A stamping process knowledge management method, device and electronic equipment

Abstract

The application provides a stamping process knowledge management method and device and electronic equipment, comprising: monitoring a first database of original files to be managed, the original files comprising stamping process knowledge of at least one information type; performing information extraction and structured processing on the stamping process knowledge of at least one information type in the original files to generate a first target file; and determining association relationship information of the first target file based on a domain vocabulary, the association relationship information indicating a second target file in a second database that has an association relationship with the first target file; and then storing the first target file and the association relationship information of the first target file to the second database, thereby automatically realizing structured and associated management of stamping process knowledge, and effectively avoiding the occurrence of the situation that the stamping process knowledge is scattered and fragmented, and the difficulty of process personnel in solving MR / TI problems is increased.

Description

Technical Field

[0001] This application relates to the field of automotive manufacturing technology, specifically to a stamping process knowledge management method, apparatus, and electronic equipment. Background Technology

[0002] With the rapid development of the automobile manufacturing industry and the increasing complexity of stamping processes, the amount of stamping process knowledge data is growing exponentially, and the construction of a stamping process knowledge base has become a core element in determining production efficiency.

[0003] However, the existing stamping process knowledge base only serves as a knowledge storage function. The stamping process knowledge is scattered across multiple heterogeneous carriers such as Excel process standard tables, PPT structural drawings, and on-site defect reports. The knowledge storage is scattered and highly fragmented, which greatly increases the difficulty for process personnel to read process knowledge to solve MR (product manufacturing) / TI (design manufacturing) problems. Summary of the Invention

[0004] Based on the aforementioned defects and shortcomings of the prior art, this application proposes a stamping process knowledge management method, device, and electronic equipment to automatically realize the structured and correlated management of stamping process knowledge, thereby avoiding the situation where the storage of stamping process knowledge in the prior art is scattered and highly fragmented, which increases the difficulty for process personnel to solve MR / TI problems.

[0005] The first aspect of this application provides a method for managing stamping process knowledge, including: Monitor the original files to be managed in the first database, the original files including stamping process knowledge of at least one information type; Information extraction and structured processing are performed on at least one type of stamping process knowledge in the original file to generate a first target file; The association information of the first target file is determined based on the domain thesaurus, and the association information indicates a second target file in the second database that has an association with the first target file; The first target file and its association information are stored in the second database.

[0006] In one possible implementation, the step of extracting and structuring the stamping process knowledge of at least one information type from the original file to generate the first target file includes: The text-type stamping process knowledge in the original document is subjected to text recognition and structured processing to obtain text recognition results. The text recognition results include at least one first data pair, which consists of process terminology and detailed content of process terminology. Extract the stamping process knowledge in the form of a table from the original file and perform structured processing to obtain the table recognition result. The table recognition result includes at least one second data pair, which consists of a field name and a field value. The field name is used to characterize the process term. The original document contains graphic-type stamping process knowledge. Graphic recognition and structuring are performed to obtain graphic recognition results. The graphic recognition results consist of process terms, process defect information and at least one third data pair. The third data pair includes structural feature name and structural feature size information. The text recognition results, the table recognition results, and the image recognition results are combined to generate a first target file. The first target file indicates the topic of the original file, and the topic of the original file is used to summarize the recognition results of the stamping process knowledge of at least one information type.

[0007] In one possible implementation, determining the association information of the first target file based on a domain thesaurus includes: A pre-set domain thesaurus is determined, the domain thesaurus including at least one type of process dictionary and the association relationships between different types of process terms in the at least one type of process dictionary; Extract keywords from the subject of the original file indicated by the first target file; Query the association relationships in the domain thesaurus to determine at least one candidate keyword that is associated with the keyword; The association information of the first target file is determined based on the keyword and the at least one candidate keyword.

[0008] In one possible implementation, determining the association information of the first target file based on the keyword and the at least one candidate keyword includes: Generate a keyword group for each of the at least one candidate keywords based on the keywords; Calculate the matching degree between the target file in the second database and the keyword group of each candidate keyword, and determine the highest matching degree as the matching degree of the target file; If the matching degree of the target file exceeds the preset target matching degree, the target file is determined to be a second target file that is associated with the first target file; Generate association information for the first target file, wherein the association information indicates the matching degree of the first target file, the keyword, the second target file, and the second target file.

[0009] In one possible implementation, the original files to be managed in the first database being monitored include: Monitor file update information in the first database, wherein the file update information indicates at least one original file that has been updated; Determine a file update sequence for the at least one original file, the file update sequence indicating the update priority of the at least one original file; Each of the original files indicated by the file update sequence is identified as the original file to be managed.

[0010] In one possible implementation, determining the file update sequence of the at least one original file includes: Obtain data information of the original file in at least one pre-set dimension, wherein the at least one dimension includes any one or more of the following: the degree of impact of the original file on production line efficiency, the severity of process defects indicated by the original file, and the update frequency of the original file. The target information of the original file is determined based on the data information of the original file in at least one dimension, and the target information represents the urgency of updating the original file. The at least one original file is sorted according to the target information of the original file to generate a file update sequence.

[0011] In one possible implementation, if the file update behavior of the original file is a file deletion behavior, the method further includes: Delete the association information between the first target file generated from the original file and the first target file in the second database.

[0012] One possible implementation also includes: Determine whether the first target file conforms to the process constraints of the domain thesaurus; If the first target file does not conform to the process constraints of the domain thesaurus, an early warning mechanism is triggered, which is used to indicate that the first target file is abnormal. The step of storing the first target file and its association information in the second database includes: if the first target file conforms to the process constraints of the domain thesaurus, storing the first target file and its association information in the second database.

[0013] A second aspect of this application provides a stamping process knowledge management device, comprising: A monitoring unit is used to monitor the original files to be managed in the first database, wherein the original files include stamping process knowledge of at least one information type; An information processing unit is used to extract and structure the stamping process knowledge of at least one information type in the original file to generate a first target file; An information association unit is used to determine the association information of the first target file based on a domain thesaurus, wherein the association information indicates a second target file in a second database that is associated with the first target file; An information storage unit is used to store the first target file and its association information in the second database.

[0014] A third aspect of this application provides an electronic device, including a memory and a processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the stamping process knowledge management method as described in the first aspect of this application or any possible implementation thereof by running a program in the memory.

[0015] The fourth aspect of this application provides a chip including a processor and a data interface, wherein the processor reads and runs a program stored in a memory through the data interface to execute the stamping process knowledge management method as described in the first aspect of this application or any possible implementation thereof.

[0016] The fifth aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the stamping process knowledge management method as described in the first aspect of this application or any possible implementation thereof.

[0017] The sixth aspect of this application provides a storage medium storing a computer program, which, when executed by a processor, implements the stamping process knowledge management method as described in the first aspect of this application or any possible implementation thereof.

