Intelligent archive management method and device, machine readable storage medium and terminal device

Abstract

The application provides a kind of intelligent archive management method, device, machine readable storage medium and terminal equipment, it is related to archive management technical field, method includes: receiving the original archive data from at least one business system and carrying out the extraction of metadata field and file content, the extracted each metadata field is valued by the preset metadata field mapping relationship, the archive entry of corresponding original archive data is generated based on the metadata field after assignment and extracted file content;According to the archive entry of each original archive data, based on the preset archive classification rule and the rule of group volume, each archive entry is classified and group volume processing, generates corresponding archive volume;Based on the delivery rule, the original archive data of the archive to be delivered is pushed to the corresponding volume receiving end, and the storage location information of the corresponding archive volume is determined according to the structure parameters of virtual warehouse.The application effectively improves the management efficiency of the whole life cycle archives of chemical enterprises.

Description

Technical Field

[0001] This application relates to the field of archives management technology, specifically to an intelligent archives management method, an intelligent archives management device, a machine-readable storage medium, and a terminal device. Background Technology

[0002] With the widespread adoption of digital delivery in chemical industry project construction, all engineering data generated during the process (including design drawings, technical documents, supply documents, approval materials, construction records, etc.) has been digitized and managed, collected, and linked through digital delivery platforms. Currently, chemical enterprises are gradually transforming their record management from traditional paper-based archiving to digital management. The development and application of digital delivery provides the foundation for the intelligent upgrading of record management. Key technologies include: digital delivery technology (enabling standardized collection and storage of engineering data), record metadata management technology (defining core information fields and association rules for records), process automation technology (automatically triggering record processing flows through rule configuration), intelligent file assembly algorithms (automatic classification and combination technology based on record attributes and management standards), and data consistency technology (ensuring the consistency and traceability of record data across various business processes).

[0003] Currently, existing chemical enterprise record management technologies are mainly divided into two categories: one is the traditional record management system, which only realizes the electronic storage and simple retrieval of records, and relies on manual completion of core links such as record collection, cataloging, compilation, and delivery; the other is the general-purpose digital record management system, which supports some batch operations and format conversion, but has not been customized in accordance with the record management standards of chemical enterprises (such as the classification rules of engineering records in the chemical industry, the regulations on retention periods, etc.), and cannot adapt to the structured data characteristics of digital delivery results. In addition, it lacks the design of data homogeneity, which leads to the disconnect between record data and business data, requiring repeated entry and verification. Summary of the Invention

[0004] The purpose of this application is to provide an intelligent file management method, an intelligent file management device, a machine-readable storage medium, and a terminal device to solve the above-mentioned problems.

[0005] To achieve the above objectives, the first aspect of this application provides an intelligent record management method, comprising: Through a preset data source interface, the system receives raw archive data from at least one business system and extracts metadata fields and file content from the received raw archive data. Through a preset metadata field mapping relationship, the extracted metadata fields are automatically mapped to assigned metadata fields. Based on the assigned metadata fields and the extracted file content, the system generates archive entries corresponding to the raw archive data and assigns a unique traceability identifier to each archive entry. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. Based on the archival entries of each original archival data, and according to the preset archival classification and file assembly rules, each archival entry is classified and assembled to generate the corresponding archival file. In response to delivery instructions, based on preset delivery rules, the system selects files to be delivered from the generated files and pushes the original file data of the files to be delivered to the corresponding file receiving end. The delivered archive files are archived to a pre-built virtual warehouse, and the corresponding archive file entry location information is determined according to the structural parameters of the virtual warehouse. The generated entry location information is then sent to the corresponding entry execution terminal.

[0006] Optionally, the process of constructing the data source interface includes: Construct an internal data adapter interface, which at least declares a unified connection method and a data extraction method; For different business systems, specific adapter classes that implement the internal data adapter interface are created respectively. The connection method encapsulates the communication protocol used by the corresponding business system to establish a connection. In the data extraction method, the metadata fields extracted from the corresponding business system are automatically mapped and output by calling the metadata field mapping relationship, so as to achieve data collection from the same source.

[0007] Optionally, archive entries corresponding to the original archive data are generated based on the assigned metadata fields and the extracted file content, including: Obtain the field information of the metadata field, which includes at least one of the following: file name, file creation date, project number, professional field, and version number; Based on the extracted file content, the corresponding file information is determined, and the file information includes at least one of the following: file format, number of pages, and file size. Obtain the business system information corresponding to the original archive data, and generate archive entries corresponding to the original archive data based on the business system information, the field information, and the file information of the original archive data.

[0008] Optionally, a unique traceability identifier is assigned to the archival entry, including: extracting the source business system identifier and business data primary key of the original archival data, and obtaining the archival archiving timestamp of the original archival data, wherein the business data primary key is the field value used to uniquely identify the data record in the database of the source business system of the original archival data; A globally unique traceability identifier is generated by calculating the source business system identifier, the business data primary key, and the archive timestamp of the original archive data using a preset hash algorithm, and then the traceability identifier is bound and stored with the archive entry.

[0009] Optionally, the method further includes: In response to a tracing request for a target file entry, the tracing identifier of the target file entry is parsed to obtain the source system identifier and business data primary key associated with the tracing identifier; By tracing back to the corresponding business system through the data source interface, the original data record corresponding to the primary key of the business data can be queried, and all file modification records associated with the source identifier can be retrieved. By integrating the original data records and all file modification records along a timeline, a full-process traceability report is generated, covering the entire process from data generation and modification to archiving.

