A method, apparatus and storage medium for industrial OT data edge governance
By establishing a PGID encoding system and a two-way mapping mechanism at the edge, the fragmentation and disjointed naming problem of industrial OT data has been solved, achieving second-level traceability and unified end-to-end data governance, ensuring data quality and security.
Patent Information
- Application Number
- CN202611126029.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-28
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies, the governance of industrial OT data is fragmented between the edge and the center, resulting in fragmented data naming, insufficient traceability, uncontrolled edge quality, security risks, and inadequate control of illegal locations, making it impossible to achieve unified and standardized data governance and end-to-end traceability.
By establishing a globally unique PGID point coding system and a five-layer mapping rule on the edge side, data format conversion, standardization processing, and quality pre-inspection are achieved. Through the two-way binding mapping between the edge side OT asset catalog and the central side data asset catalog, it is ensured that the same physical measurement point maintains a unique code throughout the entire link. Combined with the three-line defense for edge quality and the breakpoint resume mechanism, real-time data verification and synchronization are achieved.
It enables second-level tracing and location of data from business report anomalies to physical sensor failures, solves the problem of tracing infeasibility caused by data naming fragmentation, avoids high latency and high cost data governance, ensures controllable data quality and security, and achieves unified coding and asset visibility across the entire chain.
Smart Images

Figure CN123115104A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of industrial data processing technology, and more specifically to a method, apparatus, and storage medium for edge governance of industrial OT data. Background Technology
[0002] Existing process manufacturing enterprises have deployed information systems such as ERP, MES, SCADA, and LIMS on a large scale. However, high-frequency time-series data such as temperature, pressure, flow, and vibration from the field OT have long been separated from the upper-level IT operation and management data. The existing data governance system has many technical shortcomings, making it difficult to achieve unified and standardized data governance and full-chain traceability.
[0003] The existing measurement point naming is fragmented, lacking traceability capabilities. The same physical measurement point uses different naming rules across PLCs, OPC gateways, IoT platforms, and data middleware, lacking a unified and unique identifier and hierarchical mapping relationship. When data anomalies occur, it is impossible to reverse-locate physical devices and sensors, resulting in extremely low fault tracing efficiency. Furthermore, the existing edge governance capabilities are lacking, leading to uncontrolled source data quality. Edge gateways only implement protocol forwarding; data is not standardized or quality-verified throughout the entire transmission link, making it impossible to eradicate source data defects. The edge side organizes data based on physical devices, while the central side constructs an asset catalog based on business themes. There is no bidirectional mapping synchronization mechanism, resulting in severe lag and distortion of central asset data after on-site equipment upgrades and point expansions. Fourthly, there is a lack of illegal point control mechanisms, posing data security risks.
[0004] In summary, existing technologies generally suffer from poor industry adaptability, lack of end-to-end coding and mapping capabilities, lack of edge quality governance, fragmentation of edge and cloud assets, and insufficient control over illegal sites, failing to meet the core needs of the steel industry for OT / IT data integration, standardized governance, accurate traceability, and asset control. Summary of the Invention
[0005] The purpose of this invention is to provide a method, apparatus, and storage medium for edge governance of industrial OT data, which enables unified and secure management and control of OT data.
[0006] To achieve the above objectives, embodiments of the present invention provide a method for edge governance of industrial OT data. The method includes: acquiring heterogeneous data from edge devices; converting the format of the heterogeneous data to obtain a first data group; constructing an edge-side OT asset catalog and a central-side data asset catalog based on the edge devices; and performing bidirectional binding mapping between the layers of the edge-side OT asset catalog and the central-side data asset catalog; the edge-side OT asset catalog is layered according to the physical dimension of the devices, including a production line layer, a region layer, an equipment layer, and an OT measurement point layer; the central-side data asset catalog is layered according to the business dimension, including subject domain groups, subject domains, business objects, logical entities, and attributes; assigning PGID codes to OT measurement points in the OT measurement point layer; and performing edge data standardization processing and edge quality pre-inspection on the first data group based on the PGID codes to obtain a second data group for synchronization to the central platform.
[0007] Optionally, the two-way binding mapping includes: the production line layer and the subject domain are bound together, the area layer and the equipment layer are jointly mapped to the business object, and the OT measurement point layer corresponds to the mapping logical entity and attribute.
[0008] Optionally, the method further includes: when the edge-side OT asset catalog undergoes changes such as device addition, deletion, or modification, the central-side data asset catalog is synchronously updated based on a preset bidirectional mapping rule; when the central side adds business indicators based on business needs, the corresponding physical measurement point variables on the edge side are located by reverse retrieval according to the bidirectional binding mapping.
[0009] Optionally, the PGID encoding includes a factory code, a process code, an equipment code, a parameter code, and an instance number; the factory code is used to identify the physical factory, the process code is used to identify the steel production process, the equipment code is used to identify the specific equipment within the process, the parameter code is used to identify the type of physical parameter, and the instance number is used to distinguish multiple sensor instances of the same equipment and the same parameter.
[0010] Optionally, the entire PGID encoding chain includes a device layer, a gateway layer, a physical layer, and a data layer; in the device layer, the PGID is generated as a name according to the character specification; in the gateway layer, the PGID is generated as an identifier separated by dots; in the physical layer, the PGID is generated as an attribute ID separated by underscores; and in the data layer, the PGID is stored as a field value of the ODS patch source layer in its original encoding format.
[0011] Optionally, the edge data standardization processing includes range conversion, dead zone compression, outlier marking, and unit unification processing; the edge quality pre-inspection includes access verification, CRC verification, real-time rule verification, and window statistics verification.
