An intelligent decision analysis method and system for manufacturing system based on digital twin

By designing an intelligent decision analysis system in the manufacturing system, using the two-way interaction of entity monitoring and digital twin models, the problem of insufficient data interaction between physical production lines and virtual production lines is solved, and efficient dynamic control and automated optimization of the manufacturing system is realized. Improves manufacturing efficiency and product quality.

CN118071091BActive Publication Date: 2024-11-26CHINA TOBACCO YUNNAN IND
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410234189.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-03-01
Publication Date
2024-11-26
Estimated Expiration
2044-03-01

AI Technical Summary

Technical Problem

In the prior art, there is a lack of data interaction between the physical production line and the virtual production line, which makes it difficult for the data analyzed by the virtual production line to guide the physical production line, the reliability of fault prediction is low, and there is a lack of effective intelligent decision-making analysis methods to realize the manufacturing system. efficient dynamic control.

Method used

An intelligent decision-making analysis system based on digital twin manufacturing system was designed, including entity monitoring module, digital twin construction module, data analysis module, intelligent decision-making module, entity control module, data storage module and human-computer interaction module, through real-time monitoring The production information of the manufacturing system is generated, digital twin models are generated, dynamic updates and analysis are performed, real-time target feature sets are extracted, real-time decision-making optimization instructions are generated, and automated control and optimization of the manufacturing system is realized.

Benefits of technology

Realize bidirectional mapping and dynamic interaction between physical entities and virtual models, improve manufacturing efficiency and product quality, reduce invalid data acquisition and processing, ensure real-time communication, enhance system operability and automation control capabilities .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118071091B_ABST
    Figure CN118071091B_ABST
Patent Text Reader

Abstract

The present invention relates to an intelligent decision analysis method and system for a manufacturing system based on a digital twin, and relates to the technical field of intelligent manufacturing. The system comprises an entity monitoring module, which is used to obtain status information of entities in the manufacturing system; a digital twin establishment module, which is used to generate a digital twin model of the manufacturing system based on a preset model library, and dynamically update the digital twin model of the manufacturing system based on the status information of the entities in the manufacturing system; a data analysis module, which is used to process and analyze the digital twin model of the manufacturing system and extract a real-time target feature set; an intelligent decision module, which is used to generate real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module; and an entity control module, which is used to execute the real-time manufacturing decision optimization instructions generated by the intelligent decision module and control the real-time operation of the entities in the manufacturing system. The system has the advantages of realizing efficient and dynamic control of the production process of the manufacturing system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent decision-making analysis method and system for a manufacturing system based on digital twins. Background Art

[0002] Smart manufacturing has prompted large-scale manufacturing companies to transform towards intelligence. The massive amount of data owned by factories, on the one hand, brings great pressure to the intelligent development of factories. On the other hand, data itself is the core driving force for the intelligent development of factories.

[0003] Emerging information technologies (e.g., the Internet of Things, digital twins, etc.) provide a technical foundation for autonomous production and operation in manufacturing workshops. For example, the manufacturing Internet of Things technology builds ubiquitous interconnection and interoperability among people (e.g., operators), machines (e.g., processing machines, transport carts, etc.), and objects (e.g., work-in-progress, auxiliary tools, etc.) in the workshop, making each physical entity identifiable, traceable, and communicative; digital twin technology builds a closed-loop logic of "context perception-simulation calculation-collaborative decision-making-production execution" in the workshop, and through real-time interoperation between information space data / models and physical space entities, the workshop production process is transparent, efficient, and autonomous. The coupling degree between digital twins and intelligent manufacturing is getting higher and higher, and the implementation of digital twin technology is one of the key factors in promoting the development of intelligent manufacturing. Combining communication networks with basic technologies such as cloud computing and digital twins can build or optimize a large number of general technologies. The combination of general technologies and vertical industry scenarios can enable application scenarios in different fields of smart factories.

[0004] In the prior art, under the premise that the physical production line already exists, a virtual production line that is as similar as possible to the physical production line is built, and the two production lines run in parallel. The operating status and data of the virtual production line are used to analyze the status of the physical production line. However, there is a lack of data interaction between the physical production line and the virtual production line, which makes it difficult for the data analyzed by the virtual production line to guide the physical production line. At the same time, the credibility of fault predictions made based on the virtual production line is also low.

