Production process-oriented data semantic model construction method and data storage method thereof

Abstract

The invention relates to a production process-oriented data semantic model construction method and a data storage method thereof, and relates to the field of data modeling. Under the condition that enterprise production needs multi-system collaborative data calling, data of each system is collected, the collected data comprises metadata and a data ontology, mapping processing is conducted on the collected metadata information through a data dictionary, and semantic description is unified; meanwhile, endowing the data ontology with a unique identifier, and mapping the data into a data semantic model according to a data dictionary; the semantic model is divided into a data description layer, a data state description layer and a data context description layer from top to bottom; uniform data semantic modeling is performed on the operation data of the industrial internet, so that massive heterogeneous operation data can be described and managed in a unified manner, and the problem that data between different systems or between the system and equipment is not intercommunicated is solved.

Description

Technical Field

[0001] The present invention belongs to the field of semantic models for production process data, specifically for uniformly describing the massive heterogeneous data information stored in the production process and storing it according to a unified format, description, and model. Background Art

[0002] With the continuous improvement of the digital and networked level of the production and manufacturing processes of industrial enterprises in China, the demand for the exchange of a large amount of production / operation data, control data covering the factory production process, and management data, design / process data in the traditional network is increasing day by day. The interconnected entities will further expand from the machines, control systems, and information systems supported by the traditional Internet and the Internet of Things to all elements of the entire manufacturing process data including products, personnel, work-in-progress, materials, equipment tooling, tooling dies, documents, logistics units, locations, units, departments, and suppliers. Therefore, to achieve data sharing across regions, industries, and enterprises, effectively promote the digital development of China's manufacturing industry, improve the production efficiency of enterprises and their competitiveness in the international arena, and to promote data sharing and flow across regions, industries, and enterprises, and guide industrial enterprises to perform efficient data semantic processing. By introducing semantic analysis and management technologies widely recognized in the field of cross-database knowledge fusion on the Internet, realizing the interconnection, understandability, and sharing of massive heterogeneous industrial Internet data information will be an important direction and support for enterprise development. In the data information facing the production process, introducing semantic management technology and information model technology will promote the integration of heterogeneous multi-source systems and the development of comprehensive applications, and support the improvement of the industrial platform. Summary of the Invention

[0003] The object of the present invention is to provide a data semantic model for the production process, which can solve the problems of unified description and unified format of the massive and heterogeneous data in the enterprise production process.

[0004] The technical solution adopted by the present invention to achieve the above object is: A data semantic model for the production process, used to realize the mapping and storage of data by constructing different description layers, including:

[0005] A data description layer, used to describe the meaning and type of data;

[0006] A status description layer, used to describe the existing working status of the data;

[0007] A context description layer, used to store the data information related to the data.

[0008] The data description layer includes data meaning, data type, data format, and timestamp.

[0009] The data meaning includes data identification, data name, data description, and data source;

[0010] The data identifier: A data unique identifier used to identify data;

[0011] The data name: This field is the definition of the data, describing the meaning represented by the data;

[0012] The data description: This field is a supplementary description of the data name, including but not limited to text information and external links;

[0013] The data source: The data source is the origin of the extracted data. The data on the end side of the data workshop mainly comes from on-site equipment information, and the data on the edge side mainly comes from the system information of the management layer, including but not limited to ERP, PLM, and MES.

[0014] The data types include static data and dynamic data;

[0015] The static data: Data used to describe the inherent attributes of an object that distinguish it from other entities, including but not limited to:

[0016] Subject data, used to describe enterprises, institutions, and natural persons;

[0017] Location data, used to describe physical or logical locations;

[0018] Object data, used to describe physical objects.

[0019] The dynamic data, used to describe the data generated by the changes in location, status, ownership, and management rights recorded during the production, circulation, and use of industrial products, including but not limited to:

[0020] Production data, used to describe the process of industrial products from raw materials through processing into products;

[0021] Circulation data, used to describe the warehousing, transportation, and sales information generated by industrial products in the supply chain;

[0022] Usage data, used to describe the data generated during the use of industrial products.

[0023] The data format includes the data storage format and the data standard unit; The data storage format: The data storage format includes but not limited to integer type, string type, and boolean type; The data unit is used to record the physical unit of the data.

[0024] The status description layer includes the working status and access permissions.

[0025] The context description layer includes associated data and association types;

[0026] The associated data is used to describe data information that has an associated relationship with the target data; the associated data is linked to the target data through a certain association type, and the associated data and the target data are connected through industrial Internet identifiers.

[0027] A method for constructing a data semantic model for the production process, which realizes data mapping and storage by constructing different description layers, including the following steps:

[0028] Construct a data description layer to describe the meaning and type of data;

[0029] Construct a status description layer to describe the existing working status of the data;

[0030] Construct a context description layer to store data information related to the data.