[0012] Optionally, the method further includes: synchronizing the second data group to the central platform through breakpoint resume, offline caching, and incremental reconciliation mechanisms; the incremental reconciliation mechanism includes: the edge gateway periodically reporting the currently online PGID list and the actual number of data entries collected and the average collection frequency of each PGID within the statistical period to the central platform; the central platform generates a difference alarm for PGIDs whose data entry deviation exceeds a preset threshold and automatically triggers a retransmission task.
[0013] On the other hand, this application also proposes an apparatus for edge governance of industrial OT data. The apparatus includes: an acquisition module for acquiring heterogeneous data from edge devices and converting the heterogeneous data into a format to obtain a first data group; a first processing module for constructing an edge-side OT asset catalog and a central-side data asset catalog based on the edge devices, and performing bidirectional binding mapping between the layers of the edge-side OT asset catalog and the central-side data asset catalog; a second processing module for layering the edge-side OT asset catalog according to the physical dimension of the devices, including a production line layer, a region layer, an equipment layer, and an OT measurement point layer, and layering the central-side data asset catalog according to the business dimension, including subject domain grouping, subject domain, business object, logical entity, and attribute; and a third processing module for assigning PGID codes to OT measurement points in the OT measurement point layer, and performing edge data standardization processing and edge quality pre-inspection on the first data group based on the PGID codes to obtain a second data group for synchronization to the central platform.
[0014] Optionally, the two-way binding mapping includes: the production line layer and the subject domain are bound together, the area layer and the equipment layer are jointly mapped to the business object, and the OT measurement point layer corresponds to the mapping logical entity and attribute.
[0015] On the other hand, this application also proposes a machine-readable storage medium storing instructions for causing a machine to execute the industrial OT data edge governance method described above. Through the above technical solutions, this application establishes a globally unique PGID point encoding system and a five-layer mapping rule at the edge where data is generated, ensuring that the same physical measurement point maintains its PGID throughout the entire link. This enables second-level tracing and location of data from abnormal business reports to physical sensor failures, solving the problem of infeasibility of tracing due to fragmented OT data naming in existing technologies. By using three lines of defense for edge quality, it avoids the high latency and high cost of remediation after problematic data flows into the central platform, as is the case in existing technologies. Through a two-way mapping mechanism between the edge-side OT asset catalog and the central-side data asset catalog, changes on the device side are automatically synchronized to the business catalog, and business side requirements are accurately located to physical measurement points. This solves the problems of fragmented catalogs and information lag at both ends in existing technologies, achieving unified encoding, controllable quality, asset visibility, and security control for OT data throughout the entire link from physical sensors to the data platform.
[0016] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description
[0017] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a method for edge governance of industrial OT data according to this application; Figure 2 This is a schematic diagram of the edge-center collaboration mechanism of this application; Figure 3 This is a schematic diagram of the globally unique encoding structure of the PGID point in this application; Figure 4 This is a schematic diagram of the five-layer mapping relationship of PGID in this application; Figure 5 This is a schematic diagram of an industrial OT data edge governance device according to this application.
[0018] Explanation of reference numerals in the attached figures 100- Device for Industrial OT Data Edge Governance; 200 - Acquisition module; 300 - First processing module; 400 - Second processing module; 500 - Third processing module. Detailed Implementation
[0019] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.
[0020] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.
[0021] This invention provides a method for edge governance of industrial OT data, such as... Figure 1 As shown, the method includes: Step S101: Obtain heterogeneous data from the edge device, and perform format conversion on the heterogeneous data to obtain a first data group; Step S102: Construct an edge-side OT asset catalog and a central-side data asset catalog based on the edge device, and perform bidirectional binding mapping on the hierarchy of the edge-side OT asset catalog and the central-side data asset catalog; Step S103: The edge-side OT asset catalog is layered according to the physical dimension of the equipment, including the production line layer, the area layer, the equipment layer, and the OT measurement point layer; the central-side data asset catalog is layered according to the business dimension, including subject domain grouping, subject domain, business object, logical entity, and attribute. Step S104: Assign PGID codes to the OT measurement points in the OT measurement point layer, and perform edge data standardization processing and edge quality pre-inspection on the first data group according to the PGID codes to obtain a second data group for synchronization to the central platform.
[0022] The aforementioned edge devices can be PLCs / DCSs / sensors. A PLC is a programmable logic controller, and the actual data acquisition and control points on field sensors, instruments, and controllers are collectively referred to as OT measurement points. The aforementioned heterogeneous data originates from heterogeneous protocols from PLCs / DCSs / sensors. The format conversion transforms protocols such as OPC UA, Modbus TCP / RTU, MQTT, Profinet, and HTTP into a unified internal format. The first data group includes the unified internal format data after format conversion. Each data entry must contain at least a data acquisition timestamp, edge device identifier (e.g., PLC station number / sensor address), original name or temporary identifier of the OT measurement point, original physical quantity value and original unit, source protocol type marker (OPC UA / Modbus TCP / MQTT, etc.), and the edge gateway's local timestamp at the time of data reception.
[0023] The aforementioned edge-side OT asset catalog is layered according to the physical dimension of equipment, including the production line layer, area layer, equipment layer, and OT measurement point layer. For example, E-L1 production line layer (sintering, ironmaking, steelmaking, continuous casting, rolling), E-L2 area layer (converter section, LF refining section, RH vacuum section), E-L3 equipment layer (No. 1 converter furnace body, No. 1 oxygen lance system, bottom blowing system), and E-L4 measurement point layer (cooling water inlet temperature, etc.).
[0024] The aforementioned central data asset catalog is layered according to business dimensions, including subject domain groups, subject domains, business objects, logical entities, and attributes. For example, L1 subject domain groups (production, sales, procurement), L2 subject domains (sintering, ironmaking, steelmaking themes), L3 business objects (materials, customers, equipment), L4 logical entities (converter process data table), and L5 attributes (oxygen lance inlet water temperature field).