[0018] This application provides a method, apparatus, and electronic device for managing stamping process knowledge, comprising: monitoring original files to be managed in a first database, the original files including stamping process knowledge of at least one information type; extracting and structuring the stamping process knowledge of at least one information type in the original files to generate a first target file; determining the association information of the first target file based on a domain thesaurus, the association information indicating a second target file in a second database that is associated with the first target file; and storing the first target file and the association information of the first target file in the second database. Thus, this application automatically realizes the structured and associated management of stamping process knowledge, effectively avoiding the situation in the prior art where the storage of stamping process knowledge is scattered and fragmented, leading to increased difficulty for process engineers to solve MR / TI problems. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the architecture of a stamping process knowledge management system provided in an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of a stamping process knowledge management method provided in an embodiment of this application.

[0022] Figure 3 This application provides a flowchart of a method for extracting and structuring stamping process knowledge of at least one information type from an original document to generate a first target document, as provided in an embodiment of the present application.

[0023] Figure 4 A flowchart illustrating a method for determining the association information of a first target file based on a domain thesaurus, provided in an embodiment of this application.

[0024] Figure 5 This is a schematic diagram of the structure of a stamping process knowledge management device provided in an embodiment of this application.

[0025] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Exemplary methods In the field of automotive manufacturing stamping process knowledge management, to achieve the systematic capture, organization, and real-time utilization of scattered and heterogeneous process knowledge, the commonly used approach is to build a semi-automated knowledge base management system. This system typically includes a document import module, an optical character recognition (OCR) module, a manual proofreading module, a tag assignment module, and a structured database. Its basic working principle is to extract data from uploaded (imported) text or tables using OCR technology, then have engineers verify the extraction results in a manual proofreading interface, assigning preset process tags to the data, and finally storing the tagged structured data in the database. This solution is widely adopted primarily to automatically convert unstructured process documents (such as reports and standard tables) into queryable data, thereby obtaining more comprehensive process knowledge information and improving the efficiency of knowledge retrieval.

[0028] However, this solution performs poorly when applied to process documents in modern automotive stamping processes, which are highly complex and varied in source, format, and content type. A fundamental contradiction lies in the fact that, in order to optimize the recognition accuracy and processing efficiency of single-type information (such as plain text), the inherent design of this solution inevitably compromises its ability to integrate, autonomously correlate, and synchronize multi-source heterogeneous knowledge in real time. For example, when processing a comprehensive process document that simultaneously contains textual process descriptions, tabular parameters, and graphical structural diagrams, the aforementioned contradiction leads to fragmented knowledge extraction, missing knowledge connections, and delayed knowledge updates. Specifically, the optical character recognition module can only extract textual content from text and tables, and is powerless to handle dimension annotations and structural features in graphics, resulting in the loss of key graphical knowledge; the tag assignment module can only achieve a simple mapping between documents and fixed tags, and cannot automatically establish a deep cross-document association between "a specific defect of a part" and "key parameters of related processes"; the update of the entire knowledge base relies entirely on engineers manually discovering document changes and re-executing the entire process, with update cycles lasting tens of hours, which cannot meet the real-time needs of the production line for the latest process knowledge.

[0029] Through in-depth analysis, the inventors discovered that the root cause of the aforementioned contradictions lies in the lack of a closed-loop automated management perspective for the knowledge lifecycle in the design of the relevant technologies. Their various functional modules operate in isolation and sequentially, failing to form a complete closed loop from document monitoring, content parsing, semantic understanding to association construction, database storage, and synchronous updates. In particular, it lacks the ability to collaboratively parse multimodal information (text, tables, graphics), the ability to autonomously construct knowledge networks based on domain semantics, and the ability to proactively perceive and respond to changes in knowledge sources.

[0030] To overcome the aforementioned contradictions, this invention proposes a different technical approach. Its core concept lies in constructing a closed-loop knowledge management system that integrates proactive monitoring, unified extraction, semantic association, and automatic updating capabilities. This system reconstructs the entire process of knowledge generation and application, thereby significantly improving the completeness of knowledge acquisition, the relevance of knowledge organization, and the freshness of the knowledge base without significantly sacrificing the accuracy of single-type information processing. Specifically, it provides a method, device, and system for managing stamping process knowledge to solve the problems of knowledge fragmentation and lag caused by isolated processing of multi-source heterogeneous information, reliance on manual knowledge association, and passive triggering of update processes. This achieves the technical effect of transforming scattered, static process knowledge into a structured knowledge body that is relevanceable and usable in real time.

[0031] Figure 1 This is a schematic diagram of the architecture of a stamping process knowledge management system provided in an embodiment of this application. Figure 1 As shown, the stamping process knowledge management system includes a data layer, a core function layer, and an application layer.

[0032] The data layer includes a raw document repository, a structured database, and a domain thesaurus. The raw document repository stores various original documents to be managed for knowledge management, such as process standard tables from the process department, structural drawings from the design department, and defect reports from the production site. The structured database stores processed structured knowledge and its relationships. The core functional layer includes core processing units such as an information extraction module, a semantic association engine, an automatic update mechanism, and a graphical process information parsing unit, which are crucial for implementing the stamping process knowledge management method provided in this application embodiment. The application layer provides an engineer's operating interface, a management interface, and external interfaces for interfacing with other systems (such as MES (Manufacturing Execution System)). In actual operation, raw documents are stored in the raw document repository of the data layer through the external interfaces or operating interfaces of the application layer; the modules of the core functional layer work collaboratively to process the original documents in the raw document repository and write the results back to the structured database of the data layer; the various interfaces of the application layer are used to query and present knowledge from the structured database or receive data input from external systems to trigger the update process.

[0033] In this application, the term "original document" refers to any electronic document or data file containing knowledge related to the stamping process. This may include, but is not limited to: process parameter standard tables, mold structure drawings and process plan descriptions, defect report documents containing text descriptions and on-site photographs, or data reports directly exported from a manufacturing execution system. Each file may contain stamping process knowledge of one or more information types, such as text, tables, and graphics.

[0034] Figure 2 This application provides a schematic flowchart of a stamping process knowledge management method, which can be implemented by, for example... Figure 1 The core functional layer shown is executed.

[0035] like Figure 2 As shown, the method includes: Step S210: Monitor the original files to be managed in the first database. The original files include stamping process knowledge of at least one information type.

[0036] In this embodiment, the first database is... Figure 1 The original document repository in the data layer stores files called source files. The purpose of monitoring the source files to be managed in the first database is to proactively discover new files, delete files, or modify files (existing files in the first database whose content has changed) that require knowledge management in the first database.

[0037] As a specific implementation method, monitoring can be achieved by periodically scanning the file system directory of the first database and checking the last modification timestamp or file version number of the files. For example, if the file system directory is scanned every 5 minutes, and a file is found to have a last modification time later than the time it was last managed, then the file is determined to be an "original file to be managed".

