Conflict identification management method and device for multi-source heterogeneous industrial data tags

Abstract

This invention provides a method and apparatus for conflict identification and management of multi-source heterogeneous industrial data tags, comprising: acquiring tag change event information of industrial data streams and converting it into structured tag change event objects; constructing a tag graph in the form of an inheritance chain structure based on the tag change event objects; clustering the tag graph to generate a tag conflict report; generating renaming tags and conflict resolution suggestions for conflicting tags, or receiving input custom names to change the conflicting tags in the tag conflict report to generate renaming tags and generate conflict resolution suggestions; sending the renaming tags and conflict resolution suggestions to a target user terminal for review; and displaying the tag change path and tag information of the target tags in the renaming tags. This invention can solve the significant shortcomings of existing technologies in the lifecycle management, semantic consistency assurance, and cross-system traceability capabilities of industrial data tags.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and apparatus for conflict identification and management of multi-source heterogeneous industrial data tags. Background Technology

[0002] With the rapid development of Industrial Internet of Things (IIoT) platforms, more and more industrial enterprises are building digital production systems centered on data. In this process, industrial data from multiple data source systems (such as DCS, PLC, SCADA, MES, ERP, QMS, etc.) are continuously collected, processed, summarized, and transmitted to a unified platform for analysis, display, and decision-making.

[0003] Data transferred between these systems typically carries tags to identify its name, purpose, source, unit, equipment, and process location. These tags may serve different purposes in different systems, for example: In DCS / PLC systems, tags are typically used for the physical definition of underlying data variables (such as temperature, current, and speed). In SCADA, labels are converted into engineering quantity names that are more business-meaningful; In MES, the label may be renamed to "Process Parameters" or "Quality Judgment Items"; Ultimately, in the ERP / QMS system, these tags are abstracted into business attributes or statistical indicators.

[0004] To meet the personalized data semantic requirements of different systems, tags often undergo the following changes during transmission: Rename (e.g., sensor_01 → MainLine_Temp); Unit conversion (e.g., Celsius to Fahrenheit); Data abstraction (e.g., converting sensor data into "normal / abnormal"); Multiple label merging (e.g., aggregating multiple temperature labels into "average temperature"); Tag splitting (such as splitting a compound tag into multiple sub-tags), etc.

[0005] These tag changes are common in industrial platforms and are an objective result of the evolution of actual business processes and system integration. However, current mainstream IIoT platforms often lack tag change recording mechanisms and unified tag semantic management capabilities when handling multi-system tag interactions.

[0006] Although some industrial platforms already possess basic label standardization capabilities (such as label naming conventions and label template tools), the following significant problems still exist in actual business operations involving the integration, processing, and analysis of multi-source heterogeneous data: 1. The label change process is untraceable. Data flows between multiple systems, and tags undergo multiple renaming, abstraction, and merging processes. The platform cannot record or restore the evolution path of tags, making it difficult to understand and reconstruct the semantics of the data.

[0007] 2. Frequent tag conflicts In different systems, the same label name may represent different meanings (homonyms), or multiple label names may refer to the same meaning (synonyms), which brings ambiguity and conflict problems to data integration and analysis.

[0008] 3. Difficulty in data traceability affects the accuracy of decision-making. When analysts use tag data displayed in a business system, they cannot accurately determine its source, change process, or original meaning, which leads to a decrease in trust in the analysis results and even incorrect judgments.

[0009] 4. Lack of a unified cross-system label governance mechanism Currently, most platforms only manage tags within their own system, and cannot uniformly govern and audit tag changes throughout the entire data lifecycle at the platform level.

[0010] 5. Hinders model building and data asset management When building industrial AI models or data asset catalogs, inconsistent tag semantics make it difficult to achieve generalized reuse of models and unified archiving of data assets.

[0011] In summary, existing technologies have significant shortcomings in the lifecycle management, semantic consistency assurance, and cross-system traceability capabilities of industrial data tags. There is an urgent need for a new platform mechanism that can achieve complete, traceable, and conflict-detectable management capabilities for data tags "from source to end". Summary of the Invention

[0012] In view of this, it is necessary to provide a conflict identification and management device for multi-source heterogeneous industrial data tags to solve the technical problems that existing technologies have significant shortcomings in the lifecycle management, semantic consistency assurance and cross-system traceability capabilities of industrial data tags.

[0013] To address the aforementioned problems, in a first aspect, the present invention provides a conflict identification and management method for multi-source heterogeneous industrial data tags, comprising: Acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects; Based on the aforementioned tag change event object, construct a tag graph in the form of an inheritance chain structure; Cluster analysis is performed on the tag map to generate a tag conflict report; Based on the preset rule base and the tag conflict report, rename tags and conflict handling suggestions are generated for conflicting tags; or, custom names are received to modify the conflicting tags in the tag conflict report to generate rename tags and conflict handling suggestions, and the rename tags and conflict handling suggestions are sent to the target user terminal for review. In response to the tag query command, the tag change path and tag information of the target tag in the renamed tag are displayed.

[0014] In one possible implementation, the label change event information includes: the label name before and after the change, the change type, and metadata related to the label change event.

[0015] In one possible implementation, the label change event object includes: whether the label has been changed and the type of change; Based on the aforementioned tag change event object, a tag graph in the form of an inheritance chain structure is constructed, including: By constructing nodes in the inheritance chain based on the changed labels, directed edges are built between the nodes according to the change type, resulting in a label graph with an inheritance chain structure.

[0016] In one possible implementation, the tag graph includes: tag name similarity, contextual attributes, unit type, and the system to which the tag belongs; the tag conflict report includes conflicting tag pairs, conflict type, and conflict reason.

