A cloud manufacturing system digital twin migration modeling method considering preferences

Abstract

The application discloses a kind of considering preference cloud manufacturing system digital twin migration modeling method, based on the production relationship of cloud manufacturing system, the knowledge graph of cloud manufacturing system is constructed;The digital twin model set in the cloud model library of cloud manufacturing system is constructed;The digital twin model includes geometric model, behavior model, logic model and performance model;Based on the characteristics of cloud manufacturing digital twin model migration, the demand set of digital twin modeling of cloud manufacturing service demand side is constructed;According to the interaction and preference relationship between the modeling demand and the model to be selected, a recommendation algorithm is constructed to select the optimal model from the digital twin model library for migration.The migration modeling method considering preference provided by the application provides an efficient and accurate modeling method for the precise application of cloud manufacturing system digital twin, which has important value for the intelligent improvement of cloud manufacturing system.

Description

TECHNICAL FIELD

[0001] The application relates to a digital twin migration modeling method, in particular to a cloud manufacturing system digital twin migration modeling method considering preferences. BACKGROUND

[0002] With the rise of intelligent manufacturing, new technical means and tools such as Internet of Things, big data and artificial intelligence are emerging. These new technologies and tools help the transformation and upgrading of traditional manufacturing industry, so as to promote the high-quality development of manufacturing industry. Cloud manufacturing makes full use of network technology and new generation information technology to realize the cooperation of product design, manufacturing and management of enterprises in the supply chain and across the supply chain. Through the Internet, enterprises on the entire supply chain can share manufacturing resources, realize large-scale rapid response collaborative manufacturing, shorten the production cycle, quickly respond to customer demand and improve production flexibility.

[0003] The introduction of digital twin technology has greatly changed the operation mode of cloud manufacturing. The key to the application of digital twin technology in cloud manufacturing is digital twin model construction. Digital twin model construction is to construct a multi-dimensional, multi-scale, full-factor and high-fidelity physical space twin model in the virtual space through comprehensive perception and fusion of physical space information.

[0004] In the rapid evolution of today's manufacturing industry, manufacturing enterprises are facing increasingly complex market demands and production challenges. Order mixed production mode has become an important means to improve production efficiency and meet individual customer demand. This mode requires manufacturing systems to be flexible in response to diversified product demand while maintaining high efficiency and low cost. However, traditional manufacturing systems often struggle to adapt to this rapidly changing production environment, and a new solution is needed to improve production flexibility and response speed. Therefore, the industry proposes a cloud manufacturing mode, and a cloud manufacturing mode based on digital twin can predict and analyze the future through the support of a digital twin system and provide optimization solutions. However, it takes a lot of modeling time to build a complete digital twin model that virtually maps manufacturing resources to physical entities.

[0005] Under such circumstances, traditional digital twin modeling methods often cannot meet the needs of these advanced manufacturing modes. Traditional methods may lack sufficient flexibility to adapt to rapidly changing production demands, or may not effectively handle the complexity and dynamics of production. Therefore, cloud manufacturing platforms urgently need an efficient digital twin modeling method that can support cloud manufacturing. SUMMARY

[0006] Invention purpose: In view of the problems in the above technical background, the application provides a preferred cloud manufacturing system digital twin migration modeling method, which can migrate knowledge between different production tasks and manufacturing resources, improve the adaptability and generalization ability of the model, and can reflect the state of the manufacturing resource entity in the cloud manufacturing system in real time, predict future production conditions, and provide support for production decision-making.

[0007] Invention content: The preferred cloud manufacturing system digital twin migration modeling method comprises the following steps:

[0008] (1) Based on the production relationship of the cloud manufacturing system, a knowledge graph of the cloud manufacturing system is constructed;

[0009] (2) A digital twin model set in the cloud model library of the cloud manufacturing system is constructed; the digital twin model comprises a geometric model, a behavior model, a logic model and a performance model;

[0010] (3) Based on the migratable characteristics of the cloud manufacturing digital twin model, a demand set for digital twin modeling of the cloud manufacturing service demand side is constructed;

[0011] (4) According to the attention mechanism to capture the interaction and preference relationship between the modeling demand and the selected model, a recommendation algorithm is constructed to select the optimal model from the digital twin model library for migration.