[0010] Optionally, after extracting metadata fields and file content from the received raw archive data, the method further includes: If there is no value matching the extracted metadata field in the preset metadata field mapping relationship, the extracted metadata field is used as the source field, and semantic parsing is performed on the source field; The source field's field information is obtained. Based on the pre-constructed knowledge graph and the source field's field information, the semantic matching degree between the source field and each preset metadata field in the knowledge graph is determined. A mapping relationship between the source field and the target metadata field is established based on the semantic matching degree to update the current metadata field mapping relationship. The knowledge graph includes field information corresponding to different preset metadata fields.

[0011] Optionally, semantic parsing is performed on the source field, including: Based on the field information of the source field, the source field is semantically parsed using a pre-trained semantic vector model to generate the source field semantic vector of the source field. Determining the semantic matching degree between the source field and each preset metadata field in the knowledge graph includes: Using each metadata field in the knowledge graph as a candidate field, and based on the field information corresponding to each candidate field, a candidate field semantic vector is generated for each candidate field through the semantic vector model. Calculate the cosine similarity between the semantic vector of the source field and the semantic vector of each candidate field, determine the shortest path distance between the source field and each candidate field, and determine the knowledge graph enhancement weights of the source field and each candidate field based on the shortest path distance; The semantic matching degree between the source field and each candidate field is determined based on the cosine similarity between the semantic vector of the source field and the semantic vector of each candidate field, and the knowledge graph enhancement weights of the source field and each candidate field.

[0012] Optionally, the knowledge graph enhancement weights between the source field and each candidate field are determined based on the shortest path distance, including: Based on the shortest path distance, the knowledge graph enhancement weights of the source field and the corresponding candidate field are determined by a pre-constructed path weight model, wherein the path weight model is used to represent the mapping relationship in which the knowledge graph enhancement weights decrease as the shortest path distance increases.

[0013] Optionally, the semantic matching degree between the source field and each candidate field is determined based on the cosine similarity between the semantic vector of the source field and the semantic vector of each candidate field, and the knowledge graph enhancement weights of the source field and each candidate field, including: Determine the baseline adjustment value and the adjustment coefficient of the knowledge graph enhancement weights; The knowledge graph enhancement weights are scaled using the adjustment coefficient, and the scaling factor of the cosine similarity between the source field semantic vector and each candidate field semantic vector is determined based on the baseline adjustment value and the scaled knowledge graph enhancement weights. The scaling factor increases with the increase of the knowledge graph enhancement weights, and the scaling factor is greater than 1. The cosine similarity between the source field semantic vector and the semantic vectors of each candidate field is scaled by the scaling factor to obtain the semantic matching degree between the source field and each candidate field.

[0014] Optionally, the structural parameters of the virtual warehouse include the number and location of the mobile shelving units, each of which contains multiple compartments; determining the storage location information of the corresponding archives based on the structural parameters of the virtual warehouse includes: Get the total number of pages in the current file file, and calculate the number of file boxes required for the current file file based on the total number of pages in the current file file and the preset file box capacity; The remaining capacity of each compartment of each mobile shelving unit in the virtual warehouse is scanned sequentially to determine the compartment whose remaining capacity is greater than or equal to the required number of file boxes, and the compartment with the smallest remaining capacity is the target compartment. Obtain the number of the target compartment and determine the starting box number and ending box number within the target compartment based on the required number of file boxes; The target compartment number, the starting box number and the ending box number within the target compartment are determined as the current file storage location information.

[0015] A second aspect of this application provides an intelligent file management device, comprising: The archive data acquisition module is configured to receive raw archive data from at least one business system through a preset data source interface, extract metadata fields and file content from the received raw archive data, automatically map the extracted metadata fields to assigned metadata fields through a preset metadata field mapping relationship, generate archive entries corresponding to the raw archive data based on the assigned metadata fields and the extracted file content, and assign a unique traceability identifier to the archive entries. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. The classification and file assembly module is configured to classify and assemble each archival item based on the archival entries of each original archival data, according to preset archival classification and file assembly rules, and generate corresponding archival files. The automatic delivery module is configured to respond to delivery instructions, based on preset delivery rules, to select files to be delivered from the generated file files, and push the original file data of the files to be delivered to the corresponding file receiving end. The archiving and storage module is configured to archive delivered archive files to a pre-built virtual storage room, determine the storage location information of the corresponding archive files based on the structural parameters of the virtual storage room, and send the generated storage location information to the corresponding storage execution terminal.

[0016] In a third aspect, this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the intelligent file management method described above.

[0017] In a fourth aspect, this application provides a terminal device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the intelligent file management method described above.

[0018] This application enables the integration of archival data with the digital delivery business data of chemical enterprises, avoiding duplicate entry and data inconsistencies, reducing data verification costs, and improving data accuracy and traceability. At the same time, by constructing metadata field mapping relationships, it automatically classifies and assembles the original archival data, and can automatically call delivery rules to execute delivery and archiving operations based on delivery instructions, effectively improving the management efficiency of the entire life cycle of archives of chemical enterprises.

