An ontology-based digital twin workshop multi-dimensional information fusion method

Abstract

The application discloses a kind of based on ontology's digital twin workshop multi-dimensional information fusion method, comprising the following steps: S1, according to the attribute characteristics of workshop physical entity resource, physical resource is classified, so that the modeling and management of different types of resources related activities standardization is standardized.S2, the same type resource attribute is divided according to the angle of description information, constructs workshop digital twin global ontology model based on ontology modeling language, forms workshop digital twin global ontology model library.S3, according to the application of twin workshop cross-domain multi-scene, constructs the local ontology model of physical entity under each scene, and completes each local ontology and the mapping of local ontology to global ontology.S4, according to the running condition of workshop under each scene, the characteristics of ontology model attribute value acquisition mode are combined, and the collection of local ontology model information under different scenes is completed.S5, according to the mapping relationship between global ontology and each local ontology, realize multi-scene digital twin multi-dimensional information fusion.

Description

Technical Field

[0001] This invention belongs to the field of digital technology, specifically relating to an ontology-based method for multidimensional information fusion in a digital twin workshop. Background Technology

[0002] With the development of new-generation computer and communication technologies, the theoretical research and application of digital twins have attracted widespread attention from scholars and enterprises worldwide. The development of digital twin technology is seen as an important tool for the practical application of intelligent manufacturing. It uses information technology to construct virtual models corresponding to physical entities and uses data to drive the virtual models, possessing the characteristics of "synchronization between virtual and real," "reflecting the real with the virtual," and "optimizing the real with the virtual."

[0003] The application of digital twins in the workshop not only covers multiple levels such as workshop, unit, equipment, and parts, involving different scales of macro and micro, but also spans multiple fields such as design, manufacturing, production, and inspection. As a virtual mapping of physical entities, an ideal digital twin model involves the fusion of multi-dimensional information such as geometry, physics, behavior, and rules, and has the characteristics of high fidelity, high reliability, and high precision, thereby realistically depicting the characteristics of physical entities. Currently, scholars at home and abroad have conducted relevant research on the construction methods of information models of physical entities, and have given model construction methods for different stages and scenarios in design, manufacturing, and inspection. However, existing research on the construction of information models in twin space is only for specific application scenarios in a single domain, and the multi-dimensional information of physical entities at different stages still mostly needs to be manually synthesized. Moreover, because the information expression methods for the same physical entity are different in different application scenarios, it is difficult to fuse multi-dimensional information of physical entities in multi-scenario applications across domains and scales, making it impossible to achieve information sharing in different application scenarios throughout the entire life cycle, and unable to accurately construct high-fidelity models of physical entities. Therefore, in order to more realistically depict the characteristics of physical entities in the workshop, a detailed, specific, and operable multi-dimensional information fusion method is needed to guide the construction of twin space information models. Summary of the Invention

[0004] The purpose of this invention is to solve the above-mentioned problems and provide an ontology-based digital twin workshop multi-dimensional information fusion method that can solve the problems of poor correlation and difficulty in information fusion of multi-dimensional information of digital twins in cross-domain and multi-scenario applications in workshops.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is: a multi-dimensional information fusion method for an ontology-based digital twin workshop, comprising the following steps:

[0006] S1. Based on the attribute characteristics of physical resources in the workshop, classify physical resources to standardize the modeling and management of different types of resources and other related activities.

[0007] S2. Divide the attributes of the same type of resources according to the perspective of description information, and build a global ontology model of the workshop digital twin based on ontology modeling language to form a global ontology model library of workshop digital twin;

[0008] S3. Based on the cross-domain and multi-scenario application of the twin workshop, construct local ontology models of physical entities in each scenario, and complete the mapping of each local ontology and the local ontology to the global ontology.

[0009] S4. Based on the workshop operation status in each scenario and the characteristics of the ontology model attribute value acquisition method, complete the collection of local ontology model information in different scenarios.

[0010] S5. Based on the mapping relationship between the global ontology and each local ontology, multi-dimensional information fusion of digital twins in multiple scenarios is achieved according to the ontology information fusion method.

[0011] Furthermore, in step S1, based on the characteristics of the production elements in the actual physical workshop, the workshop resources are divided into four categories: manufacturing equipment resources, material resources, auxiliary hardware resources, and human resources. Each resource category can be further subdivided according to actual applications, so that the attribute characteristics of different types of resources are similar, forming a workshop resource tree, which facilitates unified management.