[0025] The aforementioned two-way binding mapping includes: the production line layer and the subject domain are bound together; the region layer and the equipment layer are jointly mapped to business objects; and the OT measurement point layer corresponds to the mapping logical entities and attributes.
[0026] The aforementioned PGID encoding includes a factory code, process code, equipment code, parameter code, and instance number. The factory code identifies the physical factory, the process code identifies the steel production process, the equipment code identifies the specific equipment within the process, the parameter code identifies the type of physical parameter, and the instance number distinguishes multiple sensor instances of the same equipment and the same parameter. The PGID (Point Global ID) encoding is a globally unique code for each point, employing a five-segment hierarchical structure (factory code - process code - equipment code - parameter code - instance number), spanning the entire L0 to L4 link. For example, PGID is assigned globally uniquely through the point master data registration center in the edge governance engine, automatically assigning a PGID code to each OT measurement point. The PGID encoding adopts a five-segment hierarchical structure: factory code - process code - equipment code - parameter code - instance number. The PGID consists of the following components: a factory code (2 letters) identifying the physical factory; a process code (3 letters) identifying the steel production process (e.g., BOF for converter, EAF for electric arc furnace, LF for refining, CCM for continuous casting, HSM for hot rolling); an equipment code (4 alphanumeric characters) identifying the specific equipment within the process, with the first letter indicating the equipment type (e.g., L for oxygen lance, B for furnace body, C for continuous casting flow); a parameter code (4 letters) identifying the type of physical parameter (e.g., TEMP for temperature, PRES for pressure, FLOW for flow rate, VIBR for vibration); and an instance number (2 digits) distinguishing multiple sensor instances of the same parameter on the same equipment. A complete PGID code would be like GG-BOF-L001-TEMP-01, representing the 01st temperature sensor in the 1st oxygen lance of the G steel converter smelting process.
[0027] The aforementioned edge data standardization processing includes range conversion, dead zone compression, outlier marking, and unit unification. For example, based on metadata associated with PGID encoding (range, unit, signal type, etc.), the edge side automatically performs range conversion (converting 4-20mA / 0-10V electrical signals into engineering unit values), dead zone compression (only the beginning and end of continuous sampling points with value changes less than the dead zone threshold are retained, which can reduce the amount of transmitted data by 80%), outlier marking (instantaneous values exceeding the range or with abnormal rates of change are marked and then allowed to pass, such as a temperature jump of 200°C in one second being marked as "suspected sensor failure"), and unit unification (all temperatures are unified to °C, pressure to kPa, and flow rate to Nm³ / h).
[0028] The aforementioned edge quality pre-inspection includes access verification, CRC verification, real-time rule verification, and window statistical verification. For example, three quality defense lines are set up on the edge side: The first defense line (access verification) performs four verifications on each data message frame: message frame integrity verification (CRC verification), timestamp continuity verification (no gaps), PGID validity verification (whitelist comparison), and data format correctness. Data that fails the verification is directly rejected and an "data not found" alarm is generated. The second defense line (real-time rule verification) performs real-time verification on the data value range (whether it is within the sensor's registered range), abrupt change threshold (whether the jump rate is reasonable), and acquisition frequency deviation (whether the deviation between the actual frequency and the registered frequency exceeds the standard). Data exceeding the limit is tagged and allowed to pass (without discarding data) and an alarm is pushed to the central platform simultaneously. The third defense line (window statistical verification) calculates the data integrity rate (actual number of received data entries ÷ expected number of data entries), mean / variance trend, and cross-validation consistency with associated measurement points every 5 minutes within a rolling window. When the integrity rate is lower than 99%, retransmission is automatically triggered, and when the trend is abnormal, an early warning work order is automatically triggered. When the network between the edge side and the central platform is interrupted, the edge gateway automatically starts the local caching mode (capacity ≥ 72 hours of continuous data), and automatically performs breakpoint resume and incremental reconciliation after the network is restored.
[0029] The second data group mentioned above includes governance data after edge data standardization and edge quality pre-inspection. Each data entry contains at least the following: PGID code (globally unique point identifier), standardized physical quantity engineering value (with range conversion and unit unification completed, temperature unified to degrees Celsius, pressure unified to kPa, and flow rate unified to Nm3 / h), standardized collection timestamp (unified to UTC millisecond timestamp), data quality marker bit (indicating whether the data has undergone dead zone compression, whether there are outlier markers, and whether it has passed the three quality defense lines verification), compression marker (indicating whether the data entry is an intermediate omission point of dead zone compression), and quality verification result code (indicating the pass status of each step of access verification / CRC verification / real-time rule verification / window statistics verification). The aforementioned central platform is deployed in the enterprise's central data center as a central data governance platform system. It includes a central data governance platform layer (responsible for receiving the second data group uploaded from the edge side, managing the PGID whitelist, registering the master data of the points, incremental reconciliation, and distributing configurations), a lake warehouse integrated storage base (storing OT data in a layered architecture from the ODS source layer to the DWD detail layer to the DWS summary layer to the ADS application layer), and a data application service layer (providing a unified API interface for upper-layer applications such as leadership dashboards, business report analysis, and AI model training). The central platform is the upstream aggregation point and unified management center for edge side data.
[0030] Through the above technical solutions, this application establishes a globally unique PGID point encoding system and a five-layer mapping rule at the edge where data is generated, ensuring that the same physical measurement point maintains its PGID throughout the entire link. This enables second-level tracing and location of data from abnormal business reports to physical sensor failures, solving the problem of infeasibility of tracing due to fragmented OT data naming in existing technologies. By using three lines of defense for edge quality, it avoids the high latency and high cost of remediation after problematic data flows into the central platform, as is the case in existing technologies. Through a two-way mapping mechanism between the edge-side OT asset catalog and the central-side data asset catalog, changes on the device side are automatically synchronized to the business catalog, and business side requirements are accurately located to physical measurement points. This solves the problems of fragmented catalogs and information lag at both ends in existing technologies, achieving unified encoding, controllable quality, asset visibility, and security control for OT data throughout the entire link from physical sensors to the data platform.