[0038] More generally, monitoring can be achieved in various ways. For example, it can include, but is not limited to: utilizing the operating system's file system event listeners to capture file creation, modification, and deletion events in real time; or calculating the hash value of file content and accurately identifying changes in file content by comparing changes in the hash value. All these methods can achieve the function of monitoring the original files to be managed in the primary database.

[0039] It should be noted that the implementation of the above monitoring behavior provides the necessary prerequisite for subsequent information extraction and knowledge update triggering. It solves the passive mode of knowledge update that relies entirely on manual triggering in related technologies, and makes database maintenance shift from "passive response" to "active perception", thereby achieving the overall technical effect of improving the real-time nature of knowledge.

[0040] Step S220: Extract and structure the stamping process knowledge of at least one information type from the original file to generate the first target file.

[0041] This step aims to address the issues of heterogeneous and unstructured information in the original files, providing a high-quality and standardized data foundation for knowledge association construction.

[0042] First, the original files to be managed are read from the original document repository. Then, the stamping process knowledge of various information types is judged and processed in parallel or sequentially. Specifically, if the original file contains text-based stamping process knowledge, the text parsing unit in the information extraction module is activated for processing. The text parsing unit has a built-in optical character recognition engine that can recognize text; it also includes a process term filtering subunit to filter out non-process text such as headers, footers, and irrelevant descriptions, while retaining process text such as "A-pillar springback compensation scheme" and "rear door outer panel tearing defect and compensation method"; finally, the first data pair consisting of process terms and detailed content of the process terms is output, for example, defect solution—side panel OP10 slip line defect solution.

[0043] If the original file contains tabular stamping process knowledge, the table parsing unit in the information extraction module is activated. The table parsing unit contains a pre-defined stamping process table template library. These templates define the field structure of common process tables (such as material parameter tables and process standard tables). The field matching subunit matches the table structure identified from the tabular stamping process knowledge with the templates to determine the process field names (i.e., process terms) corresponding to each column of data in the table structure. The data validation subunit performs a rationality check on the column data in the extracted table structure (such as unit conversion and initial range screening). Finally, a second data pair consisting of the field name (process term) and the field value information is output. For example, process number—OP10, pressure parameter—1200 kN, die clearance—0.8 mm.

[0044] If the original file contains graphic stamping process knowledge, such as drawings, diagrams, or photographs, the graphic process information parsing unit is activated. The graphic process information parsing unit first preprocesses the graphic stamping process knowledge, including noise reduction, edge detection, and image enhancement, to improve subsequent recognition accuracy. Then, based on a pre-trained process graphic recognition model, it identifies process terms (e.g., flange), structural features (e.g., flange face width), and process defect information (e.g., wrinkled areas) within the graphic. Simultaneously, it extracts dimension annotations and units from the graphic using optical character recognition technology. These recognition results are then correlated and encapsulated, outputting a third data pair consisting of process terms, process defect information, and structural feature names and dimensions.

[0045] For example, the third data pair is flange face width - 15mm, where flange face width is a structural feature with a dimension of "15mm".

[0046] For example, the process graphic recognition model identifies the process defect area corresponding to the defect annotation symbol "△" in the stamping process knowledge of the graphic type, and determines the process defect name corresponding to the process defect area by associating it with the defect dictionary, such as "wrinkled area".

[0047] Finally, the data fusion unit receives the output results from the text parsing, table parsing, and graphical process information parsing units, and integrates and encapsulates all the recognition results according to the format of "document identifier-information type-information type recognition result" to generate a unified structured file, namely the first target file.

[0048] Furthermore, the first target file not only includes the identification results of stamping process knowledge of various information types extracted from the original document, but also includes a theme for summarizing the identification results of the stamping process knowledge in the original document. That is, the first target file consists of document identifier-theme-information type-information type identification results. For example, the theme is "Solution to the OP10 slip line defect on the side panel".

[0049] Through the integrated extraction and structuring process described above, this step transforms the fragmented knowledge of scattered text, tables, and graphics in the original file into a structured knowledge record (the first target file) with a clear theme and unified format, laying a solid data foundation for subsequent semantic-based association construction. For detailed information on step S220, please refer to [link to relevant documentation]. Figure 3 This will not be elaborated upon here.

[0050] Step S230: Determine the association information of the first target file based on the domain thesaurus. The association information indicates the second target file in the second database that has an association with the first target file.

[0051] The second database is Figure 1 The structured database in the data layer already stores structured knowledge records generated from historical source files (i.e., the historical first target file, which may become the second target file in this step). It should be noted that after the first target file is stored in the second database, it can be referred to as a target file. Step S230 aims to automatically establish a connection between the currently generated knowledge record (first target file) and existing knowledge records (target files) in the second database, thus resolving the knowledge silo problem.

[0052] Figure 4 This document provides a flowchart of a method for determining the association information of a first target file based on a domain thesaurus, as illustrated in an embodiment of this application. Figure 4 As shown, the method includes: S410. Determine a pre-set domain thesaurus, which includes at least one type of process dictionary and the relationships between different types of process terms in the at least one type of process dictionary; In this embodiment of the application, the domain thesaurus is maintained by a domain thesaurus management subunit, and the domain thesaurus includes at least one type of process dictionary and the association between different types of process terms in the at least one type of process dictionary.

[0053] For example, at least one type of process dictionary includes a part dictionary (such as side panel, door inner panel), a process dictionary (such as OP10, drawing, OP20 trimming), a defect dictionary (such as slip line, wrinkling, cracking), and a parameter dictionary (such as pressure, speed, clearance).

[0054] In the embodiments of this application, the structure of the association between different types of process terms in at least one type of process dictionary includes: part—process—defect—parameter.

[0055] For example, a relationship might be described as: side panel—OP10—slip line—pressure parameters.

[0056] S420. Extract keywords from the subject of the original file indicated by the first target file; In this embodiment, the keyword extraction subunit identifies and extracts the process entity words that best represent the core content from the subject of the original document indicated by the first target document as keywords in the subject of the original document. For example, from the subject "Side OP10 Slip Line Defect Solution", the keywords "side OP10" and "slip line" are extracted.

[0057] S430. Query the association relationships in the domain thesaurus to determine at least one candidate keyword that is associated with the keyword; In this embodiment of the application, the association matching subunit matches the extracted keywords with the associations in the domain thesaurus to determine at least one candidate keyword that is associated with the keyword, so as to map the semantic content of the original document topic to the relevant concepts in the domain knowledge system.

[0058] As a preferred implementation provided in this application, the target association is queried from the association relationship of the domain thesaurus. The target association includes all keywords extracted from the original document topic indicated by the first target document. Then, each word in the target association other than the keywords in the original document topic indicated by the first target document is regarded as a candidate keyword.

