Digital twin system and construction method thereof

By introducing information source collection module, basic model module and scenario model module into the digital twin system, we automatically obtain and process key data, and solve the problems of poor synchronization and large workload in the construction process of digital twin system in the marine industry, and realize the Efficient and real-time digital twin updates reduce costs.

CN118940530APending Publication Date: 2024-11-12JIANGNAN SHIPYARD (GRP) CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411063224.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

There are problems of poor synchronization and large workload in the construction process of existing digital twin systems in the marine industry, resulting in the problems of increased costs and untimely updates.

Method used

A digital twin system was designed, including information source collection module, basic model module and scene model module. It is connected to the target system through the API interface, automatically obtains and processes key data, generates target data, and realizes real-time update of the digital twin model. .

Benefits of technology

It improves the accuracy and timeliness of the digital twin model, reduces the cost of manpower and material resources, realizes automatic updates, and enhances synchronization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940530A_ABST
    Figure CN118940530A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of digital model design, and provides a digital twinning system which comprises an information source collection module, a basic large model module, a scene large model module and a digital twinning model module. The basic large model module generates first target data according to the first key data, the scene large model module generates second target data according to the second key data, and the digital twin model module updates a digital twin model according to the first target data and the second target data. The conversion from manual data acquisition and processing to automatic data acquisition and processing is realized, the credibility, richness, timeliness and accuracy of the data are improved, and the accuracy of the digital twin model and the timeliness and synchronism of updating are further improved. The invention further provides a construction method of the digital twinning system so as to construct the digital twinning system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of digital model design, and in particular to a digital twin system and a construction method thereof. Background Art

[0002] With the development of information technology, the intelligent and digital transformation of shipbuilding enterprises has become an important factor in improving corporate competitiveness and productivity. Digital twin technology plays an important role in the intelligent and digital transformation of shipbuilding enterprises. Digital twin technology is a technology that creates virtual copies of physical entities to achieve real-time monitoring, analysis and prediction of physical entities. In the field of shipbuilding, digital twin systems can provide enterprises with more accurate decision-making support, optimize the generation process, and improve production efficiency.

[0003] The core of the digital twin system is to collect data from the physical world in real time and accurately, and feed it back to the virtual model in real time to achieve synchronization between the model and the physical entity. At present, in the construction of the digital twin system in the shipping industry, due to the wide variety of equipment in the enterprise, the workload and difficulty of docking work have been increased, which in turn has increased the workload and difficulty of building the digital twin system; when docking with the Enterprise Resource Planning (EPR) system and the Manufacturing Execution System (MES), some shipping companies need to upgrade the system, which increases the cost of building the digital twin system; and when updating relevant information, manual participation is often required, and there are problems such as untimely updates, resulting in poor synchronization between the entity and model of the digital twin system. Summary of the invention

[0004] The purpose of this application is to provide a digital twin system and a construction method thereof, so as to solve the problems of poor synchronization and large construction workload of digital twin models existing in the prior art, effectively improve the construction efficiency of digital twin models, save manpower and material costs, and realize real-time updating of digital twin models.

[0005] To achieve the above objectives and other related objectives, the present application provides a digital twin system, including:

[0006] An information source collection module, connected to a first type of target system and a second type of target system of an enterprise, wherein the information source collection module obtains first key data from the first type of target system and obtains second key data from the second type of target system;

[0007] A basic large model module, communicatively connected to the information source collection module, the basic large model module generates first target data according to the first key data;

[0008] A scene large model module, communicatively connected to the information source collection module, the scene large model module generates second target data according to the second key data;

[0009] A digital twin model module is communicatively connected to the basic large model module and the scene large model module, and the digital twin model module is used to update the digital twin model according to the first target data and the second target data.