[0005] Therefore, it is necessary to provide an intelligent decision-making analysis method and system for a manufacturing system based on digital twins, which can be used to efficiently and dynamically control the production process of the manufacturing system. Summary of the invention

[0006] The present invention provides an intelligent decision-making analysis system for a manufacturing system based on a digital twin, comprising: an entity monitoring module, used to obtain production-related information of an entity of the manufacturing system; a digital twin establishment module, used to generate a digital twin model of the manufacturing system based on a preset model library, and also used to dynamically update the digital twin model of the manufacturing system based on the production-related information of the entity of the manufacturing system obtained by the entity monitoring module; a data analysis module, used to process and analyze the digital twin model of the manufacturing system and extract a real-time target feature set; an intelligent decision-making module, used to generate a real-time manufacturing decision optimization instruction based on the real-time target feature set extracted by the data analysis module; an entity control module, used to execute the real-time manufacturing decision optimization instruction generated by the intelligent decision-making module, and control the real-time operation of the entity of the manufacturing system; a data storage module, used to store the status information of the entity of the manufacturing system obtained by the entity monitoring module, the digital twin model of the manufacturing system, the real-time target feature set and the real-time manufacturing decision optimization instruction; a human-computer interaction module, used to realize the interaction between the user and the digital twin model of the manufacturing system.

[0007] Furthermore, the entity monitoring module includes multiple human monitoring units, multiple equipment monitoring units, multiple product monitoring units and multiple edge computing units, wherein the human monitoring unit is used to obtain the status information of the operator of the manufacturing system, the equipment monitoring unit is used to obtain the status information of the manufacturing equipment of the manufacturing system, and the product monitoring unit is used to obtain the quality information of the products of the manufacturing equipment of the manufacturing system. One edge computing unit corresponds to at least one human monitoring unit, at least one equipment monitoring unit and / or at least one product monitoring unit.

[0008] Furthermore, the physical monitoring module also includes a monitoring and control unit, wherein the monitoring and control unit is used to determine the dynamic monitoring frequency of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units, and is also used to determine the correspondence between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the dynamic monitoring frequency of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units.

[0009] Furthermore, the monitoring and control unit determines the dynamic monitoring frequencies of the multiple human monitoring units, multiple equipment monitoring units and multiple product monitoring units, including: determining at least one target area based on the spatial information of the manufacturing system, and determining the dynamic monitoring frequency of the human monitoring unit based on the at least one target area; determining the failure rate of the manufacturing equipment of the manufacturing system based on the historical operation information of the manufacturing equipment of the manufacturing system, obtaining the association between any two manufacturing equipment of the manufacturing system, and determining the dynamic monitoring frequency of the equipment monitoring unit based on the association between any two manufacturing equipment of the manufacturing system and the failure rate of the manufacturing equipment; determining the dynamic monitoring frequency of the multiple product monitoring units based on the dynamic monitoring frequencies of the multiple equipment monitoring units.

[0010] Furthermore, the monitoring and control unit determines the correspondence between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the dynamic monitoring frequencies of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units, including: determining the data processing volume of each human monitoring unit in a monitoring cycle based on the dynamic monitoring frequency of each human monitoring unit; determining the data processing volume of each equipment monitoring unit in a monitoring cycle based on the dynamic monitoring frequency of each equipment monitoring unit; determining the data processing volume of each product monitoring unit in a monitoring cycle based on the dynamic monitoring frequency of the multiple product monitoring units; determining the correspondence between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the data processing volume of each human monitoring unit in a monitoring cycle, the data processing volume of each equipment monitoring unit in a monitoring cycle and the data processing volume of each product monitoring unit in a monitoring cycle.

[0011] Furthermore, the monitoring and control unit determines the correspondence between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the data processing volume of each of the human monitoring units in a monitoring cycle, the data processing volume of each of the equipment monitoring units in a monitoring cycle and the data processing volume of each of the product monitoring units in a monitoring cycle, including: generating a plurality of candidate correspondences based on the position of each of the human monitoring units and the data processing volume in a monitoring cycle, the position of each of the equipment monitoring units and the data processing volume in a monitoring cycle and the position of each of the product monitoring units and the data processing volume in a monitoring cycle through a relationship generation model; screening the plurality of candidate correspondences to determine the correspondences between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units.

[0012] Furthermore, the data analysis module processes and analyzes the digital twin model of the manufacturing system to extract a real-time target feature set, including: for each manufacturing equipment of the manufacturing system, based on the status information of the manufacturing equipment obtained by the equipment monitoring unit corresponding to the manufacturing equipment and the quality information of the product of the manufacturing equipment obtained by the product monitoring unit, generating a real-time target feature set corresponding to the manufacturing equipment.

[0013] Furthermore, the intelligent decision-making module generates real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module: for each manufacturing equipment of the manufacturing system, the real-time operating status of the manufacturing equipment is determined based on the real-time target feature set corresponding to the manufacturing equipment and the equipment fault map corresponding to the manufacturing equipment; based on the real-time operating status of each manufacturing equipment of the manufacturing system, a real-time manufacturing decision optimization instruction is generated.