[0031] A data storage method based on the data semantic model for the production process, including the following steps:

[0032] Read the data information of the production process, uniformly map the data information through a data dictionary to form a unified data description; obtain the status information and associated relationship information of the data; uniformly store the data description, status information, and associated relationship information into the data semantic model.

[0033] The step of uniformly mapping the data information through a data dictionary to form a unified data description is specifically as follows:

[0034] Map the read data information into the data semantic model through a data dictionary, match the data names in the data with the data names in the data dictionary, map the data name with the highest matching degree to the data name field of the data semantic model, and the remaining data semantic model fields are filled with data.

[0035] The present invention has the following beneficial effects and advantages:

[0036] 1. The present invention can improve the production operation efficiency of enterprises. Based on the data semantic model for the production process of the present invention, it can uniformly describe, format, and store a large amount of heterogeneous data information, enabling data with similar or close expressions to be uniformly named, solving the problems of data and system interoperability, and making the system hierarchy more flattened. Greatly improving the degree of system intelligence and automation

[0037] 2. Greatly reduce the cost of secondary development. Based on this information model for the production process, seamless transfer of data information among different information systems and hardware devices is realized. Workers do not need secondary development programming and device adaptation, greatly reducing the cost of secondary development. Brief Description of the Drawings

[0038] Figure 1Data semantic flow chart of the present invention for the production process. Detailed implementation manners

[0039] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0040] In the present invention, in view of the difficulties of massive, heterogeneous, and data information asymmetry of enterprise production data during the development of the industrial Internet, a method for constructing a data semantic model is proposed; in the case where multiple systems need to be coordinated to call data in enterprise production, data of each system is collected, and the collected data includes metadata and data ontology. The collected metadata information is mapped through a data dictionary to unify semantic descriptions; at the same time, a unique identifier is assigned to the data ontology, and the data is mapped into the data semantic model according to the data dictionary; the semantic model is divided into a data description layer, a data status description layer, and a data context description layer from top to bottom; by performing unified data semantic modeling on the operation data of the industrial Internet, massive heterogeneous operation data can be uniformly described and managed, and problems such as data non-communication between different systems or between a system and a device can be solved.

[0041] The information model for the production process is divided into three layers in total, including a data description layer, a data status description layer, and a data context description layer. The description content of each specific layer is shown in Table 1.

[0042] Table 1

[0043]

[0044] As shown in Table 1, the data description layer includes data meaning, data type, and data format. The content stored in this layer is the unified description of the data; among them

[0045] The data meaning includes data identifier, data name, data description, and data source; the data identifier is the identity identifier of the data, which is unique, and specific data information can be determined through the identifier; the data name is the definition of the data, describing the meaning of the data; the data description is the supplementary definition of the data name and the specific explanation of the data name; the data source is the origin of the extracted data, including from an information system or a device.

[0046] The data meaning, characterized in that: it includes a data identifier, a data name, a data description, and a data source;

[0047] Data identifier: The unique identifier of the data, used to identify the data;

[0048] Data name: This field is the definition of the data, describing the meaning represented by the data.

[0049] Data description: This field is a supplementary description of the data name, including but not limited to text information and external links.

[0050] Data source: The data source is the origin of the extracted data. The data on the mid-end side of the data workshop mainly comes from on-site equipment information, and the data on the edge side mainly comes from the system information of the management layer, including but not limited to ERP, PLM, and MES.

[0051] Data type describes the characteristics to which the data belongs. According to whether the data circulates in the production process, the data is divided into static data and dynamic data. Static data is the data that describes the inherent attributes and characteristics that distinguish it from other entities; dynamic data describes the data generated by the changes in position, status, ownership, management rights, etc. recorded during the production, circulation, and use of products.

[0052] The data types include static data and dynamic data;

[0053] Static data: The data that describes the inherent attributes and characteristics that distinguish the object from other entities, including but not limited to:

[0054] 1) Subject data: Used to describe enterprises, institutions, and natural persons;

[0055] 2) Location data: Used to describe physical or logical locations, such as addresses and property owners;

[0056] 3) Object data: Used to describe physical objects, such as products and equipment.

[0057] Dynamic data: The data used to describe the data generated by the changes in position, status, ownership, management rights, etc. recorded during the production, circulation, and use of industrial products, including but not limited to:

[0058] 1) Production data: Used to describe the process of industrial products from raw materials through processing into products;

[0059] 2) Circulation data: Used to describe the warehousing, transportation, and sales information generated by industrial products in the supply chain;

[0060] 3) Usage data: Used to describe the data generated during the use of industrial products.

[0061] Data format includes the data storage format type and the storage data unit. The data storage format includes integer, string, boolean, etc., and the data unit records the physical unit of the data.

[0062] The timestamp is used to record the storage time of the data, and its specific storage format is "YYYY-MM-DD HH:MM:SS".