[0010] Optionally, the information source collection module includes a first communication connection unit, a first data acquisition unit, a second data acquisition unit, a manual input unit and a dictionary acquisition unit;

[0011] The first communication connection unit connects the information source collection module to the basic large model module and the scene large model module. The dictionary acquisition unit can acquire a data target dictionary or a data dictionary from the first type of target system and / or the second type of target system. The first data acquisition unit is used to acquire first key data from the first type of target system. The second data acquisition unit is used to acquire second key data from the second type of target system. The data in the information source collection module can be manually processed through the manual input unit.

[0012] Optionally, the second key data can be labeled and labeling information can be generated based on the first key data and the manual input unit, the basic large model module can generate first algorithm data based on the second key data and the labeling information, the scene large model module is communicated with the basic large model module, and generates a scene algorithm model based on the first algorithm data and the second key data, and the scene algorithm model can generate second target data based on the second key data.

[0013] Optionally, the first type of target system includes automated office systems, legal systems and financial systems, the second type of target system includes enterprise security systems, the first key data includes first target text information, and the second key data includes second target image information and second target video information.

[0014] Optionally, the basic large model module includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data calculation unit and an intelligent prediction unit.

[0015] The present application also provides a method for constructing a digital twin system, which is used to construct any digital twin system in the aforementioned embodiments, and comprises the following steps:

[0016] Establish information source collection module and connect it to the first and second target systems of the enterprise;

[0017] Using the information source collection module to obtain first key data from the first type of target system, and to obtain second key data from the second type of target system;

[0018] Annotate the second key data according to the first key data and generate annotation information;

[0019] Constructing a basic large model module and connecting it to the information source collection module;

[0020] Generate first target data and first algorithm data using the basic large model module according to the first key data, the second key data and the annotation information;

[0021] Constructing a scene large model module according to the first algorithm data, the second key data and the annotation information, and using the scene large model module to generate second target data;

[0022] Create a digital twin model module, and update the digital twin model in the digital twin model module according to the first target data and the second target data.

[0023] Optionally, connecting the information source collection module to the first type of target system and the second type of target system comprises the following steps:

[0024] Acquire target system information, where the target system information includes first-category system information and second-category system information;

[0025] According to the first type of system information, the information source collection module is connected to the first type of target system, and the information source collection module is connected to the first type of target system having an API interface through an API interface;

[0026] According to the second type of system information, the information source collection module is connected to the second type of target system, and the information source collection module is connected to the second type of target system having an API interface through an API interface.

[0027] Optionally, obtaining first key data from the first-type target system includes the following steps:

[0028] Performing a data target dictionary acquisition operation on the first type of target system to acquire a data target dictionary;

[0029] If the data target dictionary is successfully obtained, obtaining first key data from the first type of target system according to the data target dictionary;

[0030] If the data target dictionary cannot be obtained, performing a data dictionary acquisition operation on the first type of target system to acquire a data dictionary, the data dictionary including a data target dictionary;

[0031] If the data dictionary is successfully obtained, obtaining first key data from the first type of target system according to the data dictionary;

[0032] If the data dictionary cannot be obtained, the first key data is obtained from the first type of target system using a web crawler technology.

[0033] Optionally, constructing a basic large model module includes the following steps:

[0034] Build the model load-bearing unit;

[0035] A basic model unit is built in the model carrying unit to preliminarily establish a basic large model module, wherein the basic model unit includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data calculation unit, and an intelligent prediction unit;

[0036] The initially established basic large model module is trained according to the preset first sample data and the preset second sample data to construct the basic large model module.

[0037] Optionally, the scene big model module includes multiple scene algorithm models, and after updating the digital twin model, the following steps are further included:

[0038] Testing and calibrating the scene algorithm model in the scene large model module;

[0039] Regenerate the second target data according to the calibrated scene algorithm model, and update the digital twin model again;

[0040] Testing the updated digital twin model and generating a test result;

[0041] According to the detection result, updating the first key data and / or the second key data;

[0042] The digital twin model is updated again according to the updated first key data and / or the second key data.