[0014] Furthermore, the entity control module executes the real-time manufacturing decision optimization instructions generated by the intelligent decision-making module to control the real-time operation of the entities of the manufacturing system, including: generating a verification result based on the real-time manufacturing decision optimization instructions through the digital twin model; when the verification result meets the preset requirements, executing the real-time manufacturing decision optimization instructions generated by the intelligent decision-making module to control the real-time operation of the entities of the manufacturing system.

[0015] The present invention provides an intelligent decision-making analysis method for a manufacturing system based on a digital twin, which is applied to the above-mentioned intelligent decision-making analysis system for a manufacturing system based on a digital twin, including: generating a digital twin model of the manufacturing system based on a preset model library; obtaining status information of entities of the manufacturing system; dynamically updating the digital twin model of the manufacturing system based on the status information of the entities of the manufacturing system; processing and analyzing the digital twin model of the manufacturing system to extract a real-time target feature set; generating a real-time manufacturing decision optimization instruction based on the real-time target feature set extracted by the data analysis module; executing the real-time manufacturing decision optimization instruction to control the real-time operation of the entities of the manufacturing system; storing the acquired status information of the entities of the manufacturing system, the digital twin model of the manufacturing system, the real-time target feature set and the real-time manufacturing decision optimization instruction; and realizing the interaction between the user and the digital twin model of the manufacturing system based on user instructions.

[0016] Compared with the prior art, the intelligent decision-making analysis method and system for a manufacturing system based on a digital twin provided in this specification have at least the following beneficial effects:

[0017] 1. By real-time monitoring of production-related information of the manufacturing system entities, a digital twin model of the manufacturing system is generated, realizing two-way mapping and dynamic interaction between physical entities and virtual models. At the same time, combined with data analysis, intelligent decision-making and control systems, the automated control and optimization of the manufacturing system is realized, improving manufacturing efficiency and product quality. In addition, data storage and human-computer interaction functions provide users with convenient data query and command input methods, improving the operability of the system;

[0018] 2. By determining the dynamic monitoring frequency of multiple human monitoring units, multiple equipment monitoring units and multiple product monitoring units, the acquisition and processing of invalid data is reduced, and the efficiency of intelligent decision-making analysis of the manufacturing system is improved;

[0019] 3. Based on the data processing volume of each human monitoring unit in a monitoring cycle, the data processing volume of each equipment monitoring unit in a monitoring cycle, and the data processing volume of each product monitoring unit in a monitoring cycle, determine the correspondence between multiple edge computing units and multiple human monitoring units, multiple equipment monitoring units, and multiple product monitoring units to ensure real-time communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0021] Figure 1 It is a module diagram of an intelligent decision-making analysis system for a manufacturing system based on a digital twin shown in an embodiment of the present application;

[0022] Figure 2 It is a flow chart of an intelligent decision-making analysis method for a manufacturing system based on a digital twin shown in one embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.

[0024] Figure 1 is a module diagram of an intelligent decision-making analysis system for a manufacturing system based on a digital twin shown in an embodiment of the present application, such as Figure 1 As shown, an intelligent decision-making analysis system for a manufacturing system based on a digital twin may include a physical monitoring module, a digital twin establishment module, a data analysis module, an intelligent decision-making module, a physical control module, a data storage module and a human-computer interaction module.

[0025] The entity monitoring module can be used to obtain production-related information of entities in the manufacturing system.

[0026] In some embodiments, the entity monitoring module includes multiple human monitoring units, multiple equipment monitoring units, multiple product monitoring units and multiple edge computing units, wherein the human monitoring unit is used to obtain status information of the operator of the manufacturing system, the equipment monitoring unit is used to obtain status information of the manufacturing equipment of the manufacturing system, and the product monitoring unit is used to obtain quality information of the products of the manufacturing equipment of the manufacturing system. One edge computing unit corresponds to at least one human monitoring unit, at least one equipment monitoring unit and / or at least one product monitoring unit.

[0027] For example, the human monitoring unit may include a regional image acquisition device and a positioning device, and obtain the status information of the operator of the manufacturing system, such as the operator's posture, etc., through the image acquired by the image acquisition device. The real-time position of the operator is obtained through the positioning device. The equipment monitoring unit may include a variety of sensors, such as temperature sensors, vibration sensors, sound sensors, and pressure sensors, etc., for obtaining the status information of the manufacturing equipment of the manufacturing system, such as temperature, vibration, sound, and pressure, etc. The product monitoring unit may include a product image acquisition device, an ultrasonic detection device, and a product scanning device, for obtaining the quality information of the products of the manufacturing equipment, such as the appearance of the products of the manufacturing equipment, ultrasonic detection information, point cloud information, crack information, etc.