[0059] For example, if the domain terminology stores two associations, one being sidewall-OP10-slip line-pressure parameter and the other being sidewall-OP10-slip line-drawing speed, then querying the associations in the domain terminology will identify at least one candidate keyword that is associated with the keyword, including two candidate keywords: pressure parameter and drawing speed.

[0060] S440. Determine the association information of the first target document based on the keywords and at least one candidate keyword.

[0061] In this embodiment of the application, determining the association information of the first target file based on the keyword and at least one candidate keyword includes: generating a keyword group for each candidate keyword in the at least one candidate keyword; calculating the matching degree between the target file in the second database and the keyword group of each candidate keyword, and determining the highest matching degree as the matching degree of the target file; if the matching degree of the target file exceeds the preset target matching degree, determining the target file as a second target file that is associated with the first target file; and generating association information of the first target file, wherein the association information indicates the matching degree of the first target file, the keyword, the second target file, and the second target file.

[0062] In this embodiment, for each of the at least one candidate keywords, the following process is performed: the candidate keyword is combined with keywords in the subject of the original document indicated by the first target document to generate a keyword group for that candidate keyword. Thus, a keyword group with a more specific query intent can be formed for each candidate keyword.

[0063] For example, if the keywords in the subject of the original file indicated by the first target file include sidewall, OP10, and slip line, and at least one candidate keyword includes pressure parameter and drawing speed, then keyword groups for pressure parameter and keyword groups for drawing speed are generated respectively. Specifically, the keyword group for pressure parameter includes sidewall, OP10, slip line, and pressure parameter; the keyword group for drawing speed includes sidewall, OP10, slip line, and drawing speed.

[0064] In this embodiment of the application, each file stored in the second database can be referred to as a target file. For each target file in the second database, the matching degree between the target file and the keyword group of each candidate keyword is calculated, and the highest matching degree is determined as the matching degree of the target file. The matching degree of the target file can be understood as the matching degree between the target file and the first target file.

[0065] For example, if the second database stores three target files, namely target file A, target file B, and target file C, at least one candidate keyword associated with the keywords (sidewall, OP10, and slip line) of the subject in the original file indicated by the first target file includes pressure parameter and drawing speed. The keyword group for pressure parameter (keyword group 1) includes sidewall, OP10, slip line, and pressure parameter, and the keyword group for drawing speed (keyword group 2) includes sidewall, OP10, slip line, and drawing speed; calculate the matching degree A1 between target file A and keyword group 1. The matching degree A2 between target file A and keyword group 2 is calculated, and the higher of matching degrees A1 and A2 is determined as the matching degree of target file A. Similarly, the matching degree B1 between target file B and keyword group 1, and the matching degree B2 between target file B and keyword group 2 are calculated, and the higher of matching degrees B1 and B2 is determined as the matching degree of target file B. Likewise, the matching degree C1 between target file C and keyword group 1, and the matching degree C2 between target file C and keyword group 2 are calculated, and the higher of matching degrees C1 and C2 is determined as the matching degree of target file C. Here, the matching degree of target file A can be considered as the matching degree between target file A and the first target file, the matching degree of target file B can be considered as the matching degree between target file B and the first target file, and the matching degree of target file C can be considered as the matching degree between target file C and the first target file.

[0066] In this embodiment, a target matching degree is preset. If the matching degree of the target file exceeds the preset target matching degree, it is determined that the target file has an association with the first target file. In this case, the target file can be referred to as the second target file that has an association with the first target file. If the matching degree of the target file does not exceed the preset target matching degree, it is determined that the target file does not have an association with the first target file.

[0067] For example, the target matching degree can be 85%. The above is merely a preferred value for the target matching degree provided by the embodiments of this application. Those skilled in the art can set the specific value of the target matching degree according to their own needs, and it is not limited here.

[0068] In this embodiment, the matching degree between the target file and the keyword group can be calculated by using a vector space model to calculate the cosine similarity between the keyword group and the file content of the target file, and then determining the cosine similarity as the matching degree between the target file and the keyword group. The above is merely a preferred method for calculating the matching degree between the target file and the keyword group provided by this embodiment. Those skilled in the art can set the specific calculation method according to their needs, and it is not limited here.

[0069] In this embodiment, the association information of the first target file indicates the first target file, the keywords in the original file topic corresponding to the first target file, the second target file in the second database that is associated with the first target file, and the matching degree of the second target file. It can be understood that the matching degree of the second target file is the matching degree between the second target file and the first target file.

[0070] Through the above-mentioned automated knowledge association construction based on domain semantics, new knowledge records (first target files) can be automatically "woven" into the existing knowledge network (second database), realizing the leap of knowledge from points to networks and greatly enhancing the discoverability and usability of knowledge.

[0071] Step S240: Store the first target file and its association information in the second database.

[0072] In this embodiment, the first target file and its association information are stored in a second database, realizing the structured and associated storage of stamping process knowledge in the original file. The first target file stored in the second database can be called the target file. The query interface or recommendation engine of the application layer can provide process engineers with accurate knowledge retrieval and association recommendation services based on the structured stamping process knowledge (target file) and the association information of the structured stamping process knowledge stored in the second database. This effectively avoids the situation where the storage of stamping process knowledge in the prior art is scattered and fragmented, which increases the difficulty for process engineers to solve MR / TI problems.

[0073] It should be noted that steps S210 to S240 constitute a complete knowledge management closed loop. Step S210 enables proactive awareness of the original documents to be managed; the high-quality structured data provided by step S220 forms the basis for accurate semantic association in step S230; and the rich relationships established in step S230 greatly enhance the value and usability of the knowledge stored in step S240. These four elements work together to solve the fundamental problem of fragmented and isolated knowledge in stamping processes.

[0074] Figure 3 This application provides a flowchart of a method for extracting and structuring stamping process knowledge of at least one information type from an original document to generate a first target document, as provided in this embodiment. Figure 3 As shown, the method includes: S310. Perform text recognition and structured processing on the text-type stamping process knowledge in the original document to obtain the text recognition result. The text recognition result includes at least one first data pair, which consists of process terminology and detailed content of process terminology. In this embodiment, the text recognition and structuring of the stamping process knowledge in the original document are specifically defined as obtaining a result that includes at least one first data pair, where the first data pair consists of a process term and its detailed content. This limitation aims to clarify the output format of the structured text information, making it easier to align and integrate with structured data from other information types.

[0075] Specifically, in addition to optical character recognition, the text parsing unit can also, based on a basic list of stamping process keywords, filter out process-related sentences or phrases from all the identified text through string matching or regular expressions, and break them down into "attribute (process terminology) - value (detailed content of process terminology)" pairs. For example, defect solution - side panel OP10 slip line defect solution.

[0076] Understandably, if the text-type stamping process knowledge includes detailed information on the solution to the OP10 slip line defect in the side panel, then the process details in the generated first data pair will include detailed information on the solution to the OP10 slip line defect. This approach transforms unstructured text descriptions into standard, queryable data entries, facilitating subsequent data fusion and related queries.