[0017] In one possible implementation, cluster analysis is performed on the tag map to generate a tag conflict report, including: Extract metadata information from the tag map; Convert the metadata information into high-dimensional word vectors; Based on the high-dimensional word vectors, the semantic similarity, contextual feature similarity, and inheritance chain path similarity between tags are calculated; Based on the semantic similarity, the contextual feature similarity, and the inheritance chain path similarity, multidimensional semantic fusion and conflict determination are performed on the labeled images to generate a label conflict report.

[0018] In one possible implementation, the contextual feature similarity between tags is calculated based on the high-dimensional word vectors, including: Extract the contextual attributes of each tag based on the high-dimensional word vectors; The contextual attributes of each tag are weighted and summed to obtain the contextual feature similarity between tags.

[0019] In one possible implementation, the similarity of inheritance chain paths between tags is calculated based on the high-dimensional word vectors, including: Based on the high-dimensional word vectors, the inheritance chain path of each tag is serialized into a node sequence; Based on the node sequence, the inheritance chain path similarity is calculated using dynamic time warping or path edit distance algorithms.

[0020] Secondly, the present invention also provides a conflict identification and management device for multi-source heterogeneous industrial data tags, comprising: The label change capture module is used to acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects. The tag inheritance chain generation module is used to construct a tag graph in the form of an inheritance chain structure based on the tag change event object; The tag semantic conflict identification module is used to perform cluster analysis on the tag map and generate a tag conflict report; The tag standardization and merging suggestion module is used to generate rename tags and conflict resolution suggestions for conflicting tags based on a preset rule base and the tag conflict report, or to receive input custom names to change the conflicting tags in the tag conflict report to generate rename tags and conflict resolution suggestions, and to send the rename tags and conflict resolution suggestions to the target user terminal for review. The tag tracing and restoration service module is used to respond to tag query commands and display the tag change path and tag information of the target tag in the renamed tag.

[0021] Thirdly, the present invention also provides an electronic device, including a memory and a processor, wherein, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the conflict identification and management method for multi-source heterogeneous industrial data tags as described in any of the preceding claims.

[0022] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the conflict identification and management method for multi-source heterogeneous industrial data tags as described in any of the preceding claims.

[0023] The beneficial effects of adopting the above implementation method are as follows: The conflict identification and management method and device for multi-source heterogeneous industrial data tags provided by this invention constructs a tag graph in the form of an inheritance chain structure based on the tag change event object, that is, constructs a tag inheritance chain model. This can record every change and evolution path of the tag in multiple systems, allowing users to clearly view the source, change process, and usage history of the tag, thereby improving the interpretability and controllability of data tags. It helps to quickly locate the source of data anomalies, supports problem tracing, quality auditing, and compliance supervision, and is particularly suitable for industry scenarios with high requirements for data accuracy, such as process manufacturing, energy, and power.

[0024] By performing cluster analysis on the tag graph, a tag conflict report is generated. Based on a preset rule base and the tag conflict report, renaming tags and conflict resolution suggestions are generated for conflicting tags. Alternatively, custom names are received to modify conflicting tags in the tag conflict report to generate renamed tags and conflict resolution suggestions. The renamed tags and conflict resolution suggestions are then combined to achieve a unified representation of tag semantics across different systems, solving the semantic fragmentation problem caused by differences in naming, language, and units. This enhances data asset integration capabilities, facilitates the construction of a unified industrial data knowledge graph, and supports the high requirements for tag consistency and structure in scenarios such as data platforms and large industrial models.

[0025] Furthermore, this invention automatically detects conflicts such as homonyms, synonyms, and inconsistent units between labels, and provides conflict resolution suggestions, i.e., governance suggestions, supporting label standardization. This significantly reduces the workload of manually sorting out label conflicts, lowers the risk of data misinterpretation and model mistraining caused by label ambiguity, and improves the accuracy and reliability of industrial intelligent analysis and prediction models.

[0026] Furthermore, through structured management of tag inheritance chains, this invention enables enterprises to systematically classify and version-manage their data tags, achieving semantically consistent data asset catalog construction. This enhances data reusability, significantly improves data governance efficiency, supports cross-departmental and cross-system data sharing and collaborative modeling, and increases the lifecycle value of data assets.

[0027] This invention significantly reduces the manual work of sorting out tag lineages and naming conventions by automating tag identification, chain building, conflict analysis, and governance suggestion mechanisms, thereby improving the efficiency and accuracy of tag governance. It helps enterprises build a low-cost, high-quality industrial data governance system, unlocking data value, shortening the response cycle from data collection to business decision-making, and improving the overall efficiency of digital transformation.

[0028] In summary, this invention addresses the significant shortcomings of existing technologies in terms of industrial data tag lifecycle management, semantic consistency assurance, and cross-system traceability capabilities. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A flowchart illustrating an embodiment of the conflict identification and management method for multi-source heterogeneous industrial data tags provided by the present invention; Figure 2 This is a schematic block diagram of an embodiment of the conflict identification and management device for multi-source heterogeneous industrial data tags provided by the present invention; Figure 3 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] In the description of the embodiments of this application, unless otherwise stated, "a plurality of" means two or more.

[0033] In this embodiment of the invention, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product or device that includes a series of steps or modules is not necessarily limited to those steps or modules that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product or device.

[0034] The naming or numbering of steps in the embodiments of the present invention does not mean that the steps in the method flow must be executed in the time / logical order indicated by the naming or numbering. The execution order of the named or numbered process steps can be changed according to the technical purpose to be achieved, as long as the same or similar technical effect can be achieved.

[0035] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0036] This invention provides a conflict identification and management method and apparatus for multi-source heterogeneous industrial data tags, which will be described below.