[0012] Further, the step (1) is implemented as follows:

[0013] The cloud manufacturing system production process is analyzed, and the production activities of the cloud manufacturing system are described as the entire process of processing and assembling the product on different production units according to the production progress until the final completion; the key elements of the cloud manufacturing system production process include orders, products, processes and machines; the production unit represents one or more manufacturing resource entities abstracted from each cloud manufacturing service provider;

[0014] The knowledge graph of the cloud manufacturing system describes the real-time manufacturing state of the cloud manufacturing system in the form of nodes and connection edges; the node represents an entity in the cloud manufacturing system, and the connection edge represents the relationship between the nodes; the knowledge graph G is as follows:

[0015] G=(V,E)

[0016] V=(M,O,Pd,Pc)

[0017] E=(MO,MPd,MPc,OPd,…,PdPc)

[0018] Wherein, V represents the node set, E represents the connection edge set, M is the production unit, O is the order, Pd represents the product, Pc represents the process; MO is the connection between the machine and the order, including the machine is processing the order and the machine is not processing the order; MPd is the connection between the machine and the product, including the machine is processing the product and the machine is not processing the product; determined by the position of the product in the actual production process; MPc is the connection between the machine and the process, indicating whether the machine executes a specific process; OPd indicates that the order is composed of products, and the connection can be represented by the processing progress of the product in the order; PdPc is the connection between the product and the process, indicating whether the product is executing the process.

[0019] Further, the geometric model in step (2) is:

[0020] DTM GM = (ME, GP)

[0021] Wherein, ME represents the external model, and GP represents the geographical position.

[0022] Further, the behavior model in step (2) represents the real-time response and behavior of the manufacturing resource entity under the action of the external environment and the internal operation mechanism, and is coupled by a plurality of manufacturing resource entity sub-behavior models, and is:

[0023]

[0024] Wherein, M n represents the number of manufacturing resource entities, and DTM EU represents the sub-behavior model of each manufacturing resource entity, and the formula is as follows:

[0025]

[0026] Wherein, X represents the input event set, Y represents the output event set, DI represents the input data set, DO represents the output data set, D represents the coupling member name set, {F i} represents the sub-function set, EIC represents the external input coupling function set, DIC represents the input data coupling function set, IEC represents the internal event coupling function set, IDC represents the internal data coupling function set, EOC represents the output event coupling function set, and DOC represents the output data coupling relationship set.

[0027] Further, the logic model in step (2) represents the internal logic of mapping the actual production process and operation of the manufacturing resource entity, and is:

[0028] DTM LM = (DTM GM , DTM BM , DTMOR )

[0029] wherein, DTM OR represents a set of running rules, including behavior information transmission rules between different digital twin logical models, information interaction rules between models, synchronization promotion rules between superior and subordinate models, and mutual exclusion priority rules between peer models.

[0030] Further, the performance model in step (2) includes a prediction or decision model for performance indicators, denoted as:

[0031] DTM PM = (DTM MM , DTM DDM )

[0032] wherein, DTM MM represents a mechanism model, and DTM DDM represents a data-driven model.

[0033] The jth digital twin model in the cloud model library is denoted as:

[0034]

[0035] wherein, represents the jth digital twin model in the cloud model library; respectively represent the geometric model, the behavior model, the logic model, and the performance model of the jth digital twin model;

[0036] The set of digital twin models is denoted as:

[0037]

[0038] Further, the step (3) is implemented as follows:

[0039] The construction requirement of the ith digital twin model is denoted as:

[0040]

[0041] wherein, represents the construction requirement of the ith digital twin model; respectively represent the geometric model requirement, the behavior model requirement, the logic model requirement, and the performance model requirement in the construction requirement of the ith digital twin model;

[0042] The set of construction requirements of the digital twin model of the cloud manufacturing service demand side is denoted as:

[0043]

[0044] Further, step (4) said root according to attention mechanism capture modeling needs and the interaction and preference relationship between the selected model, the implementation process is as follows:

[0045] The preference of the i th construction demand to the j th candidate model is represented as:

[0046]

[0047] Wherein, , respectively, represent the preference for geometry, behavior, logic and performance, the symbol in the parentheses represents the importance of the same group of models, that is, the preference value, the preference value is not less than 0, and x gm +x bm +x lm +x pm =1.

[0048] Further, step (4) said the construction of the recommended algorithm implementation process is as follows:

[0049] The interaction matrix between the demand of model construction and the digital twin model is denoted as Under the condition of given interaction matrix and knowledge graph G, the goal of migration modeling is to predict whether the model in the digital twin model library will be selected according to the demand of model construction, and the prediction formula is:

[0050]

[0051] Wherein, indicates the probability of selecting vm de for migration modeling according to the demand of model construction cm pr ω is the to-be-trained parameter of the prediction model.