[0019] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0020] The accompanying drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the following detailed description to explain the embodiments of this application, but do not constitute a limitation on the embodiments of this application. In the drawings: Figure 1 A system flowchart of an existing general-purpose document management system provided for a preferred embodiment of this application; Figure 2 A flowchart illustrating the intelligent file management method provided in a preferred embodiment of this application; Figure 3 This is a schematic diagram of the system architecture of the intelligent file management system provided in a preferred embodiment of this application; Figure 4 A system flowchart of the intelligent file management system provided in a preferred embodiment of this application; Figure 5 A schematic diagram of an intelligent file management device provided in a preferred embodiment of this application; Figure 6 A schematic diagram of a terminal device provided for a preferred embodiment of this application.

[0021] Explanation of reference numerals in the attached figures 10 - Terminal device, 100 - Processor, 101 - Memory, 102 - Computer program. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0023] It should be noted that the technical solutions of the various embodiments of this application can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by this application.

[0024] like Figure 1 The diagram shows the system flow of a general-purpose document management system in the prior art: business personnel upload document files - archivists manually enter metadata - manually classify documents according to type - manually assemble files - generate a delivery list - manually transfer and archive. It can be seen that the prior art has the following problems: 1. Data source consistency is not achieved. Archival data and business data (such as design documents, random supply materials, construction records, etc.) in the digital delivery results of chemical enterprise projects are independent of each other, resulting in problems such as duplicate data entry, data inconsistency, and difficulty in traceability; there is a lack of intelligent file assembly rules adapted to chemical industry standards, and file assembly relies on manual judgment, which is prone to problems such as classification errors and non-standard file structure; the archival delivery process is not connected with the digital delivery results, and manual extraction and sorting of delivery data are required, resulting in low delivery efficiency and easy errors.

[0025] 2. Manually entering metadata, manually assembling files, and repeatedly verifying data require a large number of archival management personnel, resulting in high labor costs; rework costs due to inconsistent data and non-standard file assembly increase; and general-purpose systems require additional investment in customized development costs to adapt to the needs of the chemical industry.

[0026] 3. The entire process of collecting, cataloging, compiling, and delivering archives is lengthy, which cannot meet the management needs of chemical enterprises with fast project progress and large archive volume. When manually searching and using archives, the retrieval efficiency is low due to problems such as non-standard classification and missing metadata, which affects the effectiveness of archive utilization.

[0027] To solve the above problems, such as Figure 2 As shown, the first aspect of this application provides an intelligent file management method, comprising: S100. Receive raw archive data from at least one business system through a preset data source interface, extract metadata fields and file content from the received raw archive data, automatically map the extracted metadata fields to assigned metadata fields through a preset metadata field mapping relationship, generate archive entries corresponding to the raw archive data based on the assigned metadata fields and the extracted file content, and assign a unique traceability identifier to the archive entries. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. S200. Based on the archival entries of each original archival data, and according to the preset archival classification rules and file assembly rules, classify and assemble each archival entry to generate the corresponding archival file. S300: In response to the delivery instruction, based on the preset delivery rules, it selects the files to be delivered from the generated files and pushes the original file data of the files to be delivered to the corresponding file receiving end. S400: Archive the delivered files to a pre-built virtual warehouse, determine the storage location information of the corresponding files based on the structure parameters of the virtual warehouse, and send the generated storage location information to the corresponding storage execution terminal.

[0028] Thus, this application enables the integration of archival data with the digital delivery business data of chemical enterprises, avoiding duplicate entry and data inconsistencies, reducing data verification costs, and improving data accuracy and traceability. At the same time, by constructing metadata field mapping relationships, it automatically classifies and assembles the original archival data, and can automatically call delivery rules to execute delivery and archiving operations based on delivery instructions, effectively improving the management efficiency of the entire life cycle archives of chemical enterprises.

[0029] The intelligent record management method of this application is applied to an intelligent record management system, wherein the system architecture of this application is as follows: Figure 3 As shown, the system includes a data source interface layer, a rule configuration layer, a core function layer, a storage layer, and a user interaction layer. Data transmission and interaction between the layers are achieved through a data bus.

[0030] like Figure 4 As shown, this application connects to the digital delivery platform and various business systems (design, construction, approval, etc.) of chemical enterprises through a standardized API interface pre-built in the data source interface layer. The standardized API interface extracts archive-related information (such as project number, file type, equipment tag number, responsible person, etc.) from the business data, thus eliminating the need for manual re-entry.

[0031] This application uses a pre-built data origin interface as a standardized API interface. The construction process of the data origin interface includes: building an internal data adapter interface, which declares at least a unified connection method and a data extraction method; creating specific adapter classes that implement the internal data adapter interface for different business systems; encapsulating the communication protocol used by the corresponding business system in the connection method to establish a connection; and automatically mapping and outputting the metadata fields extracted from the corresponding business system by calling the metadata field mapping relationship in the data extraction method, thereby achieving data origin collection. In this application, all business systems communicate using a unified connection method. Details such as handshakes, request and response parsing for various specific protocols such as SOAP, REST, and JDBC are completely encapsulated within the adapters corresponding to each business system. The adapters are customized for each external system, completely encapsulating the differences in various protocols, connection methods, and authentication methods within their respective adapters, preventing them from spreading to the core business logic of the archive system, thus achieving unified communication protocols. Meanwhile, the adapter internally calls a unified metadata field mapping relationship to perform a unified standard conversion mapping on the extracted raw data, thereby achieving data collection from the same source, avoiding manual re-entry, and ensuring the consistency and traceability of archival data with its business source data.