[0037] This invention provides a conflict identification and management method for multi-source heterogeneous industrial data tags. The method can be implemented by executing an application program on a computer device, such as... Figure 1 As shown, the method includes: S101. Obtain the label change event information of the industrial data stream and convert the label change event information into a structured label change event object.

[0038] It is understandable that the triggering behavior of S101 can be: when the data label changes during data acquisition, ETL process, data interface or inter-system transmission (such as renaming, unit conversion, field splitting, aggregation, etc.).

[0039] Capture the tag names before and after the change, record the change type (rename, abstract, unit conversion, etc.), collect metadata such as timestamp, operator, source system, and target system, and output a structured tag change event object.

[0040] S102. Based on the label change event object, construct a label graph in the form of an inheritance chain structure.

[0041] It is understandable that the triggering behavior of S102 could be: for each new or updated tag change event, the system synchronously updates the inheritance chain structure of the corresponding tag.

[0042] S103. Perform cluster analysis on the tag map to generate a tag conflict report.

[0043] Understandably, a tag conflict report includes the conflicting tag pair, the type of conflict, and the reason for the conflict.

[0044] S104. Based on the preset rule base and the tag conflict report, generate rename tags and conflict handling suggestions for the conflicting tags; or, receive the input custom name to change the conflicting tags in the tag conflict report to generate rename tags and conflict handling suggestions, and send the rename tags and conflict handling suggestions to the target user terminal for review.

[0045] It is understandable that the renamed tags and conflict resolution suggestions are sent to the target user's terminal for review. If the user provides modification suggestions, then modifications are made based on those suggestions.

[0046] S105. In response to the tag query command, display the tag change path and tag information of the target tag in the renamed tag.

[0047] Understandably, in order to address the problems of semantic loss, frequent conflicts, and lack of traceability in the management of multi-system data tags in existing Industrial Internet of Things (IIoT) platforms, this invention proposes an inheritance chain management and conflict identification method for multi-source heterogeneous industrial data tags. The aim is to build a unified platform mechanism that makes the tag evolution process visible, the tag semantics controllable, the tag conflicts identifiable, and the tag governance sustainable.

[0048] The main objective of this invention is: I. Construct a tag inheritance chain to achieve full lifecycle traceability of tags. By automatically capturing the tag change behavior of data in various business systems, a "tag inheritance chain" structure is constructed to record the evolution path of tags, including operations such as renaming, unit conversion, abstraction aggregation, splitting and mapping, to ensure the integrity and reproducibility of tag information in the process of flowing through multiple systems.

[0049] II. Identify and manage semantic conflicts in tags, and provide intelligent merging suggestions. By leveraging technologies such as semantic recognition, rule matching, and knowledge graphs, it automatically identifies conflicts such as homonyms and synonyms between tags, and provides standardized tag recommendations and merging suggestions based on tag attributes and context, thus resolving the problem of inconsistent tag definitions across multiple systems.

[0050] III. Enhance the platform's unified governance capabilities for industrial data semantics. By integrating the tag inheritance chain mechanism into the data governance system of the IIoT platform, the platform's semantic consistency control capabilities throughout the entire process of data collection, processing, analysis, modeling, and visualization are enhanced, supporting accurate tag tracing and semantic assurance in scenarios such as data auditing, model training, and quality traceability.

[0051] IV. Improving the quality of industrial data and the manageability of data assets By systematically managing the label change path, the semantic clarity, interpretability, and reusability of industrial data are significantly improved, providing important support for enterprises to build a high-quality, highly reliable, and highly governed data asset system.

[0052] In summary, this invention aims to introduce a novel mechanism for tag semantic lifecycle management at the IIoT platform layer, solving problems such as data understanding difficulties, analysis errors, and decision-making mistakes caused by insufficient tag management in the current integration of multi-source heterogeneous data, and providing a solid semantic foundation for industrial enterprises to achieve data-driven intelligent decision-making.

[0053] In some embodiments, the label change event information includes: the label name before and after the change, the change type, and metadata related to the label change event.

[0054] It is understandable that the change types include: renaming, abstraction, unit conversion, etc.; the metadata includes: timestamp, operator, source system, target system, etc.

[0055] In some embodiments, the label change event object includes: whether the label has been changed and the type of change; Based on the aforementioned tag change event object, a tag graph in the form of an inheritance chain structure is constructed, including: By constructing nodes in the inheritance chain based on the changed labels, directed edges are built between the nodes according to the change type, resulting in a label graph with an inheritance chain structure.

[0056] Understandably, node construction involves treating each change to a tag as a node on the chain and saving the tag's state. Edge construction: Establish directed edges of "change type" between nodes; The build process supports forking (such as when the same tag is referenced by multiple systems) and merging (such as when multiple tags are unified); Output content: Tag inheritance chain (stored in graph structure).

[0057] In some embodiments, the tag map includes: tag name similarity, contextual attributes, unit type, and the system to which the tag belongs; the tag conflict report includes conflicting tag pairs, conflict type, and conflict reason.

[0058] It is understandable that cluster analysis is performed based on dimensions such as label name similarity, contextual attributes, unit type, and system to which it belongs; to identify synonyms (such as "compression temperature" and "processing area temperature") or homonyms (such as "Temp" having different meanings in different scenarios).

[0059] In some embodiments, cluster analysis is performed on the tag map to generate a tag conflict report, including: Extract metadata information from the tag map; Convert the metadata information into high-dimensional word vectors; Based on the high-dimensional word vectors, the semantic similarity, contextual feature similarity, and inheritance chain path similarity between tags are calculated; Based on the semantic similarity, the contextual feature similarity, and the inheritance chain path similarity, multidimensional semantic fusion and conflict determination are performed on the labeled images to generate a label conflict report.