[0043] The digital twin system and construction method provided in this application have at least the following beneficial effects:

[0044] The information source collection module is used to obtain the first key data and the second key data, which realizes the transition from manual data acquisition to automatic data acquisition, improves the credibility, richness, timeliness and accuracy of the data, and thus improves the accuracy and timeliness of the digital twin model; through the manual input unit of the information source collection module, the data in the information source collection module can be manually corrected, further improving the accuracy of the data; through the basic large model module and the scene large model module, the first target data and the second target data are generated according to the first key data and the second key data respectively, which realizes the transition from manual processing to automatic processing of data, improves the efficiency of data processing, and realizes high-frequency real-time updating of the first target data and the second target data, so that the digital twin model can realize the transition from manual updating to automatic updating, and from static difficult updating to dynamic real-time updating, further improving the accuracy of the digital twin model and the timeliness and synchronization of updating. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 Shown is a structural schematic diagram of the digital twin system provided in Example 1 of the present application.

[0047] Figure 2 Shown is a flow chart of the method for constructing a digital twin system provided in Example 2 of the present application. DETAILED DESCRIPTION

[0048] In order to make the technical purpose, technical solution and technical effect of the present application clearer, the technical solution in the present application will be clearly and completely described in conjunction with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the embodiments of the present application is not intended to limit the scope of the present application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present application. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0050] In the description of the present application, it should be noted that the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the implementation or example are included in at least one implementation or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same implementation or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more implementations or examples in a suitable manner.

[0051] Embodiment 1

[0052] This embodiment provides a digital twin system, referring to Figure 1 , including information source collection module, basic large model module, scene large model module and digital twin model module.

[0053] The information source collection module is connected to the first type of target system and the second type of target system of the enterprise. The information source collection module can obtain the first key data from the first type of target system and the second key data from the second type of target system; the basic large model module is communicatively connected to the information source collection module, and the basic large model module can generate the first target data according to the first key data; the scene large model module is communicatively connected to the information source collection module, and the scene large model module can generate the second target data according to the second key data; the digital twin model module is communicatively connected to the basic large model module and the scene large model module, and the digital twin model module has a digital twin model. The digital twin model module can update the digital twin model according to the first target data and the second target data.

[0054] In an optional embodiment, the information source collection module has an API (Application Programming Interface) interface, which can collect video, image, text and other information from various systems of the enterprise, and send relevant information to other platforms, systems, modules or units, and the information source collection module also has a manual input function, which can manually input, annotate and exclude data, thereby realizing functions such as manual data supplementation and manual correction of erroneous information. Optionally, the information source collection module includes a first communication connection unit, a first data acquisition unit, a second data acquisition unit, a manual input unit and a dictionary acquisition unit.

[0055] Among them, the first communication connection unit is used to connect the information source collection module to other modules, for example, to connect the information source collection module to the basic large model module and the scene large model module; the dictionary acquisition unit can obtain data target dictionaries or data dictionaries from the first type of target system and / or the second type of target system, and according to the data target dictionary or digital dictionary, the information source collection module can obtain data from the first type of target system and / or the second type of target system; the first data acquisition unit is used to obtain first key data from the first type of target system, and the second data acquisition unit is used to obtain second key data from the second type of target system; the manual input unit can enable the information source collection module to have a manual input function, and the data in the information source collection module can be manually processed through the manual input unit, such as inputting, marking, correcting, excluding and other operations of data, thereby improving the accuracy of the data.

[0056] In an optional embodiment, a manual input unit can be used to annotate the second key data according to the first key data and generate annotation information, and the basic large model module can generate the first algorithm data according to the second key data and the annotation information; the scene large model module is communicatively connected to the basic large model module, and the scene large model module can generate a scene algorithm model according to the first algorithm data and the second key data. Optionally, the scene algorithm large model can be used manually to generate a scene algorithm model according to the first algorithm data and the second key data; the second key data is input into the scene algorithm model, and the scene algorithm model can generate the second target data according to the second key data.