[0052] Further, step (4) said the optimal model is selected from the digital twin model library for migration implementation process is as follows:

[0053] When a new modeling demand appears, the optimal digital twin model is pushed out from the cloud model library for migration modeling through the recommended algorithm f predict (.). Denoted as the dot product of the model construction demand representation and the candidate model representation The dot product of the vector is denoted as the label of the candidate model for the construction demand, which is the actual migration interaction, that is, when in cm,vm =1, it means that vm de is selected for migration according to cm pr , and in cm,vm =0, then it is not migrated.

[0054] Beneficial effects: compared with the prior art, the beneficial effects of the present application: the present application utilizes the migration learning technology to migrate knowledge between different production tasks and manufacturing resources, thereby improving the adaptability and generalization ability of the model; the present application recommends a suitable model in the model library through the preference algorithm, and after fine-tuning, quickly constructs a digital twin model, which can reflect the state of the manufacturing resource entity in the cloud manufacturing system in real time, predict the future production situation, and provide support for production decision-making; the present application provides a powerful tool for modern manufacturing industry to cope with the challenges brought by advanced manufacturing modes such as network collaborative manufacturing and cloud manufacturing, helps enterprises to improve production efficiency, reduce cost, and enhance market competitiveness. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is the principle diagram of the cloud manufacturing digital twin migration modeling method proposed by the present application;

[0056] Figure 2 is the geometric, behavior, logic and performance modeling flowchart in the cloud manufacturing model library proposed by the present application;

[0057] Figure 3 is the training method flowchart of the recommendation algorithm proposed by the present application. DETAILED DESCRIPTION

[0058] The present application will be further described below in combination with the drawings.

[0059] As shown in the drawings, the present application provides a network system manufacturing system digital twin migration modeling method considering preferences, which specifically includes the following steps: Figure 1

[0060] Step S1, based on the production relationship of the cloud manufacturing system, a knowledge graph of the cloud manufacturing system is constructed, including two processes of analyzing the production process of the cloud manufacturing system and constructing the knowledge graph of the cloud manufacturing system.

[0061] The cloud manufacturing system is a system based on the concept of "manufacturing as a service", which integrates information technology, manufacturing technology, Internet of Things technology, etc., and provides high value-added and low-cost manufacturing services through a cloud computing platform. Its core concepts include utilizing the network and cloud manufacturing platform to organize manufacturing resources according to user needs and providing various on-demand manufacturing services. The cloud manufacturing system is composed of three parties: the cloud manufacturing platform party, the cloud manufacturing service provider party and the cloud manufacturing service demander party. The cloud manufacturing service provider party virtualizes and encapsulates various manufacturing resources owned by the party into cloud service form manufacturing capabilities and stores them in the resource pool of the cloud service platform. The cloud manufacturing service demander party selects manufacturing capabilities that meet the requirements of function, performance, price, time and service quality from the resource pool of the cloud manufacturing platform according to actual needs, combines them according to the business process, meets the manufacturing task demand and organizes production.

[0062] ​The production relationship of the cloud manufacturing system is embodied as follows: network technology is used to enable the planning tasks and manufacturing links of different enterprises located in different geographical locations to be coordinated, and to realize the optimal allocation of dynamic resources and the efficient coordination of the manufacturing process. The knowledge graph representation system of the cloud manufacturing system represents the conversion relationship between the graph model data and the production data established under the production state at a certain time.

[0063] Through analysis of the production process of the cloud manufacturing system, the production activities of the cloud manufacturing system are described as the entire flow of processing and assembling a product on different production units according to the production schedule until the product is finally completed. The key elements of the production process of the cloud manufacturing system include orders, products, processes, and machines, etc. The production unit represents one or more manufacturing resource entities abstracted from each cloud manufacturing service provider.

[0064] The knowledge graph of the cloud manufacturing system describes the real-time manufacturing state of the cloud manufacturing system by using nodes and connecting edges. The nodes represent entities in the cloud manufacturing system, such as orders, products, processes, and machines, etc. The connecting edges represent the relationship between the nodes. The knowledge graph G is formulated as follows:

[0065] G=(V,E)

[0066] V represents the node set, and is formulated as:

[0067] V=(M,O,Pd,Pc)

[0068] In the formula, M is a machine (in the cloud manufacturing system, it refers to a production unit composed of multiple production devices, transfer devices, and storage devices, etc. provided by a cloud manufacturing service provider), O is an order, Pd represents a product, and Pc represents a process.