[0060] It is understandable that Chinese semantic vector models such as BERT / RoBERTa / SimCSE are used to convert the metadata information into high-dimensional word vectors.

[0061] The closer the semantic similarity value is to 1, the more likely the two tags are to be synonyms.

[0062] In some embodiments, calculating the contextual feature similarity between tags based on the high-dimensional word vectors includes: Extract the contextual attributes of each tag based on the high-dimensional word vectors; The contextual attributes of each tag are weighted and summed to obtain the contextual feature similarity between tags.

[0063] Understandably, the contextual attributes of each tag are extracted, such as equipment type, process segment, system source, and unit, and the contextual similarity is calculated by weighted summation.

[0064] In some embodiments, calculating the inheritance chain path similarity between tags based on the high-dimensional word vectors includes: Based on the high-dimensional word vectors, the inheritance chain path of each tag is serialized into a node sequence; Based on the node sequence, the inheritance chain path similarity is calculated using dynamic time warping or path edit distance algorithms.

[0065] Understandably, inheritance chain path similarity analysis For each tag's inheritance chain, its path is serialized into a sequence of nodes:

[0066] Calculate path similarity using either Dynamic Time Warping (DTW) or Path Edit Distance (PED) algorithms:

[0067] If the path similarity is high (>0.8), it indicates that the semantic evolution of the two labels is similar.

[0068] In some embodiments, the core technical principle of the multi-source heterogeneous industrial data tag inheritance chain management and conflict identification method proposed in this invention is: in an industrial Internet of Things platform, by automatically capturing and structurally modeling the change process of data tags among multiple systems, a path of all evolution paths that a tag undergoes in its life cycle is constructed, and based on this, the consistency management, conflict detection, semantic merging and traceability query of tag semantics are realized.

[0069] This invention is mainly based on the following four core technical principles: I. Principles of Tag Change Event Capture and Metadata Modeling During data flow through various business systems (such as SCADA, MES, ERP, QMS), tags may be renamed, have their units changed, undergo semantic abstraction, be aggregated, or have their mappings split. This invention captures these change events by deploying a tag change monitoring mechanism (such as API Hook, ETL log monitoring, data flow middleware, etc.) within the platform, and generates standardized tag change metadata records for each change, including: Original tag name and changed tag name; Tag change operation types (rename, transform, aggregate, split, etc.); Data source system and target system; Operator, timestamp, change description, etc.

[0070] This metadata forms an important foundation for the subsequent construction of the tag inheritance chain.

[0071] II. Construction Principles of Tag Lineage Chain This invention treats each data tag as a "semantic node," and through its modification operations in various systems, connects multiple nodes in a directed edge chain to construct a complete tag evolution path, i.e., an "inheritance chain." This inheritance chain has the following characteristics: Structurally, it presents itself as a directed graph (DAG), which can be regarded as a label evolution tree; Each node represents a label status; Each edge represents a label change operation, along with the changed metadata; Supports branch merging (e.g., aggregating multiple tags into one) and splitting (e.g., splitting a tag into multiple tags).

[0072] The formation of the inheritance chain enables the platform to fully restore the historical source path of the current state of any tag.

[0073] III. Principles of Label Conflict Identification and Semantic Merging The platform contains a large number of tags from different systems, which may include: Synonyms with different names: for example, "compression temperature", "heat treatment temperature", "Zone1.Temperature"; Same name, different meaning: For example, "Temp" has different meanings in production lines A and B.

[0074] This invention introduces a tag semantic recognition algorithm, which combines the following information to determine conflicts: The tag conflict detection module is based on the "Multi-Dimensional Semantic Conflict Detection Algorithm (MDSCDA)". It integrates four technical means: natural language processing (NLP), context feature extraction, knowledge graph comparison and historical inheritance chain analysis, to automatically complete the detection of synonyms and homonyms between tags and generate tag semantic merging suggestions.

[0075] The specific steps are as follows: Step 1: Preparation of Label Feature Data Extract the following metadata information from the tag repository or inheritance chain database: Tag Name; Belongs to (SourceSystem); Unit; Context (equipment type, process step); Historical inheritance chain path (including upstream and downstream nodes); User comments and usage frequency.

[0076] Step 2: Tag Text Semantic Vectorization (NLP Semantic Encoding) Using Chinese semantic vector models such as BERT / RoBERTa / SimCSE, the TagName and user annotations are transformed into high-dimensional word vectors:

[0077] Calculate the semantic similarity between two tags:

[0078] The closer the value is to 1, the more likely it is to be a synonym tag.

[0079] Step 3: Contextual Feature Similarity Extraction Extract the contextual attributes of each tag (equipment type, process segment, system origin, unit, etc.) and calculate the contextual similarity:

[0080] Among them each For similarity functions based on rules or word vectors, the weights Given by training data or expert rules.

[0081] Step 4: Inheritance chain path similarity analysis For each tag's inheritance chain, its path is serialized into a sequence of nodes:

[0082] Calculate path similarity using either Dynamic Time Warping (DTW) or Path Edit Distance (PED) algorithms:

[0083] If the path similarity is high (>0.8), it indicates that the semantic evolution of the two labels is similar.

[0084] Step 5: Multidimensional Semantic Fusion and Conflict Determination Similarity across three dimensions:

[0085] in: Training or experience-based settings for (human-judged samples) using historical tags from the platform; Appropriate weights can be learned automatically using logistic regression or lightweight neural networks.

[0086] Set threshold: when It was determined to be a "synonym"; when and It was determined to be "homonym"; The rest are "no conflict".