[0057] In an optional embodiment, the first type of target system includes an automated office system (OA), a legal system, and a financial system, the second type of target system includes an enterprise security system, the first key data includes a first target text information, the first target text information is text information related to product production, such as product related information and production equipment related information; the second key data is acquired video image information related to product production, for example, the second key data includes second target image information and second target video information. Optionally, the first type of target system may also include an EPR system, an MES system, a customer relationship management system (CRM), and other systems.

[0058] Among them, the legal system has the company's contract data and other related data, such as the contract agreement for purchasing production equipment. The contract agreement contains the technical parameters, brand image, installation time, etc. of the production equipment. According to the contract agreement, the product information produced by the company can be obtained; the financial system has the company's payment data and other related data. For example, according to the payment amount and payment progress in the payment data, the product's production completion node and production status can be obtained, such as the start-up advance payment node, launching node, sea trial node, and delivery node; the OA system has notification and announcement data and other related data. For example, according to the notification and announcement data, the product's generation node can be obtained, such as embarkation, section assembly, nondestructive testing, etc.; the enterprise security system covers most of the company's production areas. The enterprise security system can be used to obtain real-time information such as product production.

[0059] In this embodiment, the basic large model module includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data operation unit and an intelligent prediction unit. The language recognition processing unit, the visual analysis unit, the data operation unit and the intelligent prediction unit are used to process the first key data and the second key data and generate the first target data and the first algorithm data. The machine learning unit can train and learn the basic large model module to improve the reliability and accuracy of the basic large model module. Optionally, the language recognition processing unit adopts NLP natural language recognition processing, the visual analysis unit adopts CV computer vision analysis, and the machine learning unit adopts MMML multimodal information source machine learning.

[0060] In an optional embodiment, the first target data may include, for example, contract information, financial information, etc., and the basic big model module can be used to summarize or infer the first target data based on the first key data; the first algorithm data includes relevant algorithms for identifying and processing the second key data, and the basic big model can be used to generate the first algorithm data for subsequent establishment of a scenario algorithm model.

[0061] In this embodiment, the scene big model module has multiple scene algorithm models, and the scene algorithm model can reflect various production scenes in the actual product production process. The scene algorithm model can generate second target data based on the second key data and annotation information. The second target data can reflect the status of actual product production. Multiple scene algorithm models can be established according to the first algorithm data, the second key data and the annotation information, thereby building a scene big model module.

[0062] In an optional embodiment, the second target data includes product stage information, product location information, transportation device location information, instrument reading information, large-scale production equipment information, personnel location information, personnel work task information, personnel working hours information, and personnel information in dangerous areas; wherein, the product stage information includes various stages of ship product production, including small assembly, medium assembly, large assembly, final section, closure, slipway transverse movement, pontoon launching, dock mooring, and ship delivery ceremony; the product location information includes the product number, location, and enterprise site code of the ship product; the transportation device location information includes the flatbed truck location information and vehicle location information, and the flatbed truck location information includes the license plate of the flatbed truck. The vehicle location information includes the vehicle's license plate number, location, and enterprise site code; the instrument reading information includes electricity meter information, gas meter information, water meter information, and crane load-bearing information; the large-scale production equipment information includes the working hours, idle status, personnel on-the-job status, and temperature of the large-scale production equipment; the personnel location information includes the personnel's identity, location, and enterprise site code; the personnel work task information includes cutting, welding, grinding, painting, assembly, scaffolding, and rest; the personnel working time information includes the personnel's identity, working hours, and work location; the dangerous area personnel information includes the dangerous area information and the identity of the personnel in the dangerous area.

[0063] In this embodiment, the digital twin model module has a digital twin model, the basic large model module and the scene large model module transmit the generated first target data and second target data to the digital twin model module respectively, and the digital twin model module updates the digital twin model module according to the first target data and the second target data, so that the digital twin model can accurately reflect the current physical real world production status. In addition, the information source collection module can automatically obtain the first key data and the second key data in real time, realizing the automatic and timely update of the digital twin model.