[0031] In one embodiment, the two-way binding mapping includes: the production line layer and the subject domain are bound together, the region layer and the equipment layer are jointly mapped to business objects, and the OT measurement point layer corresponds to the mapping logical entities and attributes.
[0032] like Figure 2 As shown, the edge-side OT asset catalog is divided into four levels from top to bottom according to the physical dimension of the equipment: E-L1 production line layer (such as sintering production line, steelmaking production line) to E-L2 section / area layer (such as converter section, LF refining section) to E-L3 equipment / unit layer (such as No. 1 converter furnace body, No. 1 oxygen lance system) to E-L4 measuring point / variable layer (such as cooling water inlet temperature, oxygen lance cooling water flow rate). Figure 2 The lower half is the central data asset catalog, which is divided into five levels from top to bottom according to business dimensions: L1 subject domain grouping (such as production domain, sales domain) to L2 subject domain (such as steelmaking theme, rolling theme) to L3 business object (such as equipment, materials) to L4 logical entity (such as converter process data table) to L5 attribute (such as oxygen lance inlet water temperature field). Figure 2 The middle section uses bidirectional arrows to show the bidirectional binding mapping relationship between the edge side and the center side: the E-L1 production line layer is bidirectionally bound to the L2 subject domain (the sintering production line corresponds to the sintering subject domain, and the steelmaking production line corresponds to the steelmaking subject domain), the E-L2 section / area layer and the E-L3 equipment / device layer are jointly mapped to the L3 business object, and the E-L4 measurement point / variable layer corresponds to the mapping of L4 logical entities and L5 attributes. Figure 2The right side shows the bidirectional synchronous data flow: from bottom to top, changes to edge-side equipment (such as adding sensors or replacing PLC modules) automatically update the central-side data asset catalog through mapping relationships; from top to bottom, when new indicators are added for central-side business needs, the corresponding physical measurement point variables on the edge side are retrieved and located through mapping relationships. The L1 subject area grouping is a purely central-side management concept; no corresponding hierarchy exists on the edge side. In the bidirectional mapping, a dashed line represents the unidirectional affiliation of this hierarchy. Specifically, the edge-side OT asset catalog is organized from an equipment perspective: E-L1 production line → E-L2 section / area → E-L3 equipment / device → E-L4 measurement point / variable; the central-side data asset catalog is organized from a business perspective: L1 subject area grouping → L2 subject area → L3 business object → L4 logical entity → L5 attribute. The hierarchical binding mapping between the edge-side OT asset catalog and the central-side data asset catalog is bidirectional: E-L1 production line... L2 subject area, E-L2 section / area and E-L3 equipment / unit L3 business objects, E-L4 measurement points / variables L4 logical entities / L5 attributes. The L1 subject domain grouping is a purely central-side concept; no corresponding level is set on the edge side. When edge-side devices change, they are automatically synchronized to the central-side business catalog through bidirectional mapping; when central-side business requirements add new metrics, they are accurately located to the edge-side physical measurement points through bidirectional mapping.
[0033] This method solves the problems of fragmented directories and information lag between the two ends in the existing technology by using a two-way mapping mechanism between the edge-side OT asset directory and the central-side data asset directory. Changes on the device side are automatically synchronized to the business directory, and business side requirements are accurately located to physical measurement points.
[0034] In one embodiment, the method further includes: when the edge-side OT asset catalog is updated by adding, deleting, or modifying devices, the central-side data asset catalog is updated synchronously based on a preset bidirectional mapping rule; when the central side adds business indicators based on business needs, the corresponding physical measurement point variables on the edge side are located by reverse retrieval based on the bidirectional binding mapping.
[0035] For example, when a new oxygen lance vibration sensor is added to the No. 1 converter section of the steelmaking production line, the edge-side OT asset catalog automatically adds the oxygen lance vibration sensor (PGID GG-BOF-L001-VIBR-01) to the E-L2 converter section to the E-L3 No. 1 oxygen lance system. The edge governance engine automatically assigns a PGID code to this sensor and, based on the preset bidirectional mapping rules (mapping the E-L2 converter section plus the E-L3 No. 1 oxygen lance system to the L3 business object equipment), automatically adds the L5 attribute field oxygen lance vibration value to the L4 logical entity converter process data table in the central-side data asset catalog. The ODS source layer and DWD detail layer table structures are updated simultaneously. The entire process is completed automatically within seconds without manual intervention. When the business department proposes to add an average oxygen lance vibration trend analysis index for the converter smelting process, the central platform uses the bidirectional mapping relationship (from the L4 logical entity converter process data table to the E-L4 measuring point layer) to retrieve and locate the edge-side physical measuring point GG-BOF-L001-VIBR-01. It automatically confirms the data acquisition frequency (e.g., 100Hz), data quality status, and network reachability of the measuring point, and sends the data reporting configuration command for the measuring point to the edge gateway, eliminating the need for manual cross-system verification and configuration.
[0036] In one embodiment, the PGID encoding includes a factory code, a process code, an equipment code, a parameter code, and an instance number; the factory code is used to identify the physical factory, the process code is used to identify the steel production process, the equipment code is used to identify the specific equipment within the process, the parameter code is used to identify the type of physical parameter, and the instance number is used to distinguish multiple sensor instances of the same equipment and the same parameter.