[0012] Furthermore, step S2 also includes the following sub-steps:

[0013] S21. Based on step S1, different types of resource attributes are divided according to different perspectives of resource attribute description information. While ensuring that the expressed information is not repeated, the characteristics of workshop resources are represented as comprehensively as possible.

[0014] S22. Model resource instances using ontology modeling languages ​​to build a global ontology resource library; construct resource models based on ontology languages ​​such as XML and RDF so that resource descriptions can be understood by computers, facilitating information interaction and integration in virtual space.

[0015] Furthermore, step S3 also includes the following sub-steps:

[0016] S31. In each application scenario, based on the principle of grasping the main information and ignoring the secondary information, a local ontology model of the physical resources of the workshop required in each scenario is constructed. During the entire product life cycle, the workshop involves cross-domain and multi-scenario applications such as scheduling, design and production. The information of workshop resources that are of concern in different scenarios is different. In order to improve the application efficiency of the resource ontology model in each scenario, the information of the physical entities of concern in each scenario is extracted, and a local ontology model of resources suitable for this scenario is constructed.

[0017] S32. Based on the local ontology model information in each scenario, iteratively update the comprehensive information of workshop resources maintained in the global ontology. Based on the identity attributes in each ontology model, realize the association between local ontology and between local ontology and global ontology, so that ontology describing the same resource in cross-domain and multi-scenario scenarios can interoperate. On this basis, perform association mapping between attributes describing the same resource and the same information in the ontology to ensure semantic consistency.

[0018] Furthermore, step S4 specifically involves classifying attributes into static attributes and dynamic attributes in the local ontology model corresponding to different scenarios, based on the different methods of acquiring attribute values. Different attribute values ​​are acquired in different ways. Static attributes refer to attributes that describe the inherent information of workshop resources. They are acquired and filled manually or read into the computer during workshop activities, and their attribute values ​​do not change over time or have a long change cycle. Dynamic attributes refer to attributes whose attribute values ​​change continuously during workshop activities. They are mainly collected dynamically through processes such as data acquisition, analysis, and calculation, and are updated in real time as the physical resource status values ​​change.

[0019] Furthermore, step S5 specifically involves fusing attribute information based on the association relationship between attribute values ​​of the ontology describing the same physical entity, and on the principle of merging complementary information and removing redundant information. In multi-scenario applications, some information appears in multiple local ontology instances. These information only need to be recorded once during merging, and the rest are deleted as redundant information. Other information is scattered in local ontology instances under different application scenarios. These information are integrated into the global ontology as complementary information. In this way, the fusion of multi-dimensional information describing physical entities under cross-domain multi-scenario applications is realized, so as to display the information of all nodes of the physical entity in the digital twin space.

[0020] The beneficial effects of this invention are as follows: Compared with existing digital twin information model construction methods, the ontology-based multidimensional information fusion method for digital twin workshops provided by this invention offers an interactive fusion method for physical entity information models in cross-domain, multi-scenario applications. The ontology-based information model, while serving various application scenarios, also facilitates the sharing and interoperability of model information within the computer, showcasing the multidimensional information of physical entities in the twin space. Furthermore, it facilitates the establishment of connections between different resources in the twin space, laying the foundation for subsequent semantic reasoning and semantic querying, and interoperability between different digital twins. Attached Figure Description

[0021] Figure 1 This is a flowchart of the cross-domain, multi-scale workshop digital twin multidimensional information fusion method of the present invention;

[0022] Figure 2 This is a schematic diagram of the entity resource division in the workshop example of this invention;

[0023] Figure 3 This is a schematic diagram of the XML attributes of the workshop machine tool entity of the present invention;

[0024] Figure 4 This is a schematic diagram of the partial body attributes of a machine tool in a workshop production scheduling scenario according to the present invention;

[0025] Figure 5 This is a schematic diagram of the partial body attributes of a machine tool in the workshop process design scenario of this invention;

[0026] Figure 6 This is a schematic diagram of the partial body attributes of a machine tool in a workshop product processing scene according to the present invention;

[0027] Figure 7 This is a schematic diagram of information collection in different scenarios in the workshop of this invention;

[0028] Figure 8 This is a flowchart of the ontology information fusion of the present invention;

[0029] Figure 9 This is a schematic diagram of multi-dimensional information of machine tools in the workshop according to the present invention. Detailed Implementation

[0030] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0031] like Figure 1 As shown, the present invention provides a multi-dimensional information fusion method for an ontology-based digital twin workshop, comprising the following steps:

[0032] S1. Based on the attribute characteristics of physical resources in the workshop, classify physical resources to standardize the modeling and management of different types of resources and other related activities.