[0064] The digital twin system provided in this embodiment realizes the transformation from manual data acquisition to automatic data acquisition through the information source collection module, thereby improving the credibility, richness, timeliness and accuracy of the data, and further improving the accuracy and timeliness of the digital twin model; through the manual input unit of the information source collection module, the data in the information source collection module can be manually corrected, further improving the accuracy of the data; through the basic large model module and the scene large model module, various production scenarios in the actual product production process are effectively reflected, realizing the transformation from manual processing to automatic processing of data, and improving the efficiency of data processing; and through the basic large model module and the scene large model module, the first target data and the second target data are generated according to the first key data and the second key data respectively, further improving the efficiency of data processing, realizing high-frequency real-time updates of the first target data and the second target data, so that the digital twin model realizes the transformation from manual update to automatic update, and from static difficult update to dynamic real-time update, further improving the accuracy of the digital twin model and the timeliness and synchronization of updates.

[0065] Embodiment 2

[0066] This embodiment provides a method for constructing a digital twin system, which is used to construct any digital twin system in the first embodiment. Figure 2 , the method for constructing the digital twin system of this embodiment includes steps S1 to S7.

[0067] Step S1: Establish an information source collection module and connect it to the first type of target system and the second type of target system of the enterprise.

[0068] In this embodiment, an information source collection module is established by respectively building a first communication connection unit, a first data acquisition unit, a second data acquisition unit, a manual input unit, and a dictionary acquisition unit, wherein the information source collection module has an API interface, which can collect video, image, text and other information from various systems of the enterprise, and send relevant information to other platforms, systems or modules. Optionally, the information source collection module also includes a language conversion unit, which is used to convert the data input through the manual input unit into structured SQL (Structured Query Language) data. The information source collection module is communicatively connected to the first type of target system and the second type of target system by using the first communication connection unit, so that the information source collection module can obtain data from the first type of target system and the second type of target system.

[0069] In an optional embodiment, the information source collection module is connected to the first type of target system and the second type of target system, including steps S11 to S13.

[0070] S11: Acquire target system information, the target system information includes first-category system information and second-category system information. The target system information includes system identification information, and the system that needs to be connected to the information source collection module can be determined according to the system identification information. Specifically, the first-category system information includes first system identification information, and the first-category target system that needs to be connected to the information source collection module can be determined according to the first system identification information. The second-category system information includes second system identification information, and the second-category target system that needs to be connected to the information source collection module can be determined according to the second system identification information.

[0071] S12: According to the first-category system information, the information source collection module is connected to the first-category target system, and the information source collection module is connected to the first-category target system with the API interface through the API interface. Specifically, some of the first-category target systems are provided with the API interface, and other first-category target systems do not have the API interface. The information source collection module is connected to the first-category target system with the API interface through the API interface, and the information source collection module is connected to the first-category target system without the API interface through other appropriate communication methods.

[0072] S13: Connect the information source collection module to the second type target system according to the second type system information, and make the information source collection module docked to the second type target system having the API interface through the API interface. Specifically, when the second type target system has an API interface, the information source collection module is docked to the second type target system through the API interface; when the second type target system does not have an API interface, the information source collection module is communicatively connected to the second type target system through other appropriate means.

[0073] In this embodiment, the enterprise security system includes a first camera device, and the enterprise security system can obtain relevant information such as videos and images related to enterprise production through the first camera device. Figure 1 Before connecting the information source collection module to the second type of target system, the following steps are also included: placing the second camera device and the third camera device at a preset position; and connecting the second camera device and the third camera device to the enterprise security system for communication.