[0032] In step S100, raw archive data from different business systems is received through a standardized API interface. For example, a designer needs to upload a P&ID drawing of chemical equipment such as a cracked gas compressor. The raw archive data of this P&ID drawing is submitted to the standardized API interface along with the API request. The raw archive data includes raw metadata fields and file content. The metadata fields include, but are not limited to, at least one of the following: file name, file creation date, project number, professional field, and version number. For example, the extracted metadata fields include the following: doc_title: cracked gas compressor P&ID, drawing_id: PID-1001, tag_no: C-101, revision_code: A, responsible: Zhang San; the file content includes an encoded P&ID drawing file, such as Base64 encoded DWG binary data. In this application, the file information in the document content includes, but is not limited to, at least one of the following: file format, number of pages, and file size. Extracting the file content and determining the corresponding file information includes: performing a decoding operation on the original P&ID drawing file, restoring the Base64 encoding to DWG binary data, converting the DWG format to PDF, overlaying a watermark on each page of the PDF, extracting the number of pages in the PDF file, and obtaining the byte count of the PDF and the original DWG files. The system supports automatically converting more than 20 file formats, including TIF, JPG, and WORD, to PDF format and adding watermarks.

[0033] After extracting metadata fields and file content from the original archive data, the metadata fields of the original archive data are matched and assigned values ​​through a pre-constructed metadata field mapping relationship. This metadata field mapping relationship is pre-constructed based on the mapping relationship between different metadata fields and their corresponding values. For example, in the preset metadata field mapping relationship, the value of the metadata field `doc_title` is the archive name, the value of `drawing_id` is the drawing number, the value of `tag_no` is the equipment tag number, the value of `revision_code` is the version number, and the value of `responsible` is the responsible person. After mapping and assignment, the assigned metadata fields are: Archive Name: Pyrolysis Gas Compressor P&ID, Drawing Number: PID-1001, Equipment Tag Number: C-101, Version Number: A, Responsible Person: Zhang San.

[0034] This application defines a file metadata system that conforms to the standards of the chemical industry in the metadata database. This system includes basic metadata (file name, creation date, responsible person, etc.) and industry-specific metadata (chemical process type, equipment model, pipeline grade, etc.). The metadata is mapped one by one with the business data, ensuring that the data is from the same source and synchronized.

[0035] At this point, the data extracted from the original archival data includes data such as the archive name, drawing number, equipment tag number, responsible person, creation date, and version number from the metadata mapping, as well as data such as the storage path, page number, and file size extracted from the file content. The system merges the data from the metadata mapping and the data extracted from the file content, generates a unique traceability identifier, and automatically adds fields such as archive entry ID and creation time. Simultaneously, it obtains the business system information corresponding to the original archival data; for example, if the original archival data comes from business system A, the above information is merged to generate the corresponding archive entry for the original archival data. This application assigns a unique traceability identifier to each piece of archival data, and by associating it with its corresponding digital delivery business data source, it achieves full-process traceability of archival data from generation to archiving and utilization. Furthermore, data modification records are retained in real time, ensuring data authenticity.

[0036] In this application, assigning a unique traceability identifier to an archival entry includes: extracting the source business system identifier and business data primary key of the original archival data, and obtaining the archival archiving timestamp of the original archival data. The business data primary key is the field value used to uniquely identify the data record in the database of the source business system. For example, if the source business system of the original archival data is a construction project management system, the field value corresponding to the storage location of the original archival data in that system's database is the data's business data primary key. A globally unique traceability identifier is generated by calculating the source business system identifier, business data primary key, and archival archiving timestamp of the original archival data using a preset hash algorithm. This traceability identifier is then bound and stored with the archival entry. For example, the source business system identifier, business data primary key, and archival archiving timestamp of the original archival data are concatenated, and the concatenated string is calculated using the SHA-256 hash algorithm to obtain a 256-bit hash value, which is then converted, for example, to a 64-bit hexadecimal string. This string has globally unique, irreversible, and fixed-length characteristics, and can serve as a unique traceability identifier for the archival entry.

[0037] The method of this application further includes: responding to a traceability request for a target file entry, parsing the traceability identifier of the target file entry, obtaining the source system identifier and business data primary key associated with the traceability identifier; tracing back to the corresponding business system through the data source interface, querying the original data record corresponding to the business data primary key, and retrieving all file modification records associated with the traceability identifier; and integrating the original data records and all file modification records according to the timeline to generate a full-process traceability report from data generation, modification to archiving. For example, when a user initiates a traceability request for a certain file, the system executes the following traceability steps: Based on the traceability identifier of the file entry (i.e., the hash value mentioned above), it quickly retrieves the associated source system identifier and business data primary key from the file traceability index table; through the data source interface, it schedules the adapter of the corresponding system, queries the database of the corresponding system through the business data primary key, and obtains the current original data snapshot of the record (such as the person initially responsible for entry, the date of creation, file path, etc.); in the file modification log table, using the traceability identifier as a condition, it queries all historical modification records, each record containing information such as modification time, modifier, and comparison of the modified content before and after; taking the business source data snapshot as the starting point, it arranges all modification records in chronological order to generate a complete timeline, and finally forms a full-process traceability report.