[0087] Step 6: Generation of Merging and Standardization Recommendations For detected conflicting tag pairs, the system generates unified suggestions based on the rule engine and enterprise naming conventions: Regarding "synonyms": Select the pair with the highest confidence and the most frequent use from the conflict pairs as the standard label; Add the remaining tags to the AliasList; Generate a tag mapping table: {original tag name → standard name}; Regarding "homophones with different meanings": Automatically append context prefixes to the naming, such as: Temp → A_Temp (Production Line A); Generate governance recommendations for the following objects:

[0088] Step 7: Manual Confirmation and Automatic Execution The system can automatically merge suggestions with a confidence level higher than a set threshold (e.g., 0.9). Results with lower confidence levels are sent to the labeling management center for expert confirmation. Once approved, the system will update the tag inheritance chain and mark the processing record.

[0089] Once a potential conflict is identified, the system will prompt the user and provide suggestions for merging or standardizing tags, supporting manual confirmation or automatic merging.

[0090] IV. Safety and Systemic Principles Utilizing the tag inheritance chain structure, the platform can provide a semantic restoration path for any tag, including: The original label source (e.g., a variable from a specific DCS); All intermediate change operations and system nodes involved; The semantic meaning, unit, and data operation type of the current tag.

[0091] Application scenario example: Quality problem tracing and anomaly analysis.

[0092] Application Background: During material production or equipment assembly, the final quality control system (such as QMS) may report "abnormal heat treatment temperature" or "low welding quality." Analysts need to determine whether the problem stems from abnormal sensor data, data transmission errors, or incorrect process settings.

[0093] Tag inheritance chain application: The historical corresponding tags in the MES, SCADA, and DCS systems can be retrieved from the tag "Heat_Temp_Final" in the QMS system. Show the complete change path: DCS.TagA(Temp_sensor_01℃) → SCADA.Zone1_Temp(℃) → MES.Heat_Process_Temp(℃) → QMS.Heat_Temp_Final(decision value).

[0094] The system automatically identifies the unit conversion and semantic abstraction stages, and points out the conversion node when temperature changes from a continuous variable to "qualified / unqualified".

[0095] The present invention has the following technical effects: It enables visualized problem localization, directly locating the system or responsible link that causes data anomalies, saving manual troubleshooting costs and improving fault diagnosis efficiency.

[0096] Application Scenario 2: Data Auditing and Compliance Tracking Application Background: In industries such as pharmaceuticals, petrochemicals, and power, industrial regulatory authorities require traceability and filing of key production data. In particular, when core parameters such as process temperature, pressure, and proportions are involved, it is necessary to be able to reconstruct the source of data collection and the entire processing process.

[0097] Tag inheritance chain application: When auditors query the "Drug_Mixer_Temp" tag, the following will be automatically displayed: Source device (DCS data acquisition point number); Data collection time; Intermediate system processing actions (aggregation, unit conversion, outlier removal); The algorithm that generated the data currently being displayed.

[0098] The platform generates an exportable "data lineage report" that meets the compliance requirements of ISO / IEC or drug regulatory systems.

[0099] Technical effects: Ensure the auditability of all business tags and the reliability of data sources to meet the "data verifiability" requirement in compliance scenarios.

[0100] This ability to reconstruct data allows data analysts, operations and maintenance personnel, and auditors to accurately understand the meaning of the current tags and their upstream and downstream relationships, thereby improving the credibility and interpretability of data usage.

[0101] The multi-source heterogeneous industrial data tag inheritance chain management and conflict identification method of the present invention is built on an industrial Internet of Things (IIoT) platform and mainly consists of the following functional modules: I. The system components are shown in Table 1: Table 1: System Components

[0102] II. Logical Relationships and Data Flow Between Modules The various functional modules of this invention work collaboratively according to the lifecycle sequence of data tags, forming a closed-loop tag semantic tracing and governance process. The following is a description of the specific interaction logic and data flow between the modules: Tag change capture module → Tag inheritance chain generation module; Triggering behavior: When data labels change during data acquisition, ETL process, data interface, or inter-system transmission (such as renaming, unit conversion, field splitting, aggregation, etc.). Data processing: Capture the tag names before and after the change; Record the change type (rename, abstract, unit conversion, etc.); Collect metadata such as timestamps, operators, source systems, and target systems; Output content: a structured tag change event object; Transmission method: The event is pushed to the tag inheritance chain generation module for structured modeling.

[0103] Tag inheritance chain generation module → Tag semantic conflict identification module; Triggering behavior: Whenever a new or updated tag change event is added, the system synchronously updates the inheritance chain structure of the corresponding tag; Data processing: Node construction: Treat each change of the tag as a node on the chain and save the tag state; Edge construction: Establish directed edges of "change type" between nodes; Supports forking (such as when the same tag is referenced by multiple systems) and merging (such as when multiple tags are unified); Output content: Tag inheritance chain (stored in graph structure); Transmission method: The label image is synchronously updated to the conflict identification module, supporting periodic conflict detection tasks.

[0104] Tag semantic conflict identification module → Tag standardization and merging suggestion module; Triggering behavior: Periodically scan the existing tag graph, or actively trigger when a user queries / governance operations; Data processing: Cluster analysis is performed based on dimensions such as tag name similarity, contextual attributes, unit type, and system to which they belong. Identify synonyms (such as "compression temperature" and "processing zone temperature") or homonyms (such as "Temp" having different meanings in different scenarios). Output content: Tag conflict report, including conflicting tag pairs, conflict type, and conflict reason; Delivery method: The conflict report is delivered to the standardized merging module for analysis and suggestion generation.