[0033] Digital twin workshop resources are divided into two main categories: hard manufacturing resources and soft manufacturing resources, which can be further divided into six subcategories. The categories can be appropriately increased or decreased according to actual usage needs to meet the requirements of different usage environments and increase practicality.

[0034] In actual use, step S1 divides workshop resources into four categories based on the characteristics of production elements in the actual physical workshop: manufacturing equipment resources, material resources, auxiliary hardware resources, and human resources. Each resource category can be further subdivided according to actual application to make the attribute characteristics of different types of resources similar, forming a workshop resource tree for convenient unified management.

[0035] like Figure 2 As shown, in this embodiment, workshop resources are divided into four main categories, and further subdivided according to their characteristics. The descriptions of each resource category are as follows:

[0036] (1) Manufacturing equipment resources: Various equipment involved in workshop activities, which can be further subdivided into machine tool processing equipment, handling and assembly equipment, transportation equipment, machining equipment, measuring equipment, etc.

[0037] (2) Material resources: refers to the materials involved in workshop activities, including raw materials, blanks and semi-finished products used in production.

[0038] (3) Auxiliary hardware resources: refers to auxiliary components used for processing, transportation equipment and other resources, including cutting tools, grippers of robotic arms, etc.

[0039] (4) Human Resources: refers to the people involved in workshop activities, who are responsible for various activities such as operation and maintenance of production equipment, auxiliary production software and products during the production process, including R&D personnel, management personnel, maintenance personnel, etc.

[0040] S2. Divide the attributes of the same type of resources according to the perspective of description information, and construct a global ontology model of the workshop digital twin based on ontology modeling language to form a global ontology model library of workshop digital twin.

[0041] Step S2 also includes the following sub-steps:

[0042] S21. Based on step S1, different types of resource attributes are divided according to different perspectives of resource attribute description information. While ensuring that the expressed information is not repeated, the characteristics of workshop resources are represented as comprehensively as possible.

[0043] Ontology-based resource representation requires three elements: class, attribute, and attribute value. Relevant class information is already included in the resource partitioning process. Physical entity resource attributes can be hierarchically partitioned based on descriptive information, forming a manufacturing resource ontology attribute tree. Workshop resources are diverse, and different types of resources have different descriptive attributes, but attributes describing resource identity information must exist in every entity. Taking machine tool processing equipment in manufacturing equipment resources as an example, its first-level ontology attribute partitioning can be represented by the following four-tuple:

[0044] Attributes = {B, T, S, R}

[0045] Here, B represents Basic Attributes, primarily used to identify the physical entity's identity information, including its number, name, and workshop affiliation. T represents Technical Attributes, mainly used to describe the physical entity's physical structure and technical performance, including geometric dimensions, maximum load, maximum power, and speed range. S represents State Attributes, primarily related to the equipment's operating status, such as machine tool axis position information, axis current information, spindle speed, and alarm information. R represents Relationship Attributes, mainly used to describe the relationship between the machine tool entity and other entities. For example, a cutting tool is an auxiliary hardware resource, an indispensable part of the machine tool machining process, and its relationship with the machine tool is that of a part to the whole. Therefore, cutting tool information should be placed in the machine tool relationship attributes, such as tool number, cutting force, and tool wear value.

[0046] S22. Model resource instances using ontology modeling languages ​​to build a global ontology resource library; construct resource models based on ontology languages ​​such as XML and RDF so that resource descriptions can be understood by computers, facilitating information interaction and integration in virtual space.

[0047] The description of an ontology model needs to be transformed into an ontology language. In ontology research, there are many ontology description languages ​​and construction tools. Choosing a suitable semantic ontology description language not only improves the readability of the ontology model described in natural language, but also has a better effect on subsequent ontology knowledge parsing and multi-domain applications. Currently, commonly used ontology languages ​​in the field of ontology research include XML, RDF, and OWL. Among them, XML, as the foundation of ontology languages, has a wide range of applications, good compatibility, and strong operability. In this embodiment, the ontology model is constructed based on the XML language. Taking the above machine tool attribute analysis as an example, the construction of some attributes is as follows: Figure 3 As shown.