[0028] In some embodiments, the entity monitoring module also includes a monitoring and control unit, wherein the monitoring and control unit is used to determine the dynamic monitoring frequency of multiple human monitoring units, multiple equipment monitoring units, and multiple product monitoring units, and is also used to determine the correspondence between multiple edge computing units and multiple human monitoring units, multiple equipment monitoring units, and multiple product monitoring units based on the dynamic monitoring frequency of multiple human monitoring units, multiple equipment monitoring units, and multiple product monitoring units.

[0029] In some embodiments, the monitoring and control unit determines the dynamic monitoring frequencies of the multiple human monitoring units, the multiple equipment monitoring units, and the multiple product monitoring units, including:

[0030] Determine at least one target area based on spatial information of the manufacturing system, and determine a dynamic monitoring frequency of a human monitoring unit based on the at least one target area;

[0031] Based on the historical operation information of the manufacturing equipment of the manufacturing system, determine the failure rate of the manufacturing equipment of the manufacturing system, obtain the correlation between any two manufacturing equipment of the manufacturing system, and determine the dynamic monitoring frequency of the equipment monitoring unit based on the correlation between any two manufacturing equipment of the manufacturing system and the failure rate of the manufacturing equipment;

[0032] Based on the dynamic monitoring frequencies of the multiple equipment monitoring units, the dynamic monitoring frequencies of the multiple product monitoring units are determined.

[0033] The spatial information of the manufacturing system may include the relative positional relationship between multiple manufacturing devices. The target area may be an area where an operator is required to perform processing operations. In some embodiments, the monitoring and control unit may pre-acquire multiple sample manufacturing systems, wherein each sample manufacturing system is marked with at least one sample target area. The spatial similarity between the manufacturing system and the sample manufacturing system may be calculated based on the spatial information of the manufacturing system and the spatial information of the sample manufacturing system, and the sample manufacturing system whose spatial similarity is greater than the preset spatial similarity threshold is used as the target sample manufacturing system. The area determination model is based on the spatial information of the manufacturing system, the spatial information of the target sample manufacturing system, and the sample target area of ​​the target sample manufacturing system to determine at least one target area of ​​the manufacturing system and the activity frequency of each target area, wherein the activity frequency of the target area can characterize the frequency and duration of the processing operation performed by the operator of the manufacturing system in the target area, and the higher the activity frequency, the higher the frequency of the processing operation performed by the operator in the target area, and the longer the duration. The area determination model may be a GAN (Generative Adversarial Networks, Generative Adversarial Networks) model.

[0034] The monitoring and control unit can determine the initial monitoring frequency of the human monitoring unit corresponding to the predicted target area based on the predicted frequency of activities in the target area. The higher the predicted frequency of activities in the target area, the higher the initial monitoring frequency of the human monitoring unit corresponding to the target area. The relative position relationship between the operator and the target area can be determined according to the real-time position of the operator, and the initial monitoring frequency of the human monitoring unit corresponding to the target area can be adjusted in real time according to the relative position relationship between the operator and the target area as the dynamic monitoring frequency. For example, when the distance between all operators and the target area is greater than a preset distance threshold, the dynamic monitoring frequency of the human monitoring unit corresponding to the target area is the initial monitoring frequency. For another example, the more operators are located in the target area, the higher the dynamic monitoring frequency of the human monitoring unit corresponding to the target area.

[0035] The monitoring and control unit can determine the failure rate of each manufacturing equipment in the manufacturing system through the historical operation information of the target sample manufacturing system.

[0036] The association relationship between any two manufacturing devices in the manufacturing system can represent that the two manufacturing devices have a production upstream and downstream relationship. For example, there is an association relationship between two manufacturing devices located in the same production line.

[0037] The monitoring and control unit can determine the dynamic monitoring frequency of the equipment monitoring unit based on the association relationship between any two manufacturing equipment in the manufacturing system and the failure rate of the manufacturing equipment according to the following formula:

[0038] ;

[0039] in, is the dynamic monitoring frequency of the equipment monitoring unit corresponding to the i-th manufacturing equipment, To preset the monitoring frequency, is the failure rate of the i-th manufacturing equipment, is the failure rate of the nth manufacturing equipment associated with the ith manufacturing equipment, is the total number of manufacturing equipment that is associated with the i-th manufacturing equipment.