[0069] E represents the connecting edge set, and is formulated as:

[0070] E=(MO,MPd,MPc,OPd,…,PdPc)

[0071] In the formula, MO is the connection between a machine and an order, indicating whether the machine is executing the order, including two states that the machine is processing the order and the machine is not processing the order; MPd is the connection between a machine and a product, indicating whether the machine is processing the product, including two states that the machine is processing the product and the machine is not processing the product, which is determined by the position of the product in the actual production process; MPc is the connection between a machine and a process, indicating whether the machine is executing a specific process; OPd represents that an order is composed of a product, and the connection can be represented by the processing progress of the product in the order; and PdPc is the connection between a product and a process, indicating whether the product is executing the process.

[0072] Step S2, constructing a digital twin model set in the cloud model library of the cloud manufacturing system, including two processes of representation of the digital twin model in the cloud model library and the digital twin model set (domain).

[0073] The digital twin of the cloud manufacturing system includes a digital twin model, a manufacturing resource entity, a simulation model, a relationship model, and a data model; the digital twin model is connected with the manufacturing resource entity through the simulation model, the relationship model, and the data model; the simulation model is connected with the relationship model, and the relationship model is connected with the data model; the digital twin model is a digital mapping of the manufacturing resource entity in the network space; the simulation model is an intelligent algorithm supporting the digital twin model to run in an actual production state; and the relationship model is an operation logic relationship of the manufacturing resource entity actually performing production.

[0074] The cloud model library is a container for storing digital twin models in the cloud of the cloud manufacturing system, and supports creation, update, query, deletion, and reuse of the digital twin models. The digital twin model has diversity, covering four aspects of geometric model, behavior model, logic model, and performance model.

[0075] As shown in Figure 2 , the representation of the digital twin model in the cloud model library covers four aspects of geometric model, behavior model, logic model, and performance model.

[0076] The geometric model is a 3D model corresponding to the manufacturing resource entity, which can be constructed by 3D modeling tools such as UG, CATIA, Pro / E, SolidWorks, and Zhongwang. The geometric model is represented as:

[0077] DTM GM = (ME, GP)

[0078] Wherein, ME (Model External) represents an external model, and GP (Geographical Position) represents a geographical position.

[0079] The behavior model represents real-time response and behavior of the manufacturing resource entity under the action of external environment and internal operation mechanism, which can cover functions of manufacturing resource entities such as production equipment, transfer equipment, and storage equipment. The production equipment state machine model, the transfer equipment state machine model, and the storage equipment state machine model constructed by the finite state machine are used as sub-models of the behavior model of the manufacturing resource of the cloud manufacturing provider, realize state transition of the sub-models, and then assemble the sub-models into the behavior model of the entire cloud manufacturing system according to the production relationship of the cloud manufacturing. The finite state machine model represents a mathematical model of a limited number of states, transition between the states, and action behaviors. The state reflects the input change from the beginning of the system to the present time, the transition represents switching from one state to another state, and the action describes the activities to be performed at a given time. The behavior model is represented as:

[0080]

[0081] The behavior model is coupled by a plurality of manufacturing resource entity sub-behavior models. n DTM EU represents the sub-behavior model of each manufacturing resource entity, and the formula is as follows:

[0082]

[0083] wherein X represents the input event set, Y represents the output event set, DI represents the input data set, DO represents the output data set, D represents the coupling member name set, {F i} represents the sub-function set, EIC represents the external input coupling function set, DIC represents the input data coupling function set, IEC represents the internal event coupling function set, IDC represents the internal data coupling function set, EOC represents the output event coupling function set, and DOC represents the output data coupling relationship set.

[0084] The logic model represents the internal logic of the actual production process and operation of the mapping manufacturing resource entity, is based on the manufacturing resource entity, follows the information such as geographical layout and production mode, and is constructed according to the behavior and response (operation) rules of the production unit in the actual production process, so as to have the ability of behavior characteristics, response mechanism and state conversion; the logic model can be represented as:

[0085] DTM LM =(DTM GM ,DTM BM ,DTM OR )

[0086] wherein DTM OR represents the operation rule set, including the behavior information transmission rule between different digital twin logic models, the information interaction rule between models, the synchronous promotion rule between upper and lower models, and the mutual exclusion priority rule between models of the same level.