[0105] Label Standardization and Merging Suggestions Module → Label Governance Center; Triggering behavior: It works automatically upon receiving a tag conflict report, or a tag unification request can be manually initiated by the user; Data processing: Recommended standard tag naming (based on rule base, thesaurus, and industry standards); Provide merging suggestions (such as merging multiple tags into "HeatZone1_Temp"); Supports user-defined naming schemes or selection of platform suggestions; Output content: Executable tag name renaming / merging template; Conflict resolution recommendations record (including the replaced label, replacement method, person in charge, time, etc.); Delivery method: User review, confirmation, and execution are conducted through the label governance center.

[0106] Tag traceability and restoration service module ← Tag inheritance chain generation module; Triggering behavior: When a user clicks a tag in the analytics platform or calls the traceability API; Data processing: Query all upstream nodes in the inheritance chain for the tag; Revert all changes made (such as original labels, unit changes, and each renaming). Output a visualization path (timeline or chain structure); Output content: Tag change path (including time, system, operation type, and operator); Tag semantic explanation, version information, etc.; Transmission method: Provided for front-end visualization components; Alternatively, it can be used as interface data for external systems to call (such as data analysis platforms and AI training platforms using labels to restore information).

[0107] Tag Management Center (Unified Operation Portal) effect: It aggregates the inputs and outputs of all modules to provide a unified interactive interface; Displays tag inheritance chain graph, semantic conflict warnings, governance suggestions, and tag tracking paths; Users can audit and view the tag change history, confirm processing suggestions, and initiate tag merging. Interaction direction: Receives: metadata, change events, and analysis reports from each submodule; Distribution: Governance instructions, confirmation operations, and naming conventions are pushed to other modules of the platform.

[0108] III. System Deployment Architecture The system of this invention is mainly deployed in the data governance module of an industrial Internet of Things platform, and is divided into three levels according to function, as shown in Table 2: Table 2: Functional Division Hierarchy

[0109] IV. System Technical Features The system design of this invention has the following significant features in terms of technical implementation: Cross-system tag change tracking capability: It can automatically identify the mapping, renaming and conversion relationships of tags in multiple systems (such as DCS, MES, ERP); it can realize the full path recording and visual tracking of tags from the original collection point to the final business system.

[0110] Tag inheritance chain graph structure modeling: It is the first to model the tag evolution process with a graph structure (nodes + edges) to construct a "tag evolution tree" or "tag lineage chain"; It supports efficient storage and path query in graph databases, which facilitates tag traceability analysis.

[0111] Tag semantic conflict identification mechanism: uses multimodal information (tag semantics, unit, context, system source) to determine conflicts; supports combinations of multiple methods such as rule engine, semantic lexicon, and NLP (such as BERT, word vectors); Automated Tag Governance Recommendation Engine: For conflicting tags, it automatically recommends standard naming, naming templates, or tag merging methods; it also provides functions such as reviewing, implementing, and version management of governance recommendations.

[0112] Tag semantic restoration and transparency capabilities: Any tag can be traced through an interface or web interface to track its historical changes; information such as the tag's source system, change time, change operation, and responsible person can be displayed. The implementation process in this embodiment is as follows: The implementation process of this invention revolves around five major steps: "capturing tag changes → constructing inheritance chains → identifying conflicts → generating governance suggestions → tag tracing and visualization," forming a closed-loop tag semantic management process. The specific implementation process is as follows: I. Tag change event capture, as shown in Table 3: Table 3: Label Changes

[0113] II. Tag inheritance chain construction and updating, as shown in Table 4: Table 4: Tag Inheritance Chain Construction and Update

[0114] III. Identification of semantic conflicts in tags, as shown in Table 5: Table 5: One of the methods for identifying semantic conflicts in labels

[0115] IV. Identification of semantic conflicts in tags, as shown in Table 6: Table 6: One of the methods for identifying semantic conflicts in labels

[0116] V. Tag semantic tracing and visualization, as shown in Table 7: Table 7: Tag Semantic Tracing and Visualization

[0117] VI. Feedback on governance results and updates to the process chain, as shown in Table 8: Table 8: Governance Result Feedback and Link Update

[0118] The present invention has the following beneficial effects: The proposed method for managing and identifying multi-source heterogeneous industrial data tag inheritance chains has significant technical advantages and practical value in industrial data governance practices. Achieve traceability management throughout the entire lifecycle of labels This invention constructs a tag inheritance chain model to record every change and evolution path of tags across multiple systems, allowing users to clearly view the origin, change process, and usage history of tags, thus improving the interpretability and controllability of data tags. It helps to quickly locate the source of data anomalies, supports problem tracing, quality auditing, and compliance supervision, and is particularly suitable for industry scenarios with high data accuracy requirements, such as process manufacturing, energy, and power.

[0119] Supports consistent modeling of tag semantics across multiple systems By automatically identifying and modeling cross-system changes in tags, this invention achieves a unified representation of tag semantics across different systems, solving the semantic fragmentation problem caused by differences in naming, language, and units. It enhances data asset integration capabilities, facilitates the construction of a unified industrial data knowledge graph, and supports the high requirements for tag consistency and structure in scenarios such as data platforms and large-scale industrial models.

[0120] Automatically identify semantic conflicts to assist in tag management This invention utilizes semantic recognition algorithms and a rule engine to automatically detect conflicts such as homonyms, synonyms, and inconsistent units between tags, and provides governance suggestions to support standardized tag processing. It significantly reduces the workload of manually sorting out tag conflicts, lowers the risk of data misinterpretation and model mistraining caused by tag ambiguity, and improves the accuracy and reliability of industrial intelligent analysis and prediction models.