[0037] Specifically, such as Figure 3As shown, the PGID (Point Global ID) code is a globally unique code for each point, employing a five-segment hierarchical structure (factory code - process code - equipment code - parameter code - instance number), spanning the entire L0 to L4 link. For example, the PGID is assigned a globally unique code through the point master data registration center in the edge governance engine, automatically assigning a PGID code to each OT measurement point. The PGID code adopts a five-segment hierarchical structure: factory code - process code - equipment code - parameter code - instance number. The PGID consists of the following components: a factory code (2 letters) identifying the physical factory; a process code (3 letters) identifying the steel production process (e.g., BOF for converter, EAF for electric arc furnace, LF for refining, CCM for continuous casting, HSM for hot rolling); an equipment code (4 alphanumeric characters) identifying the specific equipment within the process, with the first letter indicating the equipment type (e.g., L for oxygen lance, B for furnace body, C for continuous casting flow); a parameter code (4 letters) identifying the type of physical parameter (e.g., TEMP for temperature, PRES for pressure, FLOW for flow rate, VIBR for vibration); and an instance number (2 digits) distinguishing multiple sensor instances of the same parameter on the same equipment. A complete PGID code would be like GG-BOF-L001-TEMP-01, representing the 01st temperature sensor in the 1st oxygen lance of the G steel converter smelting process.
[0038] In one embodiment, the entire PGID encoding chain includes a device layer, a gateway layer, a physical layer, and a data layer; in the device layer, the PGID is generated as a name according to the character specification; in the gateway layer, the PGID is generated as an identifier separated by dots; in the physical layer, the PGID is generated as an attribute ID separated by underscores; and in the data layer, the PGID is stored as a field value of the ODS source layer in its original encoding format.
[0039] Taking the No. 01 temperature sensor of the No. 1 oxygen lance in the No. 1 converter of G steel, with PGID GG-BOF-L001-TEMP-01, as an example, its name representation at each layer of the PGID chain is as follows: At the device layer (PLC / DCS layer), the PGID segments are directly concatenated according to character specifications to form the PLC variable name GG_BOF_L001_TMP01 (due to PLC naming length limitations, the parameter code TEMP is abbreviated to TMP, and the instance sequence number omits leading zeros). This name is written to the PLC's DB100.DBD40 logical address; At the gateway layer (OPC UA / MQTT layer), the PGID segments are separated by lowercase periods to generate the OPC UA Tag identifier GG.BOF.L001.TEMP01.Value. This identifier serves as the standard OPC exposed by the edge data acquisition gateway. UA variable nodes; in the physical layer (IoT platform / object model layer), each segment of PGID is connected by underscores and standardized to generate the object model attribute identifier GG_BOF_L001_TEMP_01. At the same time, metadata definitions (data type float, engineering unit degrees Celsius, range 0 to 1800 degrees Celsius, acquisition frequency 100Hz) are associated in this layer; in the data layer (data middle platform layer), the ODS source layer stores PGID in the original format GG-BOF-L001-TEMP-01 as the pgid field value, the DWD detail layer converts PGID into the wide table field name GG_BOF_L001_TEMP_01 DOUBLE, the DWS summary layer and the ADS application layer perform data retrieval and aggregation calculation through PGID, and the API query interface uses the parameter point=GG-BOF-L001-TEMP-01 as the query parameter. The PGID encoding value GG-BOF-L001-TEMP-01 remains unchanged across all layers; only the name representation is automatically derived according to the layer mapping rules, ensuring full-link traceability from PLC to data platform.
[0040] In one embodiment, the edge data standardization processing includes range conversion, dead zone compression, outlier marking, and unit unification processing; the edge quality pre-inspection includes access verification, CRC verification, real-time rule verification, and window statistics verification.
[0041] For example, based on metadata associated with PGID encoding (range, unit, signal type, etc.), the edge side automatically performs range conversion (converting 4-20mA / 0-10V electrical signals into engineering unit values), dead zone compression (only the beginning and end of continuous sampling points with value changes less than the dead zone threshold are retained, which can reduce the amount of transmitted data by 80%), outlier marking (instantaneous values that exceed the range or have abnormal rates of change are marked and then allowed to pass, such as a temperature that jumps by 200°C in one second and is marked as "suspected sensor failure"), and unit unification (all temperatures are unified to °C, pressures to kPa, and flow rates to Nm³ / h).
[0042] In one embodiment, the method further includes: synchronizing the second data group to the central platform through breakpoint resume, offline caching, and incremental reconciliation mechanisms; the incremental reconciliation mechanism includes: the edge gateway periodically reporting the currently online PGID list and the actual number of data entries collected and the average collection frequency of each PGID within the statistical period to the central platform; the central platform generates a difference alarm for PGIDs whose data entry deviation exceeds a preset threshold and automatically triggers a retransmission task.
[0043] Specifically, the breakpoint resume mechanism includes: the edge gateway monitors the network connection status with the central platform in real time; when a network interruption is detected, it automatically switches to local caching mode and records the data timestamp at the time of the interruption as a breakpoint marker; when the network is restored, the edge gateway starts from the timestamp position indicated by the breakpoint marker and sequentially uploads the locally cached data during the interruption, and sends a retransmission completion confirmation signal to the central platform after the upload is completed.
[0044] The offline caching mechanism includes: the edge gateway has a built-in local storage module, which writes the second data group, which has undergone edge data standardization and edge quality pre-inspection, into the local storage according to PGID and timestamp during network interruption; the local storage adopts a circular overwrite strategy, and the minimum cache capacity meets the full data storage requirements for more than 72 consecutive hours.
[0045] The incremental reconciliation mechanism includes: the edge gateway periodically reporting the list of currently online PGIDs and the average number of data entries and collection frequency of each PGID within the statistical period to the central platform; the central platform compares the received reported information with the whitelist of the point master data registration center and the actual amount of data entered into the ODS source layer, and generates a difference alarm for PGIDs whose data entry deviation exceeds a preset threshold and automatically triggers a retransmission task.