[0087] The performance model is an intelligent analysis model constructed according to the operation law of the cloud manufacturing system on the basis of the geometry model, the behavior model and the logic model, including the prediction or decision model of performance indexes such as production progress, yield, production cost and energy consumption, and this is the core of the digital twin model; the performance model can be represented as

[0088] DTM PM =(DTM MM ,DTM DDM )

[0089] DTM MMDTM represents a mechanism model. DDM DTM represents a data-driven model.

[0090] The above-mentioned geometric model, behavior model, logic model and performance model are integrated, and the jth digital twin model in the cloud model library is represented as:

[0091]

[0092] wherein, DTM represents the jth digital twin model in the cloud model library. , respectively represents the geometric model, behavior model, logic model and performance model of the jth digital twin model.

[0093] The digital twin model set (domain) is denoted as:

[0094]

[0095] Step S3, based on the characteristics of the cloud manufacturing digital twin model that can be migrated, the demand set of the digital twin modeling (including geometric model, behavior model, logic model and performance model) of the cloud manufacturing service demand side is constructed.

[0096] The model-migratable part and parameters are: geometric model (size, assembly relationship and material, etc.), behavior model (production equipment state machine model, transfer equipment state machine model and storage equipment state machine model, etc.), logic model (space position, process flow and constraint rule, etc., which is constructed according to the behaviors and operation rules occurring in the actual production process of cloud manufacturing, so as to have behavior characteristics, response mechanism, and state conversion capability), and performance model (prediction or decision model containing production progress, yield, production cost, energy consumption, etc. after the cloud manufacturing system total model is assembled by manufacturing resource entity sub-models).

[0097] The representation of the digital twin model construction demand of the cloud manufacturing service demand side covers four aspects of geometry, behavior, logic and performance model. Let the i th digital twin model construction demand be represented as:

[0098]

[0099] wherein, DTM represents the i th digital twin model construction demand. , respectively represents the geometric model demand, behavior model demand, logic model demand and performance model demand in the i th digital twin model construction demand.

[0100] The demand set (domain) of the digital twin model construction of the cloud manufacturing service demand side is denoted as:

[0101]

[0102] Step S4, screening the most suitable model from the digital twin model library according to the preference recommendation algorithm for migration.

[0103] The preference algorithm captures the interaction / preference relationship between the modeling requirements and the to-be-selected model according to the attention mechanism; the recommendation algorithm is a prediction algorithm converted by predicting whether the model in the digital twin model library will be selected, and the specific implementation process is as shown in Figure 3 .

[0104] The interaction / preference relationship between the modeling requirements and the to-be-selected model is captured according to the attention mechanism. The preference of the i-th construction requirement to the j-th candidate model is represented as:

[0105]

[0106] wherein, , respectively, represent the preferences for geometry, behavior, logic and performance. The symbols in the parentheses represent the importance in the same group of models, that is, the preference value, the preference value is not less than 0, and x gm +x bm +x lm +x pm =1.

[0107] For example, model B is selected as the base model, and model A is established after fine-tuning based on model B. The importance of model migration lies in That is, when model migration selection is needed, logic and performance are given high priority (preference value) in the same group of models.

[0108] The interaction matrix between the modeling requirements and the digital twin model is denoted as Under the condition of the given interaction matrix and the knowledge graph G, the goal of migration modeling is to predict whether the model in the digital twin model library will be selected according to the modeling requirements, and the prediction formula is:

[0109]

[0110] wherein, represents the probability of selecting vm de for migration modeling according to the modeling requirements cm pr , and ω is the to-be-trained parameter of the prediction model.

[0111] Due to the complex coupling relationship between geometry, behavior, logic and performance models, a digital twin model is usually a unified body of geometry model, behavior model, logic model and performance model, and cannot be selected individually. For example, in the cloud manufacturing service process, one aspect (such as behavior model) is emphasized, and when this behavior model is selected, the geometry, logic and performance models associated with it will be selected together, that is, the overall twin model to which these models are directed will be selected. For example, in the model selection in , when emphasis is placed on selecting (0.2 represents the importance of the same group of models), then associated with it are also selected together, that is, the model is selected. When emphasis is placed on selecting , then

[0112] When new modeling requirements arise, the most suitable digital twin model is pushed out from the cloud model library through the recommendation algorithm f predict (.) for migration modeling. The dot product of the model construction requirement representation and the candidate model representation can be denoted as The dot product of the vectors represents the interaction of the construction requirement to the candidate model, that is, when in cm,vm = 1, it means that the model vm de is selected according to cm pr for migration, and in cm,vm = 0, then it is not migrated.