[0077] S320. Extract the stamping process knowledge of the table type from the original file and perform structured processing to obtain the table recognition result. The table recognition result includes at least one second data pair. The second data pair consists of field name and field value information. The field name is used to characterize the process term. In this embodiment, the table-type stamping process knowledge extracted from the original file and structured to obtain the table recognition result is specifically defined as obtaining a result including at least one second data pair. The second data pair consists of field names and field value information, and the field names are used to characterize process terms. This limitation emphasizes that the purpose of table parsing is not only to extract data, but more importantly, to understand the semantics of the data (i.e., the process meaning represented by the field names).

[0078] As one implementation of this application: each template in the table template library explicitly defines the standard process field name corresponding to each column of the table; the field matching subunit analyzes the table header row, cell merging, and content features to calculate the similarity with the template and find the most matching template; thus, column headers that may be abbreviations or aliases in the original document table are mapped to standard process terms (field names). For example, if the column header in the original document table is "Press", it is matched and mapped to the field name "Pressure Parameter".

[0079] In this embodiment, the field numerical information includes the field value and the unit of measurement. For example, the second data pair is a pressure parameter—1200 kN, where the pressure parameter is the field name and 1200 kN is the field numerical information. This achieves semantic representation of the tabular data, unifying the meaning of tabular data from different sources and in different formats, greatly facilitating subsequent comparative analysis and correlation calculations.

[0080] S330. Perform graphic recognition and structuring on the stamping process knowledge of graphic type in the original file to obtain graphic recognition results. The graphic recognition results consist of process terms, process defect information and at least one third data pair. The third data pair includes the structural feature name and structural feature size information. In this embodiment, the stamping process knowledge of graphic types in the original document is subjected to graphic recognition and structuring processing to obtain graphic recognition results. Specifically, the results are defined as consisting of process terms, process defect information, and at least one third data pair. The third data pair includes the structural feature name and structural feature dimension information. This limitation is a specific design made for the characteristics of stamping process graphics, aiming to fully capture the geometric information and process semantics in the graphics. The principle is that simple dimensional values ​​have no process meaning and must be bound to the structural feature name they describe.

[0081] Specifically, the graphical process information parsing unit, based on a process graphic recognition model (such as a convolutional neural network), identifies specific process structures in the graphic, such as "punch fillet," "draw bead," and "flange," which are process terms. Through optical character recognition and geometric relationship analysis, it extracts the dimensional numbers and units labeled on the graphic and determines which structural feature each label points to. Then, it matches each identified structural feature with its corresponding dimensional information to form a third data pair, for example, flange face width -15mm. Simultaneously, if the graphic is a defect diagram, the graphical process information parsing unit further identifies defect annotation symbols on the graphic and maps them to entries in a defect dictionary, outputting process defect information, such as "wrinkled area."

[0082] The embodiments of this application can overcome the gaps in related technologies regarding graphic information processing, realize the accurate analysis and structured storage of graphical process knowledge, and effectively avoid the loss of key knowledge.

[0083] S340. The first target file is generated by integrating the text recognition results, table recognition results and graphic recognition results. The first target file indicates the topic of the original file. The topic of the original file is used to summarize the recognition results of stamping process knowledge of at least one information type.

[0084] In this embodiment, a first target file is generated by fusing text recognition results, table recognition results, and image recognition results. The first target file indicates the theme of the original document, and the theme summarizes the recognition results of at least one type of stamping process knowledge. This limitation aims to provide a high-level, condensed semantic identifier for subsequent semantic association. When integrating the recognition results of various information types in the original document, the data fusion unit generates the theme of the original document based on these results. For example, a pre-trained lightweight text generation model can generate a summary statement based on the recognition results of various information types in the original document. This generated summary statement can be considered the theme of the original document, and this theme will become the core semantic tag of the first target file. By adopting the above specific limitations, this embodiment not only completes the physical integration of data but also achieves the semantic aggregation of information, generating a self-descriptive knowledge object, thus creating conditions for accurate content-based association and retrieval.

[0085] In this embodiment, monitoring the original files to be managed in the first database can be implemented in various ways. To further optimize the monitoring mechanism of the first database in the above embodiments and improve resource scheduling efficiency and response timeliness when facing concurrent updates, in a preferred embodiment, this step is specified as an intelligent process that includes monitoring and priority scheduling.

[0086] As a preferred embodiment of this application, monitoring the original files to be managed in the first database includes: monitoring file update information in the first database, wherein the file update information indicates at least one original file that has been updated; determining the file update sequence of at least one original file, wherein the file update sequence indicates the update priority of at least one original file; and sequentially determining each original file indicated by the file update sequence as the original file to be managed.

[0087] In this embodiment, file update behavior includes file addition behavior, file modification behavior, and file deletion behavior.

[0088] If the original file is a newly added file in the first database, then the original file is determined to have been updated; if the original file is deleted from the first database, then the original file is determined to have been updated; if the content of the original file in the first database is modified, then the original file is determined to have been modified and is therefore considered to have been updated.

[0089] It should be noted that if the file update behavior of the original file to be managed in step S210 is a file addition behavior, then step S240 directly stores the first target file and the association information of the first target file corresponding to the original file into the second database; if the file update behavior of the original file to be managed in step S210 is a file modification behavior, then step S240 first deletes the target file and the association information of the target file corresponding to the original file that has been stored in the second database, and then stores the first target file and the association information of the first target file corresponding to the currently generated original file into the second database.

[0090] Furthermore, the stamping process knowledge management method provided in this application embodiment also includes: if the file update behavior of the original file is a file deletion behavior, the association information of the first target file generated based on the original file and the first target file generated in the second database is deleted.

[0091] In this embodiment of the application, if the file update behavior of the original file to be managed in step S210 is a file deletion behavior, then the target file generated based on the original file and the association information of the target file stored in the second database will be deleted.

[0092] This application not only improves the closed-loop management of stamping process knowledge "addition, deletion, and modification" and automatically maintains the synchronization between the second database and the first database, but also effectively avoids the problem that the corresponding knowledge in the second database will be invalid due to the failure of the original files in the first database, thus ensuring the accuracy and cleanliness of the second database and reducing the burden of manual maintenance.

[0093] It is understood that the embodiments of this application can monitor file update information in the first database in real time, or monitor file update information in the first database according to a preset monitoring period.

[0094] For example, the monitoring period can be 2 hours, 5 hours, or 1 day, etc. Those skilled in the art can set the monitoring period according to their own needs, and there is no limitation here.

[0095] In this embodiment, the presence of file modification can be monitored by calculating the hash value of the original file's content. For example, in the current monitoring period, the first content hash value of the original file in the first database is calculated, and it is determined whether the first content hash value is consistent with the second content hash value of the original file in the previous monitoring period. If the first content hash value is consistent with the second content hash value, it indicates that the content of the original file has not been modified in the current monitoring period (i.e., there is no file modification in the current monitoring period); if the first content hash value is inconsistent with the second content hash value, it indicates that the content of the original file has been modified in the current monitoring period (i.e., there is file modification in the current monitoring period).