[0038] Understandably, during the metadata mapping operation, not all metadata of the original archive data can be matched with corresponding values ​​in the preset metadata field mapping relationship. Therefore, after extracting the metadata fields and file content of the received original archive data in step S100, the method of this application further includes: S110. If no value matches the extracted metadata field in the preset metadata field mapping relationship, the extracted metadata field is used as the source field. Semantic parsing is performed on the source field. Specifically, based on the field information of the source field, a pre-trained semantic vector model is used to perform semantic parsing on the source field, generating a source field semantic vector. For example, using BERT-base-uncased as the base model, domain-adaptive fine-tuning is performed on chemical industry documents (including design specifications, HAZOP reports, equipment data sheets, etc.) to obtain a pre-trained semantic vector model. The input of the semantic vector model is the metadata field, and the output is a vector of a specified dimension.

[0039] S120. Obtain the field information of the source field. Based on the pre-constructed knowledge graph and the field information of the source field, determine the semantic matching degree between the source field and each preset metadata field in the knowledge graph. Establish a mapping relationship between the source field and the target metadata field according to the semantic matching degree to update the current metadata field mapping relationship. The knowledge graph includes field information corresponding to different preset metadata fields. Specifically, the knowledge graph of this application stores entities, concepts, attributes related to the chemical industry and their semantic relationships. Each entity is stored in a triple structure, such as (P&ID, synonym, process flow diagram), (responsible, synonym, responsible person), (EquipmentTagConcept, name, EquipmentTag), (EquipmentTagConcept, synonym, TagConcept), (TagConcept, belongs to, equipment tag number Concept), etc.

[0040] This includes determining the semantic matching degree between the source field and each preset metadata field in the knowledge graph, including: S121. Using each metadata field in the knowledge graph as a candidate field, and based on the field information corresponding to each candidate field, a semantic vector is generated for each candidate field through a semantic vector model. For example, the source field is EquipmentTag, which represents the device tag number. Since different business systems may use different field names, such as DeviceID or tag_no, the pre-built metadata field mapping relationship does not include a mapping relationship between EquipmentTag and device tag number. Therefore, EquipmentTag is first used as input, and the semantic vector of EquipmentTag is output through the semantic vector model. It is understandable that, in order to improve the matching accuracy, when calculating the semantic vector of the source field, in addition to inputting the source field EquipmentTag, information such as the field type, field value examples, and business system of the source field can be further combined as input to the semantic vector model. Among them, the field type, field value (e.g., C-101), and business system information can be directly obtained from the extracted metadata fields. Next, obtain the field information for each candidate field. For example, if the candidate field is the device tag number, its associated triples include (Device Tag Number Concept, Field Name, Device Tag Number), (Device Tag Number Concept, Definition, Unique Identifier Code of the Device in the Device), (Device Tag Number Concept, Synonym, Device Number), (Device Tag Number Concept, Synonym, Tag), and (Device Tag Number Concept, Associated Entity Type, Device). Then, construct the semantic vector model input using the relevant field information of the candidate field Device Tag Number Concept: "Field Name: Device Tag Number; Definition: Unique Identifier Code of the Device in the Device; Synonym: Device Number, Tag; Associated Entity: Device", to obtain the semantic vector of the candidate field Device Tag Number Concept. It is understandable that candidate fields can be determined based on the attribute information of the triples or the subject of the triples.

[0041] S122. Calculate the cosine similarity between the semantic vector of the source field and the semantic vectors of each candidate field, determine the shortest path distance between the source field and each candidate field, and determine the knowledge graph enhancement weights of the source field and each candidate field based on the shortest path distance. The calculation process of the cosine similarity between vectors is existing technology and is not limited here. The shortest path distance between the source field and each candidate field is the shortest path length from the source field name to each candidate field. For example, if the source field is EquipmentTag, and the knowledge graph includes triples (EquipmentTagConcept, name, EquipmentTag), (EquipmentTagConcept, synonym, TagConcept), and (TagConcept, belongs to, device tag number Concept), to map the source field to the device tag number Concept, the path is EquipmentTagConcept - synonym TagConcept - device tag number Concept, with a path length of 2. Traverse all mapping paths to determine the shortest path length from the source field to the candidate field.

[0042] S123. Based on the cosine similarity between the semantic vector of the source field and the semantic vector of each candidate field, and the knowledge graph enhancement weights of the source field and each candidate field, determine the semantic matching degree between the source field and each candidate field.

[0043] The determination of knowledge graph enhancement weights between the source field and each candidate field based on the shortest path distance includes: determining the knowledge graph enhancement weights between the source field and the corresponding candidate field based on the shortest path distance using a pre-constructed path weight model. The path weight model represents the mapping relationship where the knowledge graph enhancement weight decreases as the shortest path distance increases. For example, the path weight model can be expressed as P = 1 / (L+1), where P is the knowledge graph enhancement weight and L is the shortest path length between the source field and the candidate field.

[0044] Based on the cosine similarity between the semantic vectors of the source field and each candidate field, and the knowledge graph enhancement weights of the source field and each candidate field, the semantic matching degree between the source field and each candidate field is determined, including: The process involves determining a baseline adjustment value and an adjustment coefficient for the knowledge graph enhancement weights. The knowledge graph enhancement weights are then scaled using this adjustment coefficient. Based on the baseline adjustment value and the scaled knowledge graph enhancement weights, a scaling factor is determined for the cosine similarity between the source field semantic vector and the semantic vectors of each candidate field. This scaling factor increases with the increase of the knowledge graph enhancement weights and is greater than 1. The cosine similarity between the source field semantic vector and the semantic vectors of each candidate field is then scaled using this scaling factor to obtain the semantic matching degree between the source field and each candidate field. Specifically, in this application, the baseline adjustment value is set to 1, the adjustment coefficient for the knowledge graph enhancement weights is λ, and the value is 0.5. The formula for calculating the semantic matching degree between the source field and the candidate fields can be expressed as S = cos_sim × (1 + λ × P), where S represents the semantic matching degree and cos_sim represents the cosine similarity. Understandably, if there are no candidate fields in the knowledge graph that can be mapped to the source field, i.e., there is no path from the source field to the candidate field, then L is infinite and P is 0. In this case, only cosine similarity is relied upon as the semantic matching degree. The candidate field with the highest semantic matching degree is used as the value of the source field to establish a mapping relationship between the source field and the candidate field, thereby dynamically updating the metadata field mapping relationship.