[0040] In some embodiments, the monitoring and control unit determines the correspondence between the plurality of edge computing units and the plurality of human monitoring units, the plurality of equipment monitoring units, and the plurality of product monitoring units based on the dynamic monitoring frequencies of the plurality of human monitoring units, the plurality of equipment monitoring units, and the plurality of product monitoring units, including:

[0041] Based on the dynamic monitoring frequency of each human body monitoring unit, determine the data processing volume of each human body monitoring unit in a monitoring cycle;

[0042] Based on the dynamic monitoring frequency of each equipment monitoring unit, determine the data processing volume of each equipment monitoring unit in a monitoring cycle;

[0043] Based on the dynamic monitoring frequency of multiple product monitoring units, determine the data processing volume of each product monitoring unit in a monitoring cycle;

[0044] Based on the data processing volume of each human monitoring unit in a monitoring cycle, the data processing volume of each equipment monitoring unit in a monitoring cycle, and the data processing volume of each product monitoring unit in a monitoring cycle, the correspondence between multiple edge computing units and multiple human monitoring units, multiple equipment monitoring units, and multiple product monitoring units is determined.

[0045] In some embodiments, the monitoring and control unit determines the correspondence between the plurality of edge computing units and the plurality of human monitoring units, the plurality of equipment monitoring units, and the plurality of product monitoring units based on the data processing volume of each human monitoring unit in a monitoring cycle, the data processing volume of each equipment monitoring unit in a monitoring cycle, and the data processing volume of each product monitoring unit in a monitoring cycle, including:

[0046] Generate multiple candidate corresponding relationships based on the position of each human monitoring unit and the data processing amount in a monitoring cycle, the position of each equipment monitoring unit and the data processing amount in a monitoring cycle, and the position of each product monitoring unit and the data processing amount in a monitoring cycle through a relationship generation model, wherein the relationship generation model may be a GAN (Generative Adversarial Networks) model;

[0047] A plurality of candidate corresponding relationships are screened to determine the corresponding relationships between the plurality of edge computing units and the plurality of human monitoring units, the plurality of equipment monitoring units, and the plurality of product monitoring units.

[0048] Specifically, the priority score of the candidate correspondence is calculated based on the following formula:

[0049] ;

[0050] ;

[0051] in, is the priority score of the jth candidate correspondence, is the load rate of the g-th edge computing unit in the j-th candidate correspondence, is the preset load rate, is the mean communication distance of the g-th edge computing unit in the j-th candidate correspondence, is the communication distance between the gth edge computing unit and the corresponding hth monitoring unit in the jth candidate correspondence, To preset the communication distance, is the total number of monitoring units corresponding to the g-th edge computing unit in the j-th candidate correspondence.

[0052] The candidate correspondence with the largest priority score is taken as the target correspondence, and the correspondence between multiple edge computing units and multiple human monitoring units, multiple equipment monitoring units and multiple product monitoring units is determined.

[0053] The digital twin building module can be used to generate a digital twin model of the manufacturing system based on a preset model library.

[0054] Specifically, first design and build the virtual components required for the digital twin of the manufacturing system, then carry out process design in the digital process design module based on actual production needs, and then arrange the automated production line in the virtual production line planning module based on the designed components and processes. After completing the construction of the digital twin, verify the production line planning effect and control program.

[0055] Build a three-dimensional model of each production factor of the manufacturing system, set properties, and write internal drivers. The basic units of the virtual components are component nodes, behaviors, properties, and characteristics. Taking the automated production line of tobacco production as an example, build virtual components such as humidification equipment, drying equipment, deformation processing equipment, cigarette making machines, filter tip assembling machines, filter rod forming machines, and packaging machines.

[0056] Each virtual component has a unique ID to distinguish the virtual components. Different processing stages of virtual parts will also have different IDs. Taking the automated tobacco production line as an example, the parts loading stage, coding and scanning stage, loading and unloading stage at workstation No. 1, loading and unloading stage at workstation No. 2, loading and unloading stage at workstation No. 3, loading and unloading stage at workstation No. 4, high-pressure cleaning stage and unloading stage will all be distinguished by a proprietary ID.

[0057] Each virtual component contains a root node, which is the basis for building features, properties, behaviors and sub-nodes. I nodes are other component parts that are different from the properties, features and behaviors of the root node. The operation mode of the sub-node can be controlled by writing an internal driver.

[0058] Attributes are variables set for components. Taking the automated production line for tobacco production as an example, the feeding and steering unit includes a root node and a child node based on the root node. The attributes of the root node of the feeding and steering unit include the Cartesian coordinate system position coordinates of the base, the running speed and acceleration of the path, the capacity of the path, etc. The attributes of the child node include the starting position of the tray, the reference axis of the movement, the maximum and minimum values ​​of the speed and acceleration, etc.

[0059] Features include the geometric features and appearance status of the component, which are static characteristics of the component; taking the automated production line of tobacco production as an example, the features of the root node of the feeding and steering unit include the basic size, color and material of the base, and the features of the child nodes include the basic size, color and material of the tray;

[0060] Behavior is the action of a component to complete a specific task, which is the dynamic characteristic of a component. Taking the automated production line of tobacco production as an example, the behavior of the root node refers to the basic settings of the path, the basic settings of the material interface, the basic settings of the servo drive, the Python driver script and statistical data, and the behavior of the child node refers to the basic settings of the container.

