding force, the secondary ejection structure can reasonably distribute the demolding force, the situation that the product is damaged due to single-time overlarge demolding force is avoided, small demolding force evenly acts on the product during primary ejection, the periphery of the product is slowly separated from a mold, and the product quality is improved. During secondary ejection, the ejection position and force are accurately adjusted according to the specific shape and structural characteristics of the product, so that the product is ensured to be stably and completely separated from the mold, the defects of scratches, deformation and the like on the surface of the automobile product are prevented, the product quality is ensured to meet the strict automobile industry standard, and the yield of the product is improved.

1. The name of the present design product: civil aviation industry standard technical review graphical user interface for electronic equipment. 2. The use of the present design product: for an electronic device. 3. The design points of the present design product: the content of the graphical user interface display in the screen. 4. The picture or photo that best indicates the design points: front view. 5. The use of the graphical user interface: for display in a civil aviation industry standard technical review device, when a civil aviation industry standard technical review expert conducts review, the work of reviewing and reviewing the drafting of standard documents, drafting of review opinions, and technical review voting needs to be completed in the civil aviation industry technical review device. When the technical review expert is invited to conduct standard technical review, enter the technical review system as shown in the front view; the technical review expert clicks any area of the front view, enters the technical review work interface, enters interface change state diagram 1; interface change state diagram 1 is a technical review work task display, the right side of the task bar shows the activation state of the current task through the traffic light, the green light indicates that the task is activated, and the red light indicates that the task is not activated. Click the right green button to enter interface change state diagram 2; interface change state diagram 2 is the review interface of the current review task, which can edit and add comments, delete comments, add markers, delete markers, save opinions, click "next step" below the interface, enter interface change state diagram 3; interface change state diagram 3 is a review opinion filling interface, which can add, delete and save opinions above the table, click "next step" below the page, enter interface change state diagram 4; interface change state diagram 4 is a standard review voting interface, after clicking "confirm and submit" in the upper left corner of the table, the voting button in the middle and lower part is visible, select "agree", "agree, with reservations", "disagree", "disagree, agree after adopting opinions" or "abstain", and then click "next step" to enter interface change state diagram 5; interface change state diagram 5 is a standard review opinion signature interface, the left side is the standard review opinion content formed by summarizing, for experts to browse and confirm; the upper half of the right side is edited by the expert's name, position / title, work unit and contact information, and the lower half provides a signature panel for the expert to sign. After clicking "confirm and sign", enter interface change state diagram 6; interface change state 6 is the end interface of the technical review expert after completing the current review task, if the technical review expert has no unfinished review task, it will stay on this interface; if the technical expert still has unfinished review tasks, it will return to interface change state 1.

This system improves the efficiency and accuracy of CAD drawing review in design work. [Solution] The server receives the uploaded design drawing data, and the server's AI analysis module decodes the drawing file and identifies each design element. The server compares the analyzed design elements with existing design standards and industry standards. Based on the comparison results, the server generates proposed corrections for inconsistencies. Using an AI model, it proposes the optimal correction method and creates specific correction proposals that clearly indicate the areas to be corrected.

The invention discloses a single-head anti-loosening fastening device and a fastening method. The single-head anti-loosening fastening device comprises a bolt and a bolt clamping screw, the non-overlapping part of the bolt head and the bolt body in the axial direction is provided with two or more through bolt clamping screw holes parallel to the axial direction, and the bolt clamping screw holes are used for being connected with bolt clamping screws in a threaded fit mode; the diameter of the bolt clamp tail of the bolt clamp screw shrinks towards the tail end, and the bolt clamp screw is embedded into a locking groove of a contacted fastening object and is bound by sequentially penetrating through the bolt clamp tail through an iron wire. On the premise that the single-head fastening device is convenient and reliable to install, the requirements of railway industry standards are met and exceeded, the loosening torque is larger than the fastening torque, mucilage glue does not need to be smeared, using is convenient, the single-head fastening device can be put into use immediately after fastening, and loosening and disassembling are easy, convenient and rapid compared with an existing fastening device.

The application discloses a rehabilitation auxiliary tool evaluation method based on multi-agent and RAG technology. The method comprises the following steps: constructing a multi-agent system, including a rehabilitation doctor agent, an auxiliary tool engineer agent, a nursing staff agent and a user interaction agent; building a dynamic knowledge base based on RAG technology, including an auxiliary tool product database, a rehabilitation medical knowledge base, a clinical case database and an industry standard database; interacting with the user, and dynamically adjusting the questioning method, language style and question order according to the user type and answer content; retrieving in the dynamic knowledge base according to the obtained information, and respectively outputting professional evaluation opinions; and integrating the professional evaluation opinions of the rehabilitation doctor agent, the auxiliary tool engineer agent and the nursing staff agent according to a collaborative decision-making model to generate a multi-dimensional auxiliary tool evaluation report and a recommended scheme. The above method can make the evaluation process of the auxiliary tool more efficient, and the evaluation result more accurate and effective.

The invention provides an automatic generation method and system for scenarized metadata endowed by a language large model, and belongs to the technical field of data processing. The method specifically comprises the steps that a data preprocessing module is responsible for generating business terms conforming to scene specifications and definitions of the business terms, and establishing a preliminary semantic association network among the business terms; the reasoning module is responsible for automatically deducing a calculation formula, a statistical dimension and a calculation period of the business terms based on the preliminary semantic association network, and further deriving derivative business knowledge of the business terms; the semantic mapping module is responsible for establishing a precise mapping relation between technical attribute information and business terms to form integrated metadata, the verification module is responsible for establishing a rule, semantic and logic three-layer verification system, and the generated metadata is compared with historical versions, industry standards and business rules by using a language large model for analysis to form a verification result. And the metadata passing the verification is published to the metadata warehouse, so that the application value of the extraction result of the metadata is improved.