[0046] This method, through a globally unique PGID point encoding system and a five-layer mapping rule, ensures that the PGID of the same physical measurement point remains unchanged throughout the entire link from L0 sensor, L1 PLC, L2 acquisition gateway, L3 IoT platform, to L4 data platform. This enables second-level tracing and location of the source from abnormal business reports to physical sensor failures, solving the problem of infeasibility in tracing due to fragmented OT data naming in existing technologies. By employing three lines of edge quality defense (access verification, real-time rules, and window statistics), data quality issues are intercepted before they "enter the lake" (database). Illegal data is directly rejected, abnormal data is tagged and allowed, and abnormal trends trigger work orders, avoiding the high latency and high cost of handling problematic data after it flows into the central platform, as is common in existing technologies. Through a bidirectional mapping mechanism between the edge level four OT asset catalog and the central level five data asset catalog, equipment-side changes (sensor replacement / PLC expansion) are automatically synchronized to the business catalog, and business-side requirements (new indicators) are accurately located to the physical measurement point, solving the problems of fragmented catalogs and information lag in existing technologies. By employing a PGID whitelist comparison mechanism, "non-existent data" from unregistered locations can be proactively detected and blocked at the first line of defense for data access, filling the gap in existing technologies that lack code-level "non-existent data" detection. Through edge-side breakpoint resume transmission and offline caching mechanisms, it is ensured that even in extreme cases of network outages in the factory area (capacity ≥ 72 hours), OT data is not lost, and automatic retransmission and incremental reconciliation are performed after the network is restored.
[0047] In one specific embodiment, the digital transformation project of G Steel, a steel group company, is used as the application scenario. G Steel has five main production lines: sintering, ironmaking, steelmaking, continuous casting, and rolling. It has deployed over 20 heterogeneous systems, including ERP, MES, energy SCADA, and LIMS for testing and analysis. The OT (Operational Technology) site has tens of thousands of sensor points covering dozens of physical parameters such as temperature, pressure, flow rate, vibration, position, current, and voltage. For example... Figure 4As shown, the system adopts a five-layer deployment architecture: The L0 physical equipment layer includes sensors (thermocouples, pressure transmitters, flow meters, vibration sensors, etc.), actuators, PLC controllers (Siemens S7-1500 / S7-400H series), and DCS systems for each production line. The L1 edge data governance layer deploys 1-3 industrial-grade edge data acquisition gateways (configuration: embedded Linux operating system, ≥8-core CPU, ≥16GB RAM, ≥1TB SSD local cache, dual gigabit Ethernet ports) in each of G Steel's six main control rooms (sintering, ironmaking, steelmaking, continuous casting, rolling, and energy center), each with a built-in edge governance engine. The L2 central data governance platform layer is deployed in a Kubernetes containerized cluster (6-10 compute nodes) in G Steel's central computer room. The L3 lakehouse integrated storage base carries structured business data, high-frequency time-series data, and unstructured data. The L4 data application service layer provides a unified API interface for leadership dashboards, business report analysis, AI model training, and data sharing services.
[0048] In this embodiment, the PGID encoding adopts a five-segment hierarchical structure: Factory Code - Process Code - Equipment Code - Parameter Code - Instance Number. The Factory Code (2 letters), such as GG representing G steel; the Process Code (3 letters) uses internationally recognized process abbreviations, such as BOF representing converter smelting; the Equipment Code (4 alphanumeric characters) begins with the first letter indicating the equipment type (L for oxygen lance, B for furnace body, C for continuous casting stream), followed by the equipment number, such as L001 representing oxygen lance #1; the Parameter Code (4 letters), such as TEMP for temperature, PRES for pressure; and the Instance Number (2 digits) is numbered sequentially from top to bottom and left to right according to physical location, from 01 to 99. For example, GG-BOF-L001-TEMP-01 represents the 01st temperature sensor of oxygen lance #1 in the G steel converter smelting process.
[0049] The five-layer mapping of PGID encoding includes: the name of the same PGID encoding in each layer system is derived by mapping rules. L0 Sensor Layer: Type B thermocouple WRBT-5335, installed at the inlet of the oxygen lance cooling water pipe of converter #1; L1 PLC Layer: Physical address %IW256, logical address DB100.DBD40, PLC layer name deduced as GG_BOF_L01_TMP01; L2 Acquisition Gateway Layer: OPCUA Tag name deduced as GG.BOF.L001.TEMP01.Value; L3 IoT Platform Layer: Object model attribute identifier deduced as GG_BOF_L001_TEMP_01 (data type float, unit °C); L4 Data Platform Layer: ODS stores field values as pgid='GG-BOF-L001-TEMP-01', DWD uses GG_BOF_L001_TEMP_01 DOUBLE as the wide table field name, and API uses ?point=GG-BOF-L001-TEMP-01 as the query parameter. The PGID remains unchanged from L0 to L4, and the names of each layer are automatically derived from the PGID according to the mapping rules, ensuring consistency and traceability of naming across systems.
[0050] The five-layer mapping of PGID encoding (L0 sensor layer to L1 PLC layer to L2 acquisition gateway layer to L3 IoT platform layer to L4 data middle platform layer) describes the entire link flow process of data from physical sensor generation to final entry into the data middle platform, focusing on the vertical connection of data flow. Figure 4The five-layer deployment architecture (L0 physical device layer to L1 edge data governance layer to L2 central data governance platform layer to L3 lakehouse integrated storage base to L4 data application service layer) describes the physical / logical deployment location of each functional module of the system, focusing on the horizontal layering of the system architecture. The correspondence between the two is as follows: the L0 sensor layer and L1 PLC layer mapped by PGID are both physically deployed in the L0 physical device layer of the deployment architecture; the L2 acquisition gateway layer mapped by PGID corresponds to the L1 edge data governance layer of the deployment architecture, and the edge governance engine runs in this layer; the L3 IoT platform layer mapped by PGID corresponds to the L2 central data governance platform layer of the deployment architecture, and the object model management and PGID registration center of the central platform run in this layer; the L4 data middleware layer mapped by PGID corresponds to the combination of the L3 lakehouse integrated storage base and the L4 data application service layer of the deployment architecture, covering data storage and upper-layer data services at each layer of ODS / DWD / DWS / ADS. The four layers of the PGID encoding full-link – device layer, gateway layer, physical layer, and data layer – are a summary of the above five layers of PGID mapping: the device layer covers the L0 sensor layer and the L1 PLC layer (unified as the device side for data generation and initial naming), the gateway layer is the L2 acquisition gateway layer, the physical layer is the L3 IoT platform layer, and the data layer is the L4 data middleware layer.