[0096] In this embodiment, determining the file update sequence of at least one original file means not simply managing files according to the time order of their updates (first-in, first-out), but rather sorting the management tasks according to a certain strategy. For example, based on factors such as the importance and urgency of the original files, an update priority score is calculated for each original file, and the at least one original file is sorted in descending order of update priority score to form a file update sequence; then, each original file in the at least one original file is managed sequentially according to the file sequence.

[0097] The advantage of this design is that it ensures that important and urgent knowledge changes can be prioritized for management and updates to the second database, thereby optimizing the allocation of system resources and avoiding problems such as processing congestion or low-priority tasks blocking high-priority tasks during peak update request periods, thus improving the efficiency and stability of the entire knowledge management process.

[0098] In this embodiment of the application, determining the file update sequence of at least one original file can be further specified as a decision-making process based on multi-dimensional quantitative evaluation.

[0099] As a preferred embodiment of this application, determining a file update sequence for at least one original file includes: acquiring data information of the original file in at least one pre-set dimension, wherein the at least one dimension includes any one or more of the following: the degree of impact of the original file on production line efficiency, the severity of process defects indicated by the original file, and the update frequency of the original file; determining target information of the original file based on the data information of the original file in at least one dimension, wherein the target information characterizes the urgency of updating the original file; and sorting at least one original file according to the target information of the original file to generate a file update sequence.

[0100] For example, the impact of the original document on production line efficiency is pre-defined into three levels: low, medium, and high, with different levels corresponding to different data information. For instance, if the original document's impact on production line efficiency is low, its impact value is 1; if it's medium, the impact value is 3; and if it's high, the impact value is 6. The impact value of the original document on production line efficiency can be considered as the data information of the original document in the "impact of the original document on production line efficiency" dimension.

[0101] The severity of process defects indicated in the original document is pre-set to three levels: low, medium, and high, with different levels corresponding to different data information. For example, if the severity of the process defect indicated in the original document is low, the severity value is 1; if it is medium, the severity value is 3; and if it is high, the severity value is 6. The severity value indicated in the original document can be considered as the data information in the "severity of process defects indicated in the original document" dimension.

[0102] The update frequency of the original file is pre-set to three levels: low, medium, and high, with different levels corresponding to different data information. For example, if the update frequency of the original file is low, the update frequency value is 1; if the update frequency of the original file is medium, the update frequency value is 3; and if the update frequency of the original file is high, the update frequency value is 6. The update frequency value of the original file can be considered as the data information of the original file in the "update frequency" dimension.

[0103] The above are merely preferred methods for taking values ​​for data information in various dimensions provided in the embodiments of this application. Those skilled in the art can set the specific methods for taking values ​​for data information in various dimensions according to their own needs, and no limitations are imposed here.

[0104] In this embodiment, a weight for each dimension in at least one dimension is pre-set. The weights of different dimensions can be the same or different; the sum of the weights of all dimensions in at least one dimension is 1.

[0105] Taking at least one dimension including the degree of impact of the original document on production line efficiency, the severity of process defects indicated by the original document, and the update frequency of the original document as an example, the target information of the original document is determined based on the data information of the original document in at least one dimension, including: determining the target information of the original document by summing the first information, the second information, and the third information, wherein the first information is the product of the weight of the degree of impact of the original document on production line efficiency and the value of the degree of impact of the original document on production line efficiency, the second information is the product of the weight of the severity of process defects indicated by the original document and the value of the severity of process defects indicated by the original document, and the third information is the product of the weight of the update frequency of the original document and the value of the update frequency of the original document.

[0106] In this embodiment of the application, the target information of the original file can be regarded as the update priority score of the original file. By sorting the update priority scores of at least one original file, the scheduling of knowledge management tasks is closely linked to the actual production value, realizing the optimal configuration of system processing capacity and ensuring that the most important process knowledge changes can be responded to at the fastest speed.

[0107] To introduce quality control into the automated process and prevent erroneous data from contaminating the second database due to errors in the original document or deviations in information extraction, the stamping process knowledge management method provided in this application, after generating the first target file and before storing it in the second database, further includes: determining whether the first target file conforms to the process constraints of the domain thesaurus; if the first target file does not conform to the process constraints of the domain thesaurus, triggering an early warning mechanism to indicate that the first target file is abnormal; correspondingly, storing the first target file and its association information in the second database, including: if the first target file conforms to the process constraints of the domain thesaurus, storing the first target file and its association information in the second database.

[0108] In this embodiment of the application, it is determined whether the first target file conforms to the process constraints of the domain thesaurus; if it does not conform, an early warning mechanism is triggered; the first target file and the association information of the first target file are stored in the second database only if the first target file conforms to the process constraints.

[0109] For example, determining whether the first target file conforms to the process constraints of the domain thesaurus includes: comparing the file content of the first target file with the process constraints in the domain thesaurus. For instance, the parameter dictionary of the domain thesaurus not only defines parameter names but may also define their reasonable value ranges or units. The determination of whether the parameter information in the first target file conforms to the relevant definitions in the parameter dictionary is made. If it does not conform, it means that the first target file does not conform to the process constraints of the domain thesaurus. Alternatively, it can be checked whether the process terms (such as part names and operation numbers) in the first target file exist in the domain thesaurus. If the first target file includes process terms that do not exist in the domain thesaurus, it means that the first target file does not conform to the process constraints of the domain thesaurus.

[0110] If the first target file does not conform to the process constraints of the domain thesaurus, the system will trigger an early warning mechanism. This mechanism can include a prominent pop-up notification on the application-layer engineer's interface, along with a detailed anomaly report, and may even integrate SMS or email notifications to ensure relevant personnel are promptly informed. Simultaneously, the system will block the direct storage of the first target file and its associated information into the second database, awaiting manual intervention for verification. Manual intervention can correct the original file / first target file, or, if the update is deemed special, force its approval.

[0111] If the first target file fully conforms to the process constraints of the domain thesaurus, then the operation of "storing the first target file and the association information of the first target file to the second database" is executed.

[0112] By adopting the above specific limitations, this application embodiment, while pursuing automation efficiency, introduces an automated quality inspection checkpoint based on domain knowledge, which can effectively intercept obviously illogical abnormal data, reduce the risk of erroneous knowledge entering the database at the source, improve the reliability and authority of the database, and achieve a good balance between automation and reliability.

[0113] Exemplary device Accordingly, this application also provides a stamping process knowledge management device.