[0045] In step S200, the file classification rules and file assembly rules can be configured in advance. After the file entries are generated, the system calls the classification rules and file assembly rules of the rule configuration layer to automatically classify the extracted file entries, generate file numbers according to the file naming rules, arrange the files in the file according to the file sorting rules, automatically calculate the number of pages in the file (the sum of the number of pages of all files), automatically determine the file retention period (take the maximum retention period of the files in the file), complete the intelligent file assembly, and support drag and drop to adjust the file content (manual intervention optimization).

[0046] The classification rules can be configured based on the engineering file management standards of chemical enterprises. Classification rules can be preset at the rule configuration layer, for example, by project stage (preliminary, design, construction, completion), file type (drawings, documents, records, reports), and professional field (process, equipment, electrical, instrumentation), etc., allowing enterprises to customize and expand classification dimensions. Project stage, file type, and professional field can be directly obtained from the metadata fields of the original document data.

[0047] In this application, the intelligent file assembly rule configuration includes: configuring a file assembly rule library, including file naming rules (project number + classification code + file serial number), document sorting rules (creation date + importance level), automatic retention period determination rules (associating retention periods with file type and project level, such as permanent retention for design drawings of major projects), and security classification rules (automatically labeling security levels based on the technical confidentiality level involved in the files). The relevant information for the file assembly rules can also be obtained directly from the metadata fields of the original document data.

[0048] In step S300, after the file assembly is completed, the system receives the delivery instruction, automatically filters the files to be delivered according to the delivery rules, generates a delivery list, packages the delivery files in the standard format (EEP data package), and automatically verifies the integrity and compliance of the delivered files (such as checking whether the number of pages in the file matches the number of pages in the table of contents, and whether the resolution meets the standard). If the verification is successful, it is pushed to the delivery receiving end (such as the owner's file system or the regulatory department's platform). If the verification fails, an error report is generated and a reminder for rectification is issued.

[0049] In step S400, after delivery, the system automatically transfers the archive data from the sorting repository to the archive repository. Based on the virtual structure of the repository (supporting custom repository and mobile shelving layout) and the thickness of the archive boxes and the width of the mobile shelving partitions, the system automatically calculates the archive storage location and updates the repository location field, supporting panoramic 720° virtual repository viewing and positioning. The delivery rule configuration includes: preset digital delivery package generation rules, supporting automatic filtering of the scope of archives to be delivered according to acceptance requirements, generating EEP format delivery packages according to the DA / T48-2009 standard, and automatically verifying the integrity of the delivered archives (such as whether essential metadata is missing and whether the file format is compliant). For example, the delivery scope rule can be configured as: Project Stage = "Completed" AND Archive Type IN ("Drawings", "Report", "Certificate") AND Security Class ≠ "Confidential" AND File Format = "PDF". After selecting file entries that meet the delivery rules, the EEP packaging service is invoked to generate a delivery package according to the national standard DA / T48-2009. The system automatically verifies the generated EEP package to ensure delivery quality, specifically including the following checks: Metadata integrity: Checking whether the file number, project number, creation date, and responsible person for each file are not empty; if missing information is found, an error is recorded; File format compliance: Confirming that all file extensions are within the PDF, TIF, and JPG whitelist; Page count and table of contents consistency: Comparing the page number field of each file's table of contents with the actual number of PDF pages, allowing a deviation of ±1 page; Resolution verification: For scanned PDFs, checking whether the resolution of each page is not less than 300 DPI. If all checks pass, the system marks the delivery package as qualified; if there are any failures, the system generates a detailed error report and notifies the administrator via site message and email. The delivery is then re-executed after the administrator corrects the problem.

[0050] The structural parameters of the virtual warehouse include the number and location of the mobile shelving units, each containing multiple compartments. Based on these parameters, the warehouse location information for corresponding archival files is determined, including: obtaining the total number of pages in the current archival file; calculating the required number of archive boxes based on the total number of pages and the preset archive box capacity (where the thickness of the archive boxes is known and can be estimated based on the total number of pages); sequentially scanning the remaining capacity of each compartment in each mobile shelving unit of the virtual warehouse, determining the compartment with the remaining capacity greater than or equal to the required number of archive boxes, and identifying the compartment with the smallest remaining capacity as the target compartment; checking the current warehouse occupancy status, and finding the first mobile shelving compartment with sufficient continuous space or sufficient continuous space with the smallest available space; obtaining the target compartment number and determining the starting and ending box positions within the target compartment based on the required number of archive boxes; and determining the target compartment number, starting and ending box positions within the target compartment as the current archival file's warehouse location information.