[0113] Finally, the preferred migration modeling method is used to complete the efficient migration of the digital twin model of the cloud manufacturing system. In the running process of the digital twin model, the virtual data acquisition terminal collects the data of the virtual manufacturing resource entity in real time, and transmits the data to the cloud manufacturing platform. The data acquisition terminal reads the data from the cloud manufacturing platform and acts on the manufacturing resource entity. The manufacturing resource entity performs production, transportation or warehousing actions, obtains the data of the manufacturing resource entity, and transmits the data of the manufacturing resource entity to the cloud manufacturing platform and the digital twin model for comparison. If the data of the manufacturing resource entity and the digital twin model are consistent, it means that the model migration is successful.

[0114] The preferred cloud manufacturing system digital twin migration modeling method provided by the application fully utilizes data and knowledge in cloud manufacturing, effectively migrates existing models (parameter fine-tuning), is simple in theoretical method, easy to implement, greatly improves the digital twin modeling efficiency, and meets the requirements of accurate and efficient model migration. For the precise application of cloud manufacturing system digital twin, it has important value for improving the intelligent level of cloud manufacturing service. The application can be used to guide the accurate and rapid construction process of the cloud manufacturing system digital twin model, solve the problems of long modeling time and high cost of the twin model, and effectively improve the construction efficiency of the twin model.

[0115] The above only describes the preferred embodiments of the application, and it should be pointed out that those skilled in the art can make several improvements and refinements without departing from the principles of the application, and these improvements and refinements should also be considered as the protection scope of the application.

Claims

1. A method for digital twin transfer modeling of cloud manufacturing systems that considers preferences, characterized in that, Includes the following steps: (1) Based on the production relations of the cloud manufacturing system, construct a knowledge graph of the cloud manufacturing system; (2) Construct a set of digital twin models in the cloud model library of the cloud manufacturing system; the digital twin models include geometric models, behavioral models, logical models and performance models; (3) Based on the transferability of cloud manufacturing digital twin models, construct a demand set for digital twin modeling of cloud manufacturing service demanders; (4) Based on the attention mechanism, capture the interaction and preference relationship between modeling needs and the models to be selected, construct a recommendation algorithm, and select the best model from the digital twin model library for transfer. Step (4) involves capturing the interaction and preference relationship between modeling requirements and the models to be selected based on the attention mechanism. The implementation process is as follows: The preference of the i-th construction requirement for the j-th candidate model is expressed as: in, This represents the preference of the i-th construction requirement for the j-th candidate model; Let x represent the preferences for the i-th construction requirement with respect to the j-th geometric model, the i-th construction requirement with respect to the j-th behavioral model, the i-th construction requirement with respect to the j-th logical model, and the i-th construction requirement with respect to the j-th performance model, respectively; gm x bm x lm x pm The sign inside the parentheses indicates the importance of x within the same model group, i.e., the preference value. The preference value is not less than 0, and x gm +x bm +x lm +x pm =1; The process of constructing the recommendation algorithm in step (4) is as follows: The interaction matrix between the requirements for model building and the digital twin model is denoted as: Among them, cm de Indicating the need for building digital twin models, CM de This represents the set of requirements for building a digital twin model; vm pr Representing the digital twin model, VM pr Represents a set of digital twin models; in cm,vm cm de With VM pr The interaction status between them takes a value of 0 or 1, where 0 indicates no interaction and 1 indicates interaction; given an interaction matrix Given a knowledge graph G, the goal of transfer modeling is to predict whether a model in the digital twin model library will be selected based on the needs of model building. The prediction formula is: in, This indicates the requirement (cm) based on the model construction. de Select VM pr The probability of performing transfer modeling; cm de Requirements for building digital twin models; VM pr ω represents the candidate digital twin model; f is the prediction model. predict The parameters to be trained in (.); The process of selecting the optimal model from the digital twin model library for transfer in step (4) is as follows: When new modeling needs arise, the recommendation algorithm f is used. predict (.) Push the best digital twin model from the cloud model library for transfer modeling; Let denot be the dot product of the model construction requirement representation and the candidate model representation. Representing the dot product of vectors, constructing the interaction between the requirements and the candidate model's label as the actual transfer, i.e., when in cm,vm When = 1, it means according to cm de Select VM pr To migrate, in cm,vm When the value is 0, no migration occurs.