[0051] The edge-side OT asset catalog of this embodiment includes: E-L1 production lines (sintering, ironmaking, steelmaking, continuous casting, rolling), E-L2 sections / areas (converter section, LF refining section, RH vacuum section), E-L3 equipment / devices (1# converter furnace body, 1# oxygen lance system, bottom blowing system), and E-L4 measuring points / variables (cooling water inlet temperature, etc.). The central-side data asset catalog includes L1 subject domain groups (production, sales, procurement), L2 subject domains (sintering, ironmaking, steelmaking themes), L3 business objects (materials, customers, equipment), L4 logical entities (converter process data table), and L5 attributes (oxygen lance inlet temperature field). A bidirectional mapping relationship is established; the L1 subject domain groups are purely central-side concepts, and no corresponding levels are set on the edge side.
[0052] This embodiment also sets up three lines of defense for edge quality, including: the first line of defense (access verification) compares the PGID whitelist at the edge gateway entry point, rejecting illegal data directly and recording a "data not found" alarm; the second line of defense (real-time rules) verifies the value range, mutation threshold, and frequency deviation in real time, allowing data exceeding the limits after labeling it and simultaneously pushing an alarm; the third line of defense (window statistics) calculates the integrity rate every 5 minutes, automatically triggering retransmission if it is below 99%. In the event of a network interruption, local caching is automatically started (≥72 hours), and after recovery, interrupted transmission and incremental reconciliation are performed to ensure zero data loss.
[0053] This embodiment also achieves a two-way closed loop between the edge governance engine and the central data governance platform through a five-dimensional collaborative mechanism, including incremental point registration, daily reconciliation, configuration distribution, detection of "non-existent data" points, and firmware / rule upgrades, for data reporting (edge → center) and configuration distribution (center → edge).
[0054] This method, through the deployment of an edge data governance system, achieves 100% PGID encoding coverage for OT points (newly accessed points); reduces end-to-end traceability time from several hours to ≤10 seconds (one-click traceability across the entire link based on PGID); achieves end-to-end latency of <100ms on the edge side (L0 sensor → edge gateway exit); eliminates the number of points with "non-existent data" (automatic detection during daily reconciliation); achieves an edge-to-center data integrity rate of ≥99.9% (point-level data volume reconciliation); and eliminates data loss due to edge network outages (local cache ≥72h + automatic retransmission).
[0055] On the other hand, this application also proposes a device for edge governance of industrial OT data, such as Figure 5 As shown, the industrial OT data edge governance device 100 includes: an acquisition module 200, used to acquire heterogeneous data from edge devices and perform format conversion on the heterogeneous data to obtain a first data group; a first processing module 300, used to construct an edge-side OT asset catalog and a central-side data asset catalog based on the edge devices, and to perform bidirectional binding mapping on the hierarchical levels of the edge-side OT asset catalog and the central-side data asset catalog; a second processing module 400, used to layer the edge-side OT asset catalog according to the physical dimension of the devices, including the production line layer, the area layer, the equipment layer, and the OT measurement point layer, and to layer the central-side data asset catalog according to the business dimension, including subject domain grouping, subject domain, business object, logical entity, and attribute; and a third processing module 500, used to assign PGID codes to OT measurement points in the OT measurement point layer, and to perform edge data standardization processing and edge quality pre-inspection on the first data group according to the PGID codes to obtain a second data group, which is used to synchronize to the central platform.
[0056] Specifically, the two-way binding mapping includes: the production line layer and the subject domain are bound together, the area layer and the equipment layer are jointly mapped to the business object, and the OT measurement point layer corresponds to the mapping logical entity and attribute.
[0057] Through the above technical solutions, this application establishes a globally unique PGID point encoding system and a five-layer mapping rule at the edge where data is generated, ensuring that the same physical measurement point maintains its PGID throughout the entire link. This enables second-level tracing and location of data from abnormal business reports to physical sensor failures, solving the problem of infeasibility of tracing due to fragmented OT data naming in existing technologies. By using three lines of defense for edge quality, it avoids the high latency and high cost of remediation after problematic data flows into the central platform, as is the case in existing technologies. Through a two-way mapping mechanism between the edge-side OT asset catalog and the central-side data asset catalog, changes on the device side are automatically synchronized to the business catalog, and business side requirements are accurately located to physical measurement points. This solves the problems of fragmented catalogs and information lag at both ends in existing technologies, achieving unified encoding, controllable quality, asset visibility, and security control for OT data throughout the entire link from physical sensors to the data platform.
[0058] The industrial OT data edge governance device 100 includes a processor and a memory. The aforementioned acquisition module 200, first processing module 300, second processing module 400, third processing module 500, etc., are all stored in the memory as program units. The processor executes the aforementioned program units stored in the memory to realize the corresponding functions.
[0059] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and unified security management of OT data can be achieved by adjusting kernel parameters.
[0060] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.
[0061] This invention provides a storage medium storing a program that, when executed by a processor, implements the method for edge governance of industrial OT data.
[0062] This invention provides a processor for running a program, wherein the program executes the method for edge governance of industrial OT data during runtime.