[0051] To further improve the efficiency of record management, the system in this application also has the following auxiliary functions: Batch processing function: Supports batch generation of file numbers, batch modification of file attributes (such as year, responsible person), batch matching of electronic files and directories, and batch download of file files.

[0052] Expiration reminder function: Set expiration reminders according to the storage period rules, and push reminder notifications to designated personnel through in-site messages, emails, etc., and support customizing the number of days in advance reminder.

[0053] Query and Utilization Functions: Provides quick and accurate retrieval across types and archival groups, supports searching by keywords and metadata combinations, allows setting the number of times or duration for electronic archive borrowing, and automatically retrieves archives upon expiration, and supports online reservation, borrowing registration, renewal and reminder management for physical archives.

[0054] Statistics and reporting functions: Supports statistics on workload, collection size, and utilization by dimensions such as archival collection, department, archive type, and time interval. Automatically generates archive reports (documents, infrastructure, scientific research, etc.) that conform to chemical industry standards. Supports exporting to Excel and PDF formats and chart display (pie charts, bar charts, etc.).

[0055] In summary, this application integrates archival data with the digital delivery business data of chemical enterprises, avoiding duplicate entry and data inconsistencies, reducing data verification costs, improving data accuracy and traceability, and solving the problem of data disconnect in existing technologies. Based on industry-specific rules, this application achieves automatic classification, intelligent file assembly, and automatic delivery of archives, significantly reducing manual operations, improving file assembly efficiency by over 80%, and shortening the delivery cycle by 70%, thus solving the problems of low efficiency and non-standard file assembly in existing technologies. Based on a deep alignment with the archival management standards of chemical enterprises, it pre-sets industry-specific rules for classification, file assembly, and delivery, eliminating the need for additional... Customized development ensures compliance in document management and resolves the poor compatibility issues of existing general-purpose systems. This application integrates full-process functions such as data collection, intelligent processing, document assembly, delivery, archiving, querying, statistics, and reporting. It also supports auxiliary functions such as virtual warehouse management, expiration reminders, and identification and destruction, meeting the full lifecycle document management needs of chemical enterprises and providing more comprehensive functionality. Document security is ensured through functions such as file format conversion, watermark addition, access control, and data traceability. At the same time, it supports personalized interfaces and permission settings for multiple user roles (document administrators, leaders, and ordinary users), making it more practical than existing systems.

[0056] like Figure 5 As shown, in a second aspect, this application provides an intelligent file management device, comprising: The archive data acquisition module is configured to receive raw archive data from at least one business system through a preset data source interface, extract metadata fields and file content from the received raw archive data, automatically map the extracted metadata fields to assigned metadata fields through a preset metadata field mapping relationship, generate archive entries corresponding to the raw archive data based on the assigned metadata fields and the extracted file content, and assign a unique traceability identifier to the archive entries. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. The classification and file assembly module is configured to classify and assemble each archival item based on the archival entries of each original archival data, according to preset archival classification and file assembly rules, and generate corresponding archival files. The automatic delivery module is configured to respond to delivery instructions, based on preset delivery rules, to select files to be delivered from the generated file files, and push the original file data of the files to be delivered to the corresponding file receiving end. The archiving and storage module is configured to archive delivered archive files to a pre-built virtual storage room, determine the storage location information of the corresponding archive files based on the structure parameters of the virtual storage room, and send the generated storage location information to the corresponding storage execution terminal.

[0057] It is understood that those skilled in the art will clearly recognize that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0058] In a third aspect, this application provides a machine-readable storage medium storing instructions that, when executed by a processor, configure the processor to perform the intelligent file management method described above.

[0059] In a fourth aspect, this application provides a terminal device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the intelligent file management method described above.

[0060] like Figure 6 The diagram shown is a schematic representation of a terminal device provided in an embodiment of this application. Figure 6 As shown, the terminal device 10 of this embodiment includes a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and executable on the processor 100. When the processor 100 executes the computer program 102, it implements the steps in the above method embodiments. Alternatively, when the processor 100 executes the computer program 102, it implements the functions of each module / unit in the above device embodiments.

[0061] For example, computer program 102 may be divided into one or more modules / units, one or more of which are stored in memory 101 and executed by processor 100 to complete this application. One or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 102 in terminal device 10.

[0062] Terminal device 10 may be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. Terminal device 10 may include, but is not limited to, a processor 100 and a memory 101. Those skilled in the art will understand that... Figure 6 This is merely an example of terminal device 10 and does not constitute a limitation on terminal device 10. It may include more or fewer components than shown, or combine certain components, or different components. For example, terminal device may also include input / output devices, network access devices, buses, etc.

[0063] Processor 100 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0064] The memory 101 can be an internal storage unit of the terminal device 10, such as a hard disk or RAM of the terminal device 10. The memory 101 can also be an external storage device of the terminal device 10, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or FlashCard equipped on the terminal device 10. Furthermore, the memory 101 can include both internal and external storage units of the terminal device 10. The memory 101 is used to store computer programs and other programs and data required by the terminal device 10. The memory 101 can also be used to temporarily store data that has been output or will be output.