2. The method for digital twin migration modeling of cloud manufacturing systems considering preferences, as described in claim 1, is characterized in that... The implementation process of step (1) is as follows: The production process of the cloud manufacturing system is analyzed, and the production activities of the cloud manufacturing system are described as the entire process in which products are processed and assembled on different production units according to the production schedule until final completion; the key elements of the cloud manufacturing system production process include orders, products, processes and machines; the production unit represents one or more manufacturing resource entities that each cloud manufacturing service provider is abstracted into. The knowledge graph of a cloud manufacturing system describes the real-time manufacturing status of the system using nodes and edges. Nodes represent entities within the cloud manufacturing system, and edges represent the relationships between nodes. The formula for the knowledge graph G is as follows: G = (V, E) V = (M, O, Pd, Pc) E = (MO, MPd, MPc, OPd, ..., PdPc) Where V represents the node set, E represents the edge set, M is the production unit, O is the order, Pd represents the product, and Pc represents the process; MO is the connection between the machine and the order, including two states: the machine is processing the order and the machine is not processing the order; MPd is the connection between the machine and the product, including two states: the machine is processing the product and the machine is not processing the product; it is determined by the position of the product in the actual production process; MPc is the connection between the machine and the process, indicating whether the machine is executing a specific process; OPd indicates that the order consists of products, and its connection can be represented by the processing progress of the products in the order; PdPc is the connection between the product and the process, indicating whether the product is executing the process.

3. The method for digital twin migration modeling of cloud manufacturing systems considering preferences, as described in claim 1, is characterized in that... The geometric model described in step (2) is as follows: DTM GM =(ME,GP) Where ME represents the external model and GP represents the geographical location.

4. The method for digital twin migration modeling of cloud manufacturing systems considering preferences, as described in claim 1, is characterized in that... The behavioral model described in step (2) represents the real-time response and behavior of manufacturing resource entities under the influence of the external environment and internal operating mechanisms. It is composed of multiple sub-behavioral models of manufacturing resource entities coupled together, as follows: Among them, M n DTM represents the number of manufacturing resource entities. EU The sub-behavior model representing each manufacturing resource entity is represented by the following formula: Where X represents the set of input events, Y represents the set of output events, DI represents the set of input data, DO represents the set of output data, D represents the set of coupled member names, and {F i } represents the set of sub-functions, EIC represents the set of external input coupling functions, DIC represents the set of input data coupling functions, IEC represents the set of internal event coupling functions, IDC represents the set of internal data coupling functions, EOC represents the set of output event coupling functions, and DOC represents the set of output data coupling relationships.

5. The method for digital twin migration modeling of a cloud manufacturing system considering preferences, as described in claim 1, is characterized in that... The logical model described in step (2) represents the actual production process and internal logic of the mapping manufacturing resource entity, as follows: DTM LM =(DTM GM ,DTM BM ,DTM OR ) Among them, DTM OR It represents the set of operating rules, including rules for the transmission of behavioral information between different digital twin logical models, rules for information interaction between models, rules for synchronous advancement between upper and lower level models, and rules for mutual exclusion and priority between peer models.

6. The method for digital twin migration modeling of cloud manufacturing systems considering preferences, as described in claim 1, is characterized in that... The performance model described in step (2) includes a prediction or decision model for performance indicators, expressed as: DTM PM =(DTM MM ,DTM DDM ) Among them, DTM MM Representational Mechanism Model, DTM DDM Represents a data-driven model; The j-th digital twin model in the cloud model library is represented as: in, This represents the j-th digital twin model in the cloud-based model library; Let each represent the geometric model, behavioral model, logical model, and performance model of the j-th digital twin model, respectively. The set of digital twin models is denoted as:

7. The method for digital twin migration modeling of cloud manufacturing systems considering preferences, as described in claim 1, is characterized in that... The implementation process of step (3) is as follows: The requirements for building the i-th digital twin model are expressed as follows: in, This represents the construction requirements for the i-th digital twin model; These represent the geometric model requirements, behavioral model requirements, logical model requirements, and performance model requirements in the construction requirements of the i-th digital twin model, respectively. The set of requirements for building digital twin models of cloud manufacturing services demanders is denoted as:

Images