[0121] Enhance the manageability and reusability of data assets. Through structured management of tag inheritance chains, enterprises can systematically classify and version-manage their data tags, achieving semantically consistent data asset catalog construction. This enhances data reusability, significantly improves data governance efficiency, supports cross-departmental and cross-system data sharing and collaborative modeling, and increases the lifecycle value of data assets.

[0122] Effectively reduce enterprise data governance costs This invention significantly reduces the manual work of sorting out tag lineages and naming conventions by automating tag identification, chain building, conflict analysis, and governance suggestion mechanisms, thereby improving the efficiency and accuracy of tag governance. It helps enterprises build a low-cost, high-quality industrial data governance system, unlocking data value, shortening the response cycle from data collection to business decision-making, and improving the overall efficiency of digital transformation.

[0123] Explanation of innovation points: A "tag inheritance chain" structure is proposed to record the tag evolution process: This invention innovatively introduces a "Tag Inheritance Chain" structure as a modeling carrier for the semantic evolution of data tags. Based on a graph model, this structure abstracts the process of tag changes between systems into a chain of node relationships. Each node represents the state of a tag at a certain moment / system, and each edge represents a specific change event (such as renaming, unit transformation, semantic abstraction, etc.). Through this structure, the source path, change history, and evolutionary logic of tags can be fully reconstructed.

[0124] Compared to traditional tag dictionaries or static mapping methods, the tag inheritance chain of this invention is scalable, traceable, and versionable, and can dynamically adapt to the actual environment of frequent changes in industrial systems and normalized tag reconstruction, filling the technical gap in existing data platforms in tag semantic lineage management.

[0125] A label change capture mechanism and event logging model are proposed: This invention further designs a tag change capture mechanism that can automatically identify tag change behaviors in data flow, interface calls, ETL processing, and other stages, capturing various change types including tag name changes, unit adjustments, field merging / splitting, and application semantic transformation. It also proposes a structured tag change event model to standardize the recording of the evolution details of each tag, including fields such as the state before and after the change, the operator, timestamp, system source, and change type.

[0126] This mechanism enables label governance to shift from "static management" to "dynamic perception," possessing real-time capability, auditability, and behavioral explainability, thus providing a solid foundation for subsequent construction of inheritance chains, conflict identification, and semantic tracing.

[0127] Intelligent identification and recommendation of tag conflicts by integrating semantic recognition algorithms: This invention integrates Natural Language Processing (NLP) and rule engine technologies to develop an algorithm module for identifying semantic conflicts in tags. This module comprehensively considers multi-dimensional features such as tag name, unit, context, system location, and usage frequency. Through methods such as word vector matching, semantic similarity calculation, and hyponym / hypernym recognition, it automatically identifies semantic conflicts such as synonyms, homonyms, and inconsistent units. Furthermore, it combines enterprise naming conventions to generate governance suggestions (such as tag merging, replacement, retention, or marking discrepancies).

[0128] This innovation significantly enhances the intelligence level of label governance, changing the inefficient model that previously relied on manual comparison and review, and enabling label conflict identification and governance to enter the stage of intelligent recommendation and semi-automatic processing, thereby improving governance efficiency and quality.

[0129] For the first time, cross-system semantic traceability capability for data tags has been achieved in an IIoT platform: Existing industrial data platforms generally lack the ability to systematically track the semantic changes of tags, especially when facing multi-system collaboration (such as DCS, MES, ERP, etc.), where issues such as tag fragmentation, renaming, and semantic drift between systems are difficult to resolve. This invention is the first to achieve cross-system semantic tracing of data tags in an Industrial Internet of Things (IIoT) platform, enabling the retrieval of the historical state, evolution path, and source system of any tag across multiple systems, effectively solving the industry pain points of "unknown origin" and "uncontrollable semantics" of tags.

[0130] This capability not only enhances the platform's data transparency and interpretability, but also provides strong support for scenarios such as data problem localization, anomaly tracing, and model auditability, and has significant practical application value and promotion prospects.

[0131] like Figure 2 As shown, the present invention also provides a conflict identification and management device 200 for multi-source heterogeneous industrial data tags, comprising: The label change capture module 201 is used to acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects. The tag inheritance chain generation module 202 is used to construct a tag graph in the form of an inheritance chain structure based on the tag change event object; The tag semantic conflict identification module 203 is used to perform cluster analysis on the tag map and generate a tag conflict report; The tag standardization and merging suggestion module 204 is used to generate rename tags and conflict resolution suggestions for conflicting tags based on a preset rule base and the tag conflict report, or to receive input custom names to change the conflicting tags in the tag conflict report to generate rename tags and conflict resolution suggestions, and to send the rename tags and conflict resolution suggestions to the target user terminal for review. The tag tracing and restoration service module 205 is used to display the tag change path and tag information of the target tag in the renamed tag in response to the tag query command.

[0132] The conflict identification and management device for multi-source heterogeneous industrial data tags provided in the above embodiments can realize the technical solutions described in the above embodiments of the conflict identification and management method for multi-source heterogeneous industrial data tags. The specific implementation principles of each module or unit can be found in the corresponding content in the above embodiments of the conflict identification and management method for multi-source heterogeneous industrial data tags, and will not be repeated here.

[0133] like Figure 3 As shown, the present invention also provides an electronic device 300. The electronic device 300 includes a processor 301, a memory 302, and a display 303. Figure 3 Only some components of the electronic device 300 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0134] In some embodiments, memory 302 may be an internal storage unit of electronic device 300, such as a hard disk or memory of electronic device 300. In other embodiments, memory 302 may also be an external storage device of electronic device 300, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 300.

[0135] Furthermore, the memory 302 may include both internal storage units of the electronic device 300 and external storage devices. The memory 302 is used to store application software and various types of data installed on the electronic device 300.