[0065] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0067] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. An intelligent record management method, characterized in that, include: Through a pre-defined data source interface, raw archive data from at least one business system is received. Metadata fields and file content are extracted from the received raw archive data. Through a pre-defined metadata field mapping relationship, the extracted metadata fields are automatically mapped to assigned metadata fields. Based on the assigned metadata fields and the extracted file content, corresponding archive entries for the raw archive data are generated, and a unique traceability identifier is assigned to each archive entry. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. Based on the archive entries of each raw archive data, and according to pre-defined archive classification rules and file assembly rules, each archive entry is classified and assembled into a file, generating corresponding archive files. In response to delivery instructions, based on preset delivery rules, the system selects files to be delivered from the generated files and pushes the original file data of the files to be delivered to the corresponding file receiving end. The delivered archive files are archived to a pre-built virtual warehouse, and the corresponding archive file entry location information is determined according to the structural parameters of the virtual warehouse. The generated entry location information is then sent to the corresponding entry execution terminal.

2. The intelligent file management method according to claim 1, characterized in that, The construction process of the data source interface includes: Construct an internal data adapter interface, which at least declares a unified connection method and a data extraction method; For different business systems, specific adapter classes that implement the internal data adapter interface are created respectively. The connection method encapsulates the communication protocol used by the corresponding business system to establish a connection. In the data extraction method, the metadata fields extracted from the corresponding business system are automatically mapped and output by calling the metadata field mapping relationship, so as to achieve data collection from the same source.

3. The intelligent file management method according to claim 1, characterized in that, Based on the assigned metadata fields and the extracted file content, archive entries corresponding to the original archive data are generated, including: Obtain the field information of the metadata field, which includes at least one of the following: file name, file creation date, project number, professional field, and version number; Based on the extracted file content, the corresponding file information is determined, and the file information includes at least one of the following: file format, number of pages, and file size. Obtain the business system information corresponding to the original archive data, and generate archive entries corresponding to the original archive data based on the business system information, the field information, and the file information of the original archive data.

4. The intelligent file management method according to claim 1, characterized in that, Assigning a unique traceability identifier to an archival entry includes: extracting the source business system identifier and business data primary key of the original archival data, and obtaining the archival archiving timestamp of the original archival data, wherein the business data primary key is the field value used to uniquely identify the data record in the database of the source business system of the original archival data; A globally unique traceability identifier is generated by calculating the source business system identifier, the business data primary key, and the archive timestamp of the original archive data using a preset hash algorithm, and then the traceability identifier is bound and stored with the archive entry.

5. The intelligent file management method according to claim 4, characterized in that, The method further includes: In response to a tracing request for a target file entry, the tracing identifier of the target file entry is parsed to obtain the source system identifier and business data primary key associated with the tracing identifier; By tracing back to the corresponding business system through the data source interface, the original data record corresponding to the primary key of the business data can be queried, and all file modification records associated with the source identifier can be retrieved. By integrating the original data records and all file modification records along a timeline, a full-process traceability report is generated, covering the entire process from data generation and modification to archiving.

6. The intelligent file management method according to claim 1, characterized in that, After extracting metadata fields and file content from the received raw archive data, the method further includes: If there is no value matching the extracted metadata field in the preset metadata field mapping relationship, the extracted metadata field is used as the source field, and semantic parsing is performed on the source field; The source field's field information is obtained. Based on the pre-constructed knowledge graph and the source field's field information, the semantic matching degree between the source field and each preset metadata field in the knowledge graph is determined. A mapping relationship between the source field and the target metadata field is established based on the semantic matching degree to update the current metadata field mapping relationship. The knowledge graph includes field information corresponding to different preset metadata fields.

7. The intelligent record management method according to claim 1, characterized in that, The structural parameters of the virtual warehouse include the number and location of the mobile shelving units, with each shelving unit containing multiple compartments; the storage location information of the corresponding archives is determined based on the structural parameters of the virtual warehouse, including: Get the total number of pages in the current file file, and calculate the number of file boxes required for the current file file based on the total number of pages in the current file file and the preset file box capacity; The remaining capacity of each compartment of each mobile shelving unit in the virtual warehouse is scanned sequentially to determine the compartment whose remaining capacity is greater than or equal to the required number of file boxes, and the compartment with the smallest remaining capacity is the target compartment. Obtain the number of the target compartment and determine the starting box number and ending box number within the target compartment based on the required number of file boxes; The target compartment number, the starting box number and the ending box number within the target compartment are determined as the current file storage location information.

8. An intelligent file management device, characterized in that, include: The archive data acquisition module is configured to receive raw archive data from at least one business system through a preset data source interface, extract metadata fields and file content from the received raw archive data, automatically map the extracted metadata fields to assigned metadata fields through a preset metadata field mapping relationship, generate archive entries corresponding to the raw archive data based on the assigned metadata fields and the extracted file content, and assign a unique traceability identifier to the archive entries. The metadata field mapping relationship includes the values ​​corresponding to different metadata fields. The classification and file assembly module is configured to classify and assemble each archival item based on the archival entries of each original archival data, according to preset archival classification and file assembly rules, and generate corresponding archival files. The automatic delivery module is configured to respond to delivery instructions, based on preset delivery rules, to select files to be delivered from the generated file files, and push the original file data of the files to be delivered to the corresponding file receiving end. The archiving and storage module is configured to archive delivered archive files to a pre-built virtual storage room, determine the storage location information of the corresponding archive files based on the structural parameters of the virtual storage room, and send the generated storage location information to the corresponding storage execution terminal.

9. A machine-readable storage medium storing instructions thereon, characterized in that, When executed by a processor, this instruction causes the processor to be configured to perform the intelligent file management method as described in any one of claims 1-7.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent file management method as described in any one of claims 1-7.

Images