[0061] The digital twin establishment module can also be used to dynamically update the digital twin model of the manufacturing system based on the production-related information of the entities of the manufacturing system obtained by the entity monitoring module.

[0062] After planning and verifying the digital twin, the real production line is built. In order to realize the true digital twin, the operating data of the real production line is collected, and the synchronization between the real production line data and the digital twin data is completed. The virtual and real synchronization and operating data are displayed through the data visualization module, and the data processing results of the digital twin are used to guide the actual production.

[0063] The data analysis module can be used to process and analyze the digital twin model of the manufacturing system and extract real-time target feature sets.

[0064] Specifically, for each manufacturing equipment in the manufacturing system, based on the status information of the manufacturing equipment obtained by the equipment monitoring unit corresponding to the manufacturing equipment and the quality information of the products of the manufacturing equipment obtained by the product monitoring unit, a real-time target feature set corresponding to the manufacturing equipment is generated.

[0065] The intelligent decision-making module can be used to generate real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module.

[0066] Specifically include:

[0067] For each manufacturing equipment of the manufacturing system, based on a real-time target feature set corresponding to the manufacturing equipment and an equipment fault map corresponding to the manufacturing equipment, determine the real-time operation status of the manufacturing equipment;

[0068] Generate real-time manufacturing decision optimization instructions based on the real-time operating status of each manufacturing equipment in the manufacturing system.

[0069] Specifically, the equipment failure map may include a target feature set of manufacturing equipment under different failure modes.

[0070] The entity control module can be used to execute the real-time manufacturing decision optimization instructions generated by the intelligent decision module to control the real-time operation of the entity of the manufacturing system.

[0071] Specifically include:

[0072] Optimize instructions based on real-time manufacturing decisions through digital twin models and generate verification results;

[0073] When the verification result meets the preset requirements, the real-time manufacturing decision optimization instructions generated by the intelligent decision-making module are executed to control the real-time operation of the entities in the manufacturing system.

[0074] Specifically, a simulation result is generated based on the real-time manufacturing decision optimization instruction through the digital twin model. According to the simulation result, it is determined whether a failure will occur when the manufacturing system executes the real-time manufacturing decision optimization instruction. When it is determined that a failure will not occur when the manufacturing system executes the real-time manufacturing decision optimization instruction, the verification result meets the preset requirements, and the real-time manufacturing decision optimization instruction generated by the intelligent decision-making module is executed to control the real-time operation of the entities of the manufacturing system.

[0075] The data storage module can be used to store the entity status information of the manufacturing system obtained by the entity monitoring module, the digital twin model of the manufacturing system, the real-time target feature set and the real-time manufacturing decision optimization instructions.

[0076] The human-computer interaction module can be used to realize the interaction between users and the digital twin model of the manufacturing system.

[0077] Figure 2is a flow chart of an intelligent decision-making analysis method for a manufacturing system based on a digital twin shown in an embodiment of the present application, such as Figure 2 As shown, an intelligent decision analysis method for a manufacturing system based on a digital twin may include the following steps:

[0078] Step 210, generating a digital twin model of the manufacturing system based on a preset model library;

[0079] Step 220, obtaining status information of entities of the manufacturing system;

[0080] Step 230, dynamically updating the digital twin model of the manufacturing system based on the status information of the entities of the manufacturing system;

[0081] Step 240, processing and analyzing the digital twin model of the manufacturing system to extract a real-time target feature set;

[0082] Step 250, generating real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module;

[0083] Step 260, executing the real-time manufacturing decision optimization instruction to control the real-time operation of the entity of the manufacturing system;

[0084] Step 270 , storing the acquired entity status information of the manufacturing system, the digital twin model of the manufacturing system, the real-time target feature set, and the real-time manufacturing decision optimization instructions;

[0085] Step 280 , enabling interaction between the user and the digital twin model of the manufacturing system based on user instructions.

[0086] An intelligent decision-making analysis method for a manufacturing system based on a digital twin can be executed by an intelligent decision-making analysis system for a manufacturing system based on a digital twin. For more descriptions of an intelligent decision-making analysis method for a manufacturing system based on a digital twin, please refer to the relevant description of an intelligent decision-making analysis system for a manufacturing system based on a digital twin, which will not be repeated here.