[0048] S3. Based on the cross-domain and multi-scenario application of the twin workshop, construct local ontology models of physical entities in each scenario, and complete the mapping of each local ontology and the local ontology to the global ontology.

[0049] Step S3 also includes the following sub-steps:

[0050] S31. In each application scenario, based on the principle of grasping the main information and ignoring the secondary information, a local ontology model of the physical resources of the workshop required in each scenario is constructed. During the entire product life cycle, the workshop involves cross-domain and multi-scenario applications such as scheduling, design and production. The information of workshop resources that are of concern in different scenarios is different. In order to improve the application efficiency of the resource ontology model in each scenario, the information of the physical entities of concern in each scenario is extracted, and a local ontology model of resources suitable for this scenario is constructed.

[0051] Throughout the entire product lifecycle, digital twin workshops involve various activities such as scheduling, design, and production. The physical entities within the workshop can exist in multiple application scenarios, each with different attributes relevant to its physical entity. A global ontology displays comprehensive information about the physical entity, requiring continuous refinement based on the functional requirements of different application scenarios. However, applying the global ontology to all scenarios would result in redundant information for some applications, potentially leading to an overly large ontology model and reduced application efficiency. Therefore, it is necessary to construct local ontology models of the physical entities specific to each scenario's requirements.

[0052] This embodiment illustrates the application of machine tools in three scenarios: production scheduling, process design, and product processing.

[0053] Please see Figure 4 In production scheduling scenarios, in order to complete production tasks, it is necessary to schedule and select machine tools in the workshop to perform the tasks. At this time, it is necessary to locate the machine tools based on the basic attributes of the corresponding entity, such as ID, workshop, and unit. At the same time, it is also necessary to pay attention to technical attributes such as processing time and status attributes such as fault information to provide a basis for scheduling algorithms.

[0054] Please see Figure 5 In process design scenarios, the formulation and execution of process plans are related to the functions of machine tool processing. Therefore, in this scenario, in addition to basic attributes, technical attributes such as maximum power, speed range, and machining accuracy also need to be considered. When writing process documents, relational attributes such as tool information also need to be taken into account.

[0055] Please see Figure 6 In product processing scenarios, the primary focus should be on machine tool status information during the machining process. Therefore, in addition to basic attributes, attention should be paid to status information such as current program information, current information, axis position information, spindle speed information, feed rate information, and alarm information. Furthermore, the tool number and its corresponding cutting force and tool wear value should also be considered.

[0056] S32. Based on the local ontology model information in each scenario, iteratively update the comprehensive information of workshop resources maintained in the global ontology. Based on the identity attributes in each ontology model, realize the association between local ontology and between local ontology and global ontology, so that ontology describing the same resource in cross-domain and multi-scenario scenarios can interoperate. On this basis, perform association mapping between attributes describing the same resource and the same information in the ontology to ensure semantic consistency.

[0057] Each ontology model is associated with the ID values ​​of the physical entities described in their basic attributes. The global ontology needs to be iteratively updated while designing local ontology models. For example, in production scheduling, process design, and product processing scenarios, the corresponding local ontology for each scenario is found through the machine tool ID value. Since the attribute names of different local ontology models describing the same information about physical entities may differ, to ensure semantic consistency, attribute mappings are established between attributes in each ontology. For instance, in the production scheduling scenario, the fault information attribute and the alarm information attribute in product processing should describe the same information about physical entities, and the attributes should be equivalent. Simultaneously, to ensure that the global ontology displays comprehensive information about physical entities, iterative updates are needed in scenario applications, such as updating existing attribute names in the global ontology or expanding attributes not displayed in a particular scenario application.

[0058] S4. Based on the workshop operation status in each scenario and the characteristics of the ontology model attribute value acquisition method, complete the collection of local ontology model information in different scenarios.