[0114] Please see Figure 5 In one exemplary embodiment, a stamping process knowledge management device 500 is provided, the stamping process knowledge management device 500 including: The monitoring unit 510 is used to monitor the original files to be managed in the first database, which include stamping process knowledge of at least one type of information. The information processing unit 520 is used to extract and structure the stamping process knowledge of at least one type of information in the original file to generate a first target file; The information association unit 530 is used to determine the association information of the first target file based on the domain thesaurus, and the association information indicates the second target file in the second database that has an association relationship with the first target file; The information storage unit 540 is used to store the first target file and the association information of the first target file into the second database.

[0115] In one possible implementation, the information processing unit includes: The text parsing unit is used to perform text recognition and structured processing on the stamping process knowledge of the text type in the original document to obtain the text recognition result. The text recognition result includes at least one first data pair, which consists of process terminology and detailed content of process terminology. The table parsing unit is used to extract the stamping process knowledge of the table type in the original file and perform structured processing to obtain the table recognition result. The table recognition result includes at least one second data pair, which consists of field name and field value information. The field name is used to characterize the process term. The graphical process information parsing unit is used to perform graphical recognition and structural processing on the stamping process knowledge of graphical type in the original file to obtain graphical recognition results. The graphical recognition results consist of process terms, process defect information and at least one third data pair. The third data pair includes the structural feature name and structural feature size information. The data fusion unit is used to fuse text recognition results, table recognition results, and graphic recognition results to generate a first target file. The first target file indicates the topic of the original file, and the topic of the original file is used to summarize the recognition results of stamping process knowledge of at least one information type.

[0116] In one possible implementation, the information association unit includes: The domain thesaurus determination unit is used to determine a pre-set domain thesaurus, which includes at least one type of process dictionary and the relationship between different types of process terms in the at least one type of process dictionary; The keyword extraction unit is used to extract keywords from the subject of the original file indicated by the first target file; The candidate keyword determination unit is used to query the association relationships in the domain thesaurus and determine at least one candidate keyword that is associated with the keyword. The association determination unit is used to determine the association information of the first target document based on the keyword and at least one candidate keyword.

[0117] In one possible implementation, the association determination unit includes: The keyword group generation unit is used to generate keyword groups for each candidate keyword from at least one candidate keyword, based on the keyword. The matching degree determination unit is used to calculate the matching degree between the target file in the second database and the keyword group of each candidate keyword, and to determine the highest matching degree as the matching degree of the target file; The second target file determination unit is used to determine the target file as a second target file that is associated with the first target file if the matching degree of the target file exceeds the preset target matching degree. The association information generation unit is used to generate association information of the first target file. The association information indicates the matching degree of the first target file, keywords, second target file, and second target file.

[0118] In one possible implementation, the monitoring unit includes: The file update information monitoring unit is used to monitor file update information in the first database. The file update information indicates at least one original file that has been updated. The file update sequence determination unit is used to determine the file update sequence of at least one original file, wherein the file update sequence indicates the update priority of at least one original file. The original file determination unit is used to sequentially determine each original file indicated by the file update sequence as the original file to be managed.

[0119] In one possible implementation, the file update sequence determination unit includes: The data information acquisition unit is used to acquire data information of the original document in at least one pre-set dimension. The at least one dimension includes any one or more of the following: the degree of impact of the original document on the production line efficiency, the severity of the process defects indicated by the original document, and the update frequency of the original document. The target information generation unit is used to determine the target information of the original file based on the data information of the original file in at least one dimension. The target information represents the urgency of updating the original file. The sequence generation unit is used to sort at least one original file according to the target information of the original file and generate a file update sequence.

[0120] In one possible implementation, the stamping process knowledge management device further includes: The deletion unit is used to delete the first target file and its association information generated from the original file in the second database if the file update behavior of the original file is a file deletion behavior.

[0121] In one possible implementation, the stamping process knowledge management device further includes: The process constraint judgment unit is used to determine whether the first target file conforms to the process constraints of the domain thesaurus; The early warning mechanism triggering unit is used to trigger the early warning mechanism to start if the first target file does not meet the process constraints of the domain thesaurus. The early warning mechanism is used to indicate that the first target file is abnormal. Correspondingly, the information storage unit is specifically used to: if the first target file conforms to the process constraints of the domain thesaurus, store the first target file and the association information of the first target file in the second database.

[0122] The stamping process knowledge management device 500 provided in this embodiment belongs to the same application concept as the stamping process knowledge management method provided in the above embodiments of this application. It can execute the stamping process knowledge management method provided in any of the above embodiments of this application and has the corresponding functional modules and beneficial effects of executing the stamping process knowledge management method. Technical details not described in detail in this embodiment can be found in the corresponding method embodiments of this application, and will not be repeated here.

[0123] The functions implemented by each unit in the above device can be implemented by the same or different processors, and this application embodiment does not limit this.

[0124] It should be understood that the units in the above device can be implemented by a processor calling software. For example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit in the device. The processor can be a general-purpose processor, such as a CPU or microprocessor, and the memory can be internal or external to the device. Alternatively, the units in the device can be implemented as hardware circuits. By designing the hardware circuits, some or all of the unit functions can be implemented. The hardware circuits can be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are implemented by designing the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a PLD, such as an FPGA, which can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files to implement the functions of some or all of the above units. All units in the above device can be implemented entirely by a processor calling software, entirely by hardware circuits, or partially by a processor calling software with the remaining parts implemented by hardware circuits.

[0125] In this application embodiment, a processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction reading and execution capabilities, such as a CPU, microprocessor, GPU, or DSP. In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. These logical relationships are fixed or reconfigurable. For example, the processor may be a hardware circuit implemented as an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the processor loading instructions to implement the functions of some or all of the above units. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, or DPU.

[0126] As can be seen, each unit in the above device can be one or more processors (or processing circuits) configured to implement the above methods, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0127] Furthermore, the units in the above devices can be integrated in whole or in part, or they can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a System-on-Chip (SoC). The SoC may include at least one processor for implementing any of the above methods or implementing the functions of the units in the device. The at least one processor may be of different types, such as CPU and FPGA, CPU and artificial intelligence processor, CPU and GPU, etc.

[0128] Exemplary electronic devices Another embodiment of this application also proposes an electronic device. See [link to relevant documentation]. Figure 6 As shown, the electronic device may include: a memory 600 and a processor 610; wherein the memory 600 is connected to the processor 610 and is used to store programs; the processor 610 is used to implement the stamping process knowledge management method disclosed in any of the above embodiments by running the programs stored in the memory 600.

[0129] Specifically, the aforementioned electronic device may also include: a bus, a communication interface 620, an input device 630, and an output device 640.

[0130] The processor 610, memory 600, communication interface 620, input device 630, and output device 640 are interconnected via a bus. Among them: A bus can include a pathway for transmitting information between various components of a computer system.

[0131] The processor 610 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0132] The processor 610 may include a main processor, as well as a baseband chip, modem, etc.