[0087] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An intelligent decision-making analysis system for a manufacturing system based on a digital twin, characterized in that: include: An entity monitoring module, used to obtain production-related information of entities in the manufacturing system; A digital twin establishment module, used to generate a digital twin model of the manufacturing system based on a preset model library, and also used to dynamically update the digital twin model of the manufacturing system based on production-related information of entities of the manufacturing system acquired by the entity monitoring module; A data analysis module, used to process and analyze the digital twin model of the manufacturing system and extract a real-time target feature set; An intelligent decision-making module, used to generate real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module; An entity control module, used to execute the real-time manufacturing decision optimization instructions generated by the intelligent decision module, and control the real-time operation of the entity of the manufacturing system; A data storage module, used to store the entity status information of the manufacturing system acquired by the entity monitoring module, the digital twin model of the manufacturing system, the real-time target feature set and the real-time manufacturing decision optimization instruction; A human-computer interaction module, used to enable interaction between a user and the digital twin model of the manufacturing system; The entity monitoring module includes multiple human monitoring units, multiple equipment monitoring units, multiple product monitoring units and multiple edge computing units, wherein the human monitoring unit is used to obtain the status information of the operator of the manufacturing system, the equipment monitoring unit is used to obtain the status information of the manufacturing equipment of the manufacturing system, and the product monitoring unit is used to obtain the quality information of the product of the manufacturing equipment of the manufacturing system. One edge computing unit corresponds to at least one human monitoring unit, at least one equipment monitoring unit and / or at least one product monitoring unit; the entity monitoring module also includes a monitoring and control unit, wherein the monitoring and control unit is used to determine the dynamic monitoring frequency of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units, and is also used to determine the correspondence between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the dynamic monitoring frequency of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units; the monitoring and control unit determines the dynamic monitoring frequency of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units, including: Based on the spatial information of the manufacturing system, at least one target area is determined, and based on the at least one target area, the dynamic monitoring frequency of the human monitoring unit is determined, wherein the spatial information of the manufacturing system includes the relative position relationship between multiple manufacturing devices, and the target area is an area where an operator needs to perform processing operations, specifically including: pre-acquiring multiple sample manufacturing systems, wherein each sample manufacturing system is marked with at least one sample target area, calculating the spatial similarity between the manufacturing system and the sample manufacturing system based on the spatial information of the manufacturing system and the spatial information of the sample manufacturing system, taking the sample manufacturing system whose spatial similarity is greater than a preset spatial similarity threshold as the target sample manufacturing system, and determining at least one target area of ​​the manufacturing system and the activity frequency of each target area based on the spatial information of the manufacturing system, the spatial information of the target sample manufacturing system and the sample target area of ​​the target sample manufacturing system through a region determination model, wherein the activity frequency of the target area represents the frequency and duration of the processing operation performed by the operator of the manufacturing system in the target area; Based on the historical operation information of the manufacturing equipment of the manufacturing system, the failure rate of the manufacturing equipment of the manufacturing system is determined, and the association relationship between any two manufacturing equipment of the manufacturing system is obtained. Based on the association relationship between any two manufacturing equipment of the manufacturing system and the failure rate of the manufacturing equipment, the dynamic monitoring frequency of the equipment monitoring unit is determined. The association relationship between any two manufacturing equipment of the manufacturing system indicates that the two manufacturing equipment have an upstream and downstream production relationship. Specifically, the monitoring and control unit can determine the dynamic monitoring frequency of the equipment monitoring unit based on the association relationship between any two manufacturing equipment of the manufacturing system and the failure rate of the manufacturing equipment according to the following formula: Among them, f i is the dynamic monitoring frequency of the equipment monitoring unit corresponding to the i-th manufacturing equipment, f0 is the preset monitoring frequency, P i is the failure rate of the i-th manufacturing equipment, P (i,n) is the failure rate of the nth manufacturing equipment associated with the ith manufacturing equipment, and N is the total number of manufacturing equipment associated with the ith manufacturing equipment; Based on the dynamic monitoring frequencies of the multiple equipment monitoring units, the dynamic monitoring frequencies of the multiple product monitoring units are determined. Specifically, based on the predicted activity frequency of the target area, the initial monitoring frequency of the human monitoring unit corresponding to the predicted target area is determined. The higher the predicted activity frequency of the target area, the higher the initial monitoring frequency of the human monitoring unit corresponding to the target area. According to the real-time position of the operator, the relative position relationship between the operator and the target area is determined. According to the relative position relationship between the operator and the target area, the initial monitoring frequency of the human monitoring unit corresponding to the target area is adjusted in real time as the dynamic monitoring frequency. The monitoring and control unit determines the corresponding relationship between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units based on the dynamic monitoring frequencies of the multiple human monitoring units, the multiple equipment monitoring units and the multiple product monitoring units, including: Determining the data processing amount of each of the human body monitoring units in one monitoring cycle based on the dynamic monitoring frequency of each of the human body monitoring units; Based on the dynamic monitoring frequency of each of the equipment monitoring units, determining the data processing volume of each of the equipment monitoring units in a monitoring cycle; Determining the data processing volume of each of the product monitoring units in a monitoring cycle based on the dynamic monitoring frequencies of the multiple product monitoring units; Based on the data processing amount of each human monitoring unit in a monitoring cycle, the data processing amount of each device monitoring unit in a monitoring cycle, and the data processing amount of each product monitoring unit in a monitoring cycle, determining the correspondence between the plurality of edge computing units and the plurality of human monitoring units, the plurality of device monitoring units, and the plurality of product monitoring units specifically includes: Generate multiple candidate corresponding relationships based on the position of each human monitoring unit and the amount of data processed in a monitoring cycle, the position of each device monitoring unit and the amount of data processed in a monitoring cycle, and the position of each product monitoring unit and the amount of data processed in a monitoring cycle through a relationship generation model; The multiple candidate corresponding relationships are screened to determine the corresponding relationships between the multiple edge computing units and the multiple human monitoring units, the multiple equipment monitoring units, and the multiple product monitoring units. Specifically, the priority scores of the candidate corresponding relationships are calculated based on the following formula: Among them, S j is the priority score of the jth candidate correspondence, R g is the load rate of the g-th edge computing unit in the j-th candidate correspondence, R presets is the preset load rate, L (g,mean) is the mean communication distance of the g-th edge computing unit in the j-th candidate correspondence, L (g,h) is the communication distance between the gth edge computing unit and the corresponding hth monitoring unit in the jth candidate correspondence, L presets is the preset communication distance, H is the total number of monitoring units corresponding to the g-th edge computing unit in the j-th candidate correspondence; The candidate correspondence with the largest priority score is taken as the target correspondence, and the correspondence between multiple edge computing units and multiple human monitoring units, multiple equipment monitoring units and multiple product monitoring units is determined.