[0063] The data status description layer includes the working status and access rights; the data working status records the working status of the devices, sensors, or systems from which the data originates, and the specific status includes but is not limited to the running status, maintenance status, standby status, offline status, etc. The data access rights record the rights of users to access or modify the data, and the operations on the data are divided into create, delete, update, and query.

[0064] The working status data mainly describes the working status of the devices, sensors, or systems from which the data originates, and the specific status includes the running status, maintenance status, standby status, offline status.

[0065] The data context description layer includes associated data information and association types. Each associated data information contains an association type, and the data can contain one or more associated data;

[0066] The associated data mainly describes the data information that has an association relationship with the target data. The associated data is connected to the target data through a certain association type. The associated data also has the same semantic description framework, including data description, status description, and context description. The associated data and the target data are connected through industrial Internet identifiers.

[0067] The association type describes the association relationship between the associated data and the target data. The association type connects the data by defining the association relationship. The association types include progressive relationship and reciprocal relationship.

[0068] The data semanticization process is to uniformly describe massive heterogeneous data through specific methods, which includes three parts: data semantic model, data dictionary mapping, and semantic library storage.

[0069] The data semanticization process, its steps include: first, read the data information in the production process, uniformly map the data information through the data dictionary to form a unified data description, and obtain the status information and association relationship information of the data, and uniformly store the information into the data semantic model. Finally, store the semantic information model into the semantic library.

[0070] Such as Figure 1 As shown in the data semanticization flowchart for the production process, first collect enterprise data. The data information for the production process includes system data and on-site data. The read data information is massive and heterogeneous and cannot be directly used, and semantic processing is required. Secondly, map the read data information into the semantic model through the data dictionary, match the data names in the data with the data names in the data dictionary, and map the most suitable data name to the data name field of the information model. The remaining information model fields are filled with data and can be empty. Finally, store the semantic information model into the semantic library.

Claims

1. A data semantic model for the production process, characterized in that For realizing the mapping and storage of data by constructing different description layers, including: The data description layer is used to describe the meaning and type of data; The status description layer is used to describe the existing working status of data; The context description layer is used to store data information related to the data.

2. The data semantic model for the production process according to claim 1, wherein The data description layer includes data meaning, data type, data format, and timestamp.

3. The data semantic model for the production process according to claim 2, wherein The data meaning includes data identifier, data name, data description, and data source; The data identifier: a unique identifier for the data, used to identify the data; The data name: this field is the definition of the data, describing the meaning represented by the data; The data description: this field is a supplementary description of the data name, including but not limited to text information, external links; The data source: the data source is the origin of the extracted data. The data on the end side in the data workshop mainly comes from on-site equipment information, and the data on the edge side mainly comes from the system information of the management layer, including but not limited to ERP, PLM, and MES.

4. The data semantic model for the production process according to claim 2, characterized in that, The data type includes static data and dynamic data; The static data: data used to describe the inherent attributes of an object that distinguish it from other entities, including but not limited to: Subject data, used to describe enterprises, institutions, and natural persons; Location data, used to describe physical or logical locations; Object data, used to describe physical objects. The dynamic data is used to describe the data generated by the changes in location, status, ownership, and management rights recorded during the production, circulation, and use of industrial products, including but not limited to: Production data, used to describe the process of industrial products from raw materials through processing into products; Circulation data, used to describe the warehousing, transportation, and sales information generated by industrial products in the supply chain; Usage data, used to describe the data generated during the use of industrial products.

5. The data semantic model for the production process according to claim 2, wherein The data format includes data storage format and data standard unit; the data storage format: the storage format of the data includes but not limited to integer type, string type, boolean type; the data unit is used to record the physical unit of the data.

6. The data semantic model for the production process according to claim 1, wherein The status description layer includes working status and access rights.

7. The data semantic model for the production process according to claim 1, wherein The context description layer includes associated data and association type; The associated data is used to describe the data information related to the target data; the associated data is connected to the target data through a certain association type, and the associated data and the target data are connected through industrial Internet identifiers.

8. A method for constructing a data semantic model for the production process, characterized in that, By constructing different description layers, the mapping and storage of data are realized, including the following steps: Construct a data description layer to describe the meaning and type of data; Construct a status description layer to describe the existing working status of data; Construct a context description layer to store data information related to the data.

9. A data storage method based on a data semantic model for a production process, characterized in that, Including the following steps: Read the data information in the production process, uniformly map the data information through a data dictionary to form a unified data description; obtain the status information and associated relationship information of the data; uniformly store the data description, status information, and associated relationship information into a data semantic model.

10. The data storage method based on the data semantic model for the production process according to claim 9, wherein The uniformly mapping the data information through a data dictionary to form a unified data description is specifically: Map the read data information to the data semantic model through the data dictionary. Match the data names in the data with the data names in the data dictionary, and map the data name with the highest matching degree to the data name field of the data semantic model. The remaining data semantic model fields are filled with data.

Images