[0059] Step S4 specifically involves classifying attributes into static and dynamic attributes in the local ontology model corresponding to different scenarios, based on the different methods of acquiring attribute values. The acquisition methods for different attribute values ​​are different. Static attributes refer to attributes that describe the inherent information of workshop resources. They are acquired and filled manually or read into the computer during workshop activities, and the attribute values ​​do not change over time or have a long change cycle. Dynamic attributes refer to attributes whose values ​​change continuously during workshop activities. They are mainly collected dynamically through processes such as data acquisition, analysis, and calculation, and are updated in real time with changes in the physical resource status values.

[0060] Please see Figure 7Based on the method of obtaining ontology attribute values, attributes can be divided into static attributes and dynamic attributes. Static attributes refer to attributes that describe the inherent information of a physical entity. Their attribute values ​​come from files, etc., are determined during the initialization of the ontology instance, and are subsequently manually filled and modified. Examples include basic attributes such as machine tool ID and machine tool brand, and technical attributes such as maximum power and speed range in a machine tool. Dynamic attributes refer to attributes whose values ​​change continuously during workshop activities. They are mainly collected through data acquisition and calculation processes and updated in real time as the physical entity's state values ​​change. For example, in a machine tool, attribute values ​​such as spindle speed and alarm information are obtained through data acquisition and transmission components; status attributes such as cutting force and relational entity attributes such as tool wear values, in addition to data acquisition, also need to be obtained from tool wear algorithms, etc.

[0061] Specifically, physical entities in the workshop collect data through third-party interfaces or sensors, transmit the data to the host computer via Kafka, and then form dynamic data through algorithm modules or data parsing modules. The host computer also generates static data information through resource registration. Finally, by fusing dynamic and static data, a digital twin model is formed.

[0062] S5. Based on the mapping relationship between the global ontology and each local ontology, multi-dimensional information fusion of digital twins in multiple scenarios is achieved according to the ontology information fusion method.

[0063] Step S5 specifically involves fusing attribute information based on the association relationships between attribute values ​​of the ontology describing the same physical entity, and on the principles of merging complementary information and removing redundant information. In multi-scenario applications, some information appears in multiple local ontology instances. This information only needs to be recorded once during merging, and the rest are deleted as redundant information. Other information is scattered in local ontology instances under different application scenarios. This information is integrated into the global ontology as complementary information. In this way, the fusion of multi-dimensional information describing physical entities under cross-domain multi-scenario applications is realized, so as to display the information of all nodes of the physical entity in the digital twin space.

[0064] Please see Figure 8 Information fusion is achieved based on the mapping relationships between ontology attributes, and follows the principles of merging complementary information and eliminating redundant information. In multi-scenario applications, some attributes describing physical entities appear only once and are recorded as complementary information during information fusion. For example, the current spindle speed and current axis position information only appear in the product processing scenario and are directly filled into the global ontology during fusion. Others appear in multiple scenarios, such as basic machine tool attribute information and fault alarm information, which appear in multiple scenarios such as production scheduling and product processing. These information only need to be recorded once during fusion, and the rest are deleted as redundant information.

[0065] Specifically, the local ontology information and the global ontology information are used to determine whether they describe the same entity. If they do not describe the same entity, they are considered unrelated. If they describe the same entity, an attribute relationship mapping is established, and then the attributes are checked for redundancy. If the attributes are not redundant, attribute values ​​are fused; if the attributes are redundant, the redundant attributes are removed.

[0066] The global ontology information of the machine tool after integration in the three scenarios of production scheduling, process design and product processing, such as Figure 9 As shown, the scheduling ontology model, process ontology model, and manufacturing ontology model are merged into a global ontology model. The machine tool global ontology model includes basic information, technical information, status information, and relational information. Basic information includes Name, ID, brand, serial number, workshop, and unit. Technical information includes machining time, maximum speed, machining accuracy, and axis travel information, with axis travel including corresponding coordinate values. Status information includes alarm information, spindle speed, feed rate, and position information, with position information including corresponding coordinate values. Relational information includes tool number, cutting force, and tool wear value, with cutting force including cutting forces in the X, Y, and Z directions.

[0067] This ontology-based multidimensional information fusion method for workshop digital twins categorizes resources based on their physical characteristics. It then further classifies the attributes of different resource types according to the information they express. A generalizable and shareable resource ontology model is then standardized using an ontology modeling language, ensuring consistency in the description, expression, and understanding of manufacturing resource models. Next, based on cross-domain, multi-scale applications in different scenarios, it focuses on key information and constructs local ontology models of physical entities in different scenarios, collecting information through a combination of dynamic and static data. Finally, it achieves the fusion of multidimensional information in the digital twin across cross-domain, multi-scenario applications through the relationships between ontology models and information fusion methods.