[0133] The memory 600 stores a program for executing the technical solution of this application, and may also store an operating system and other key business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory 600 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0134] Input device 630 may include a device for receiving user input data and information, such as a keyboard, mouse, camera, scanner, light pen, voice input device, touch screen, pedometer, or gravity sensor.

[0135] Output device 640 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0136] The communication interface 620 may include a device that uses any transceiver to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0137] The processor 610 executes the program stored in the memory 600 and calls other devices, which can be used to implement any of the steps of the stamping process knowledge management method provided in the above embodiments of this application.

[0138] This application also proposes a chip, which includes a processor and a data interface. The processor reads and runs a program stored in the memory through the data interface to execute any of the stamping process knowledge management methods provided in the above embodiments. For the specific processing procedure and its beneficial effects, please refer to the above-described embodiments of the stamping process knowledge management method.

[0139] Exemplary computer program products and storage media In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the stamping process knowledge management method according to various embodiments of this application as described in any of the above embodiments of this specification.

[0140] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0141] Furthermore, embodiments of this application may also be storage media storing a computer program, which is executed by a processor of the steps in the stamping process knowledge management method according to various embodiments of this application as described in any of the foregoing embodiments of this specification.

[0142] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0143] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For apparatus embodiments, since they are basically similar to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0144] The steps in the methods of the various embodiments of this application can be adjusted, merged, or deleted in order according to actual needs, and the technical features described in each embodiment can be replaced or combined.

[0145] The modules and sub-modules in the apparatus and terminal in the various embodiments of this application can be merged, divided, and deleted according to actual needs.

[0146] It should be understood that the disclosed terminals, devices, and methods can be implemented in other ways, given the several embodiments provided in this application. For example, the terminal embodiments described above are merely illustrative. For instance, the division of modules or sub-modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple sub-modules or modules may be combined or integrated into another module, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or modules, and may be electrical, mechanical, or other forms.

[0147] The modules or submodules described as separate components may or may not be physically separate. The components that constitute a module or submodule may or may not be physical modules or submodules; that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0148] Furthermore, the functional modules or sub-modules in the various embodiments of this application can be integrated into one processing module, or each module or sub-module can exist physically separately, or two or more modules or sub-modules can be integrated into one module. The integrated modules or sub-modules described above can be implemented in hardware or in the form of software functional modules or sub-modules.

[0149] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0150] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software unit executed by a processor, or a combination of both. The software unit can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0151] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0152] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily 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 this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for managing knowledge of stamping processes, characterized in that, include: Monitor the original files to be managed in the first database, the original files including stamping process knowledge of at least one information type; Information extraction and structured processing are performed on at least one type of stamping process knowledge in the original file to generate a first target file; The association information of the first target file is determined based on the domain thesaurus, and the association information indicates a second target file in the second database that has an association with the first target file; The first target file and its association information are stored in the second database.

2. The method according to claim 1, characterized in that, The step of extracting and structuring the stamping process knowledge of at least one information type from the original file to generate the first target file includes: The text-type stamping process knowledge in the original document is subjected to text recognition and structured processing to obtain text recognition results. The text recognition results include at least one first data pair, which consists of process terminology and detailed content of process terminology. Extract the stamping process knowledge in the form of a table from the original file and perform structured processing to obtain the table recognition result. The table recognition result includes at least one second data pair, which consists of a field name and a field value. The field name is used to characterize the process term. The original document contains graphic-type stamping process knowledge. Graphic recognition and structuring are performed to obtain graphic recognition results. The graphic recognition results consist of process terms, process defect information and at least one third data pair. The third data pair includes structural feature name and structural feature size information. The text recognition results, the table recognition results, and the image recognition results are combined to generate a first target file. The first target file indicates the topic of the original file, and the topic of the original file is used to summarize the recognition results of the stamping process knowledge of at least one information type.

3. The method according to claim 2, characterized in that, The step of determining the association information of the first target file based on the domain thesaurus includes: A pre-set domain thesaurus is determined, the domain thesaurus including at least one type of process dictionary and the association relationships between different types of process terms in the at least one type of process dictionary; Extract keywords from the subject of the original file indicated by the first target file; Query the association relationships in the domain thesaurus to determine at least one candidate keyword that is associated with the keyword; The association information of the first target file is determined based on the keyword and the at least one candidate keyword.

4. The method according to claim 3, characterized in that, Determining the association information of the first target file based on the keyword and the at least one candidate keyword includes: Generate a keyword group for each of the at least one candidate keywords based on the keywords; Calculate the matching degree between the target file in the second database and the keyword group of each candidate keyword, and determine the highest matching degree as the matching degree of the target file; If the matching degree of the target file exceeds the preset target matching degree, the target file is determined to be a second target file that is associated with the first target file; Generate association information for the first target file, wherein the association information indicates the matching degree of the first target file, the keyword, the second target file, and the second target file.

5. The method according to claim 1, characterized in that, The original files to be managed in the first monitoring database include: Monitor file update information in the first database, wherein the file update information indicates at least one original file that has been updated; Determine a file update sequence for the at least one original file, the file update sequence indicating the update priority of the at least one original file; Each of the original files indicated by the file update sequence is identified as the original file to be managed.

6. The method according to claim 5, characterized in that, Determining the file update sequence of the at least one original file includes: Obtain data information of the original file in at least one pre-set dimension, wherein the at least one dimension includes any one or more of the following: the degree of impact of the original file on production line efficiency, the severity of process defects indicated by the original file, and the update frequency of the original file. The target information of the original file is determined based on the data information of the original file in at least one dimension, and the target information represents the urgency of updating the original file. The at least one original file is sorted according to the target information of the original file to generate a file update sequence.

7. The method according to claim 5, characterized in that, If the file update behavior of the original file is a file deletion behavior, the method further includes: Delete the association information between the first target file generated from the original file and the first target file in the second database.

8. The method according to claim 1, characterized in that, Also includes: Determine whether the first target file conforms to the process constraints of the domain thesaurus; If the first target file does not conform to the process constraints of the domain thesaurus, an early warning mechanism is triggered, which is used to indicate that the first target file is abnormal. The step of storing the first target file and its association information in the second database includes: if the first target file conforms to the process constraints of the domain thesaurus, storing the first target file and its association information in the second database.

9. A stamping process knowledge management device, characterized in that, include: A monitoring unit is used to monitor the original files to be managed in the first database, wherein the original files include stamping process knowledge of at least one information type; An information processing unit is used to extract and structure the stamping process knowledge of at least one information type in the original file to generate a first target file; An information association unit is used to determine the association information of the first target file based on a domain thesaurus, wherein the association information indicates a second target file in a second database that is associated with the first target file; An information storage unit is used to store the first target file and its association information in the second database.

10. An electronic device, characterized in that, Including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the stamping process knowledge management method as described in any one of claims 1 to 8 by running the program in the memory.

Images