2. According to claim 1, an intelligent decision-making analysis system for a manufacturing system based on a digital twin, characterized in that: The data analysis module processes and analyzes the digital twin model of the manufacturing system to extract a real-time target feature set, including: For each manufacturing equipment of the manufacturing system, a real-time target feature set corresponding to the manufacturing equipment is generated based on the status information of the manufacturing equipment obtained by the equipment monitoring unit corresponding to the manufacturing equipment and the quality information of the product of the manufacturing equipment obtained by the product monitoring unit.

3. The intelligent decision-making analysis system based on digital twin manufacturing system according to claim 2 is characterized in that: The intelligent decision module generates a real-time manufacturing decision optimization instruction based on the real-time target feature set extracted by the data analysis module: For each manufacturing device of the manufacturing system, determining the real-time operating status of the manufacturing device based on a real-time target feature set corresponding to the manufacturing device and a device fault map corresponding to the manufacturing device; Based on the real-time operating status of each manufacturing equipment in the manufacturing system, a real-time manufacturing decision optimization instruction is generated.

4. The intelligent decision-making analysis system based on digital twin manufacturing system according to claim 3 is characterized in that: The entity control module executes the real-time manufacturing decision optimization instruction generated by the intelligent decision module to control the real-time operation of the entity of the manufacturing system, including: generating a verification result based on the real-time manufacturing decision optimization instruction through the digital twin model; When the verification result meets the preset requirements, the real-time manufacturing decision optimization instructions generated by the intelligent decision module are executed to control the real-time operation of the entity of the manufacturing system.

5. An intelligent decision analysis method for a manufacturing system based on a digital twin, applied to an intelligent decision analysis system for a manufacturing system based on a digital twin as claimed in any one of claims 1 to 4, characterized in that: include: Generating a digital twin model of the manufacturing system based on a preset model library; Acquiring status information of entities of the manufacturing system; Dynamically updating the digital twin model of the manufacturing system based on status information of entities of the manufacturing system; Processing and analyzing the digital twin model of the manufacturing system to extract a real-time target feature set; Generate real-time manufacturing decision optimization instructions based on the real-time target feature set extracted by the data analysis module; Executing the real-time manufacturing decision optimization instruction to control the real-time operation of the entity of the manufacturing system; storing the acquired entity status information of the manufacturing system, the digital twin model of the manufacturing system, the real-time target feature set, and the real-time manufacturing decision optimization instruction; The user interacts with the digital twin model of the manufacturing system based on the user instructions.

Citation Information

Patent Citations

  • Intelligent production system and method based on edge calculation and digital twinning

    CN111857065A

  • Composite machine tool digital twinning monitoring system

    CN112162519A

  • Method for constructing digital twin model of aluminum casting manufacturing production line for operation optimization

    CN116774660A

  • Digital twin processing method and digital twin system

    WO2023246347A1