[0063] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: acquiring heterogeneous data from an edge device; converting the format of the heterogeneous data to obtain a first data group; constructing an edge-side OT asset catalog and a central-side data asset catalog based on the edge device; and performing a bidirectional binding mapping between the hierarchical levels of the edge-side OT asset catalog and the central-side data asset catalog; the edge-side OT asset catalog is layered according to the physical dimension of the device, including a production line layer, a region layer, a device layer, and an OT measurement point layer; the central-side data asset catalog is layered according to the business dimension, including subject domain groups, subject domains, business objects, logical entities, and attributes; assigning PGID codes to OT measurement points in the OT measurement point layer; and performing edge data standardization processing and edge quality pre-inspection on the first data group based on the PGID codes to obtain a second data group for synchronization to the central platform. The device in this document can be a server, PC, PAD, mobile phone, etc.
[0064] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: acquiring heterogeneous data from edge devices, converting the format of the heterogeneous data to obtain a first data group; constructing an edge-side OT asset catalog and a central-side data asset catalog based on the edge devices, and performing bidirectional binding mapping on the hierarchy of the edge-side OT asset catalog and the central-side data asset catalog; the edge-side OT asset catalog is layered according to the physical dimension of the devices, including a production line layer, a region layer, a device layer, and an OT measurement point layer; the central-side data asset catalog is layered according to the business dimension, including subject domain groups, subject domains, business objects, logical entities, and attributes; assigning PGID codes to OT measurement points in the OT measurement point layer, and performing edge data standardization processing and edge quality pre-inspection on the first data group based on the PGID codes to obtain a second data group for synchronization to the central platform.
[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] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0068] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0069] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0070] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.
[0071] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.
[0072] 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.
[0073] 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. A method for edge governance of industrial OT data, characterized in that, The method includes: Obtain heterogeneous data from edge devices, and perform format conversion on the heterogeneous data to obtain a first data group; An edge-side OT asset catalog and a central-side data asset catalog are constructed based on the edge devices, and a two-way binding mapping is performed on the hierarchy of the edge-side OT asset catalog and the central-side data asset catalog; The edge-side OT asset catalog is layered according to the physical dimension of equipment, including the production line layer, area layer, equipment layer, and OT measurement point layer. The central-side data asset catalog is layered according to the business dimension, including subject domain grouping, subject domain, business object, logical entity, and attribute. PGID codes are assigned to OT measurement points in the OT measurement point layer. Based on the PGID codes, the first data group is subjected to edge data standardization processing and edge quality pre-inspection to obtain a second data group, which is used to synchronize to the central platform.
2. The method according to claim 1, characterized in that, The bidirectional binding mapping includes: bidirectional binding between the production line layer and the subject domain; mapping of the region layer and the equipment layer to business objects; and mapping of the OT measurement point layer to logical entities and attributes.
3. The method according to claim 1, characterized in that, The method also includes: When devices are added, deleted, or modified in the edge-side OT asset catalog, the central-side data asset catalog is updated synchronously based on preset bidirectional mapping rules. When the central side adds new business indicators based on business needs, it uses the two-way binding mapping to retrieve and locate the corresponding physical measurement point variables on the edge side.
4. The method according to claim 1, characterized in that, The PGID encoding includes factory code, process code, equipment code, parameter code, and instance number; The factory code is used to identify the physical factory, the process code is used to identify the steel production process, the equipment code is used to identify the specific equipment within the process, the parameter code is used to identify the type of physical parameter, and the instance number is used to distinguish multiple sensor instances of the same equipment and the same parameter.
5. The method according to claim 1, characterized in that, The entire PGID encoding chain includes the device layer, gateway layer, physical layer, and data layer; In the device layer, the PGID is generated as a name according to the character specification; in the gateway layer, the PGID is generated as an identifier separated by dots; in the physical layer, the PGID is generated as an attribute ID separated by underscores; and in the data layer, the PGID is stored as a field value of the ODS source layer in its original encoding format.
6. The method according to claim 1, characterized in that, The edge data standardization process includes range conversion, dead zone compression, outlier marking, and unit unification. The edge quality pre-inspection includes access verification, CRC verification, real-time rule verification, and window statistics verification.
7. The method according to claim 1, characterized in that, The method also includes: The second data group is synchronized to the central platform through breakpoint resume, offline caching and incremental reconciliation mechanisms; The incremental reconciliation mechanism includes: the edge gateway periodically reporting the list of currently online PGIDs and the average number of data entries and collection frequency of each PGID within the statistical period to the central platform; the central platform generates a difference alarm for PGIDs whose data entry deviation exceeds a preset threshold and automatically triggers a retransmission task.
8. A device for edge governance of industrial OT data, characterized in that, The device includes: An acquisition module is used to acquire heterogeneous data from edge devices and perform format conversion on the heterogeneous data to obtain a first data group. The first processing module is used to construct an edge-side OT asset catalog and a central-side data asset catalog based on the edge device, and to perform bidirectional binding mapping on the hierarchy of the edge-side OT asset catalog and the central-side data asset catalog; The second processing module is used to layer the edge-side OT asset catalog according to the physical dimension of the equipment, including the production line layer, the area layer, the equipment layer, and the OT measurement point layer; and to layer the central-side data asset catalog according to the business dimension, including subject domain grouping, subject domain, business object, logical entity, and attribute. The third processing module is used to assign PGID codes to OT measurement points in the OT measurement point layer, and to perform edge data standardization processing and edge quality pre-inspection on the first data group according to the PGID codes to obtain a second data group, which is used to synchronize to the central platform.
9. The apparatus according to claim 8, characterized in that, The bidirectional binding mapping includes: bidirectional binding between the production line layer and the subject domain; mapping of the region layer and the equipment layer to business objects; and mapping of the OT measurement point layer to logical entities and attributes.
10. A machine-readable storage medium storing instructions thereon, characterized in that, This instruction is used to cause the machine to perform any one of the industrial OT data edge governance methods of claims 1-7 of this application.