[0068] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A method for multi-dimensional information fusion in an ontology-based digital twin workshop, characterized in that, Includes the following steps: S1. Based on the attribute characteristics of physical resources in the workshop, classify physical resources to standardize the modeling and management activities related to different types of resources. S2. Divide the attributes of the same type of resources according to the perspective of description information, and build a global ontology model of the workshop digital twin based on ontology modeling language to form a global ontology model library of workshop digital twin; S3. Based on the cross-domain and multi-scenario application of the twin workshop, construct local ontology models of physical entities in each scenario, and complete the mapping of each local ontology and the local ontology to the global ontology. Step S3 further includes the following sub-steps: S31. In each application scenario, based on the principle of grasping the main information and ignoring the secondary information, a local ontology model of the physical resources of the workshop required in each scenario is constructed. During the entire product life cycle, the workshop involves scheduling, design and production across multiple scenarios. The information of workshop resources that are of concern in different scenarios is different. In order to improve the application efficiency of the resource ontology model in each scenario, the information of the physical entities of concern in each scenario is extracted, and a local ontology model of resources suitable for this scenario is constructed. S32. Based on the local ontology model information in each scenario, iteratively update the comprehensive information of workshop resources maintained in the global ontology. Based on the identity attributes in each ontology model, realize the association between local ontology and between local ontology and global ontology, so that ontology describing the same resource in cross-domain and multi-scenario scenarios can interoperate. On this basis, perform association mapping between attributes describing the same resource and the same information in the ontology to ensure semantic consistency. S4. Based on the workshop operation status in each scenario and the characteristics of the ontology model attribute value acquisition method, complete the collection of local ontology model information in different scenarios. S5. Based on the mapping relationship between the global ontology and each local ontology, multi-dimensional information fusion of digital twins in multiple scenarios is achieved according to the ontology information fusion method.

2. The ontology-based digital twin workshop multi-dimensional information fusion method according to claim 1, characterized in that, In step S1, based on the characteristics of the production elements in the actual physical workshop, the workshop resources are divided into four categories: manufacturing equipment resources, material resources, auxiliary hardware resources, and human resources. Each resource category can be further subdivided according to actual applications, so that the attribute characteristics of different types of resources are similar, forming a workshop resource tree, which facilitates unified management.

3. The ontology-based digital twin workshop multi-dimensional information fusion method according to claim 1, characterized in that, Step S2 further includes the following sub-steps: S21. Based on step S1, different types of resource attributes are divided according to different perspectives of resource attribute description information. While ensuring that the expressed information is not repeated, the characteristics of workshop resources are represented as comprehensively as possible. S22. Model resource instances using ontology modeling language to build a global ontology resource library; construct resource models based on XML and RDF ontology languages ​​so that resource descriptions can be understood by computers, facilitating information interaction and integration in virtual space.

4. The ontology-based digital twin workshop multi-dimensional information fusion method according to claim 1, characterized in that, Specifically, in the local ontology model corresponding to different scenarios, attributes can be divided into static attributes and dynamic attributes according to the different methods of obtaining attribute values. Different attribute values ​​are collected in different ways. Static attributes refer to attributes that describe the inherent information of workshop resources. They are obtained and filled manually or read by computer during workshop activities. The attribute values ​​do not change with time or have a long change period. Dynamic attributes refer to attributes whose attribute values ​​change continuously during workshop activities. They are mainly collected dynamically through data collection, analysis and calculation processes and updated in real time with changes in the physical resource status values.

5. The ontology-based multidimensional information fusion method for digital twin workshops according to claim 1, characterized in that, Step S5 specifically involves fusing attribute information based on the association relationships between attribute values ​​of the ontology describing the same physical entity, and on the principles of merging complementary information and removing redundant information. In multi-scenario applications, some information appears in multiple local ontology instances. This information only needs to be recorded once during merging, and the rest are deleted as redundant information. Other information is scattered in local ontology instances under different application scenarios. This information is integrated into the global ontology as complementary information. In this way, the fusion of multi-dimensional information describing physical entities under cross-domain multi-scenario applications is realized, so as to display the information of all nodes of the physical entity in the digital twin space.

Images