[0136] In some embodiments, processor 301 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 302 or process data, such as the conflict identification and management method for multi-source heterogeneous industrial data tags in this invention.

[0137] In some embodiments, display 303 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 303 is used to display information from electronic device 300 and to display a visual user interface. Components 301-303 of electronic device 300 communicate with each other via a system bus.

[0138] In some embodiments of the present invention, when the processor 301 executes the conflict identification and management program for multi-source heterogeneous industrial data tags in the memory 302, the following steps can be implemented: Acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects; Based on the aforementioned tag change event object, construct a tag graph in the form of an inheritance chain structure; Cluster analysis is performed on the tag map to generate a tag conflict report; Based on the preset rule base and the tag conflict report, rename tags and conflict handling suggestions are generated for conflicting tags; or, custom names are received to modify the conflicting tags in the tag conflict report to generate rename tags and conflict handling suggestions, and the rename tags and conflict handling suggestions are sent to the target user terminal for review. In response to the tag query command, the tag change path and tag information of the target tag in the renamed tag are displayed.

[0139] It should be understood that when the processor 301 executes the conflict identification and management program for multi-source heterogeneous industrial data tags in the memory 302, in addition to the functions mentioned above, it can also perform other functions, as detailed in the description of the corresponding method embodiments above.

[0140] Furthermore, the embodiments of the present invention do not specifically limit the type of electronic device 300 mentioned. Electronic device 300 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, electronic device 300 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0141] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the conflict identification and management method for multi-source heterogeneous industrial data tags provided by the methods described above, the method comprising: Acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects; Based on the aforementioned tag change event object, construct a tag graph in the form of an inheritance chain structure; Cluster analysis is performed on the tag map to generate a tag conflict report; Based on the preset rule base and the tag conflict report, rename tags and conflict handling suggestions are generated for conflicting tags; or, custom names are received to modify the conflicting tags in the tag conflict report to generate rename tags and conflict handling suggestions, and the rename tags and conflict handling suggestions are sent to the target user terminal for review. In response to the tag query command, the tag change path and tag information of the target tag in the renamed tag are displayed.

[0142] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0143] The conflict identification and management method and apparatus for multi-source heterogeneous industrial data tags provided by the present invention have been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A conflict identification and management method for multi-source heterogeneous industrial data tags, characterized in that, include: Acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects; Based on the aforementioned tag change event object, construct a tag graph in the form of an inheritance chain structure; Cluster analysis is performed on the tag map to generate a tag conflict report; Based on the preset rule base and the tag conflict report, rename tags and conflict handling suggestions are generated for conflicting tags; or, custom names are received to modify the conflicting tags in the tag conflict report to generate rename tags and conflict handling suggestions, and the rename tags and conflict handling suggestions are sent to the target user terminal for review. In response to the tag query command, the tag change path and tag information of the target tag in the renamed tag are displayed.

2. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 1, characterized in that, The label change event information includes: the label name before and after the change, the change type, and metadata related to the label change event.

3. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 1, characterized in that, The label change event object includes: whether the label has been changed and the type of change; Based on the aforementioned tag change event object, a tag graph in the form of an inheritance chain structure is constructed, including: By constructing nodes in the inheritance chain based on the changed labels, directed edges are built between the nodes according to the change type, resulting in a label graph with an inheritance chain structure.

4. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 1, characterized in that, The tag map includes: tag name similarity, context attributes, unit type, and the system to which the tag belongs; the tag conflict report includes conflicting tag pairs, conflict type, and conflict reason.

5. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 1, characterized in that, Cluster analysis is performed on the tag map to generate a tag conflict report, including: Extract metadata information from the tag map; Convert the metadata information into high-dimensional word vectors; Based on the high-dimensional word vectors, the semantic similarity, contextual feature similarity, and inheritance chain path similarity between tags are calculated; Based on the semantic similarity, the contextual feature similarity, and the inheritance chain path similarity, multidimensional semantic fusion and conflict determination are performed on the labeled images to generate a label conflict report.

6. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 5, characterized in that, Calculating the contextual feature similarity between tags based on the high-dimensional word vectors includes: Extract the contextual attributes of each tag based on the high-dimensional word vectors; The contextual attributes of each tag are weighted and summed to obtain the contextual feature similarity between tags.

7. The conflict identification and management method for multi-source heterogeneous industrial data tags according to claim 5, characterized in that, The similarity of inheritance chain paths between tags is calculated based on the high-dimensional word vectors, including: Based on the high-dimensional word vectors, the inheritance chain path of each tag is serialized into a node sequence; Based on the node sequence, the inheritance chain path similarity is calculated using dynamic time warping or path edit distance algorithms.

8. A conflict identification and management device for multi-source heterogeneous industrial data tags, characterized in that, include: The label change capture module is used to acquire label change event information from industrial data streams and convert the label change event information into structured label change event objects. The tag inheritance chain generation module is used to construct a tag graph in the form of an inheritance chain structure based on the tag change event object; The tag semantic conflict identification module is used to perform cluster analysis on the tag map and generate a tag conflict report; The tag standardization and merging suggestion module is used to generate rename tags and conflict resolution suggestions for conflicting tags based on a preset rule base and the tag conflict report, or to receive input custom names to change the conflicting tags in the tag conflict report to generate rename tags and conflict resolution suggestions, and to send the rename tags and conflict resolution suggestions to the target user terminal for review. The tag tracing and restoration service module is used to respond to tag query commands and display the tag change path and tag information of the target tag in the renamed tag.

9. An electronic device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the conflict identification and management method for multi-source heterogeneous industrial data tags as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the conflict identification and management method for multi-source heterogeneous industrial data tags as described in any one of claims 1 to 7.

Images