[0074] In an optional embodiment, the second camera device is a camera, for example, the second camera device can be a fast ball camera, and the position of the second camera device can be set according to the design requirements of the digital twin system. For example, multiple second cameras can be arranged in the area where the instrument is located to identify the instrument type and monitor the instrument reading, or the second camera device can be arranged in an area that the first camera device cannot monitor to meet the design requirements; the third camera device is a temperature measuring camera, for example, it can be an infrared temperature measuring camera, and the layout position of the third camera device can be set according to the design requirements of the digital twin system. For example, multiple second cameras can be arranged in the area where large-scale production equipment is located to monitor the temperature of large-scale production equipment, and then monitor the operating status and abnormal conditions of large-scale production equipment. By additionally installing the second camera device and the third camera device on the original enterprise security system, the data source of the second type of target system can be expanded, and all-round monitoring of related equipment in production can be achieved, making the established digital twin model more accurate.

[0075] Step S2: using the information source collection module to obtain first key data from the first type of target system, and to obtain second key data from the second type of target system.

[0076] In this embodiment, the first key data is obtained from the first target system by using the first data acquisition unit of the information source collection module, and the second key data is obtained from the second target system by using the second data acquisition unit of the information source collection module. Optionally, the first key data includes the first target text information. The first target system has text information related to enterprise production. The first data acquisition unit can be used to obtain text information related to product production, that is, the first target text information. The first target text information can include product specification information, product type information, payment information, production equipment parameter information, etc.; the second key data includes the second target image information and the second target video information. The second target system can obtain video information and image information related to product production in real time, that is, the second key data. The second key data can include flatbed vehicle moving video information, ship delivery ceremony video information, pontoon launching video information, personnel operation video information, equipment image information, instrument reading information, etc.

[0077] In an optional embodiment, obtaining the first key data from the first type of target system includes steps S21 to S25.

[0078] S21: Perform a data target dictionary acquisition operation on the first type of target system to acquire a data target dictionary. Specifically, when the first type of target system has a data target dictionary, the data target dictionary is acquired from the first type of target system; optionally, when the data target dictionary cannot be acquired from the first type of target system and the enterprise or equipment manufacturer can provide the data target dictionary, the data target dictionary can be acquired by other appropriate methods. Furthermore, the dictionary acquisition unit of the information source collection module is used to acquire the data target dictionary.

[0079] S22: If the data target dictionary is successfully obtained, first key data is obtained from the first type of target system according to the data target dictionary. Optionally, the first data acquisition unit of the information source collection module is used to obtain the first key data from the first type of target system according to the data target dictionary.

[0080] S23: If the data target dictionary cannot be obtained, a data dictionary acquisition operation is performed on the first type target system to obtain the data dictionary; wherein the data dictionary includes the data target dictionary, and the data target dictionary is the part of the data dictionary related to product production. Specifically, when the first type target system has a data dictionary, the dictionary acquisition unit is used to acquire the data dictionary from the first type target system; optionally, when the data dictionary cannot be acquired from the first type target system, and the enterprise or equipment manufacturer can provide the data dictionary, other suitable methods can be used to acquire the data dictionary.

[0081] S24: If the data dictionary is successfully obtained, first key data is obtained from the first type of target system according to the data dictionary. Optionally, the first key data is obtained from the first type of target system according to the data dictionary using a first data obtaining unit.

[0082] S25: If the data dictionary cannot be obtained, the first key data is obtained from the first target system using a web crawler technology. The first data acquisition unit of the information source collection module is in communication connection with the first target system, and the first key data is obtained from the first target system using a web crawler technology through the first data acquisition unit.

[0083] In this embodiment, the second data acquisition unit is connected to the second type of target system through the first communication connection unit, and the second key data can be acquired from the second type of target system in real time by using the second data acquisition unit.

[0084] Through the information source collection module, the first key data can be directly obtained from the first type of target system, and the second key data can be obtained in real time from the second type of target system, realizing the transformation from manual data acquisition to automatic data acquisition, bypassing the need to transform and upgrade the relevant systems of the enterprise, reducing material and labor costs, and improving the credibility, richness, timeliness and accuracy of the data, thereby improving the accuracy of the established digital twin model.

[0085] Step S3: labeling the second key data according to the first key data and generating labeling information.

[0086] In this embodiment, the second key data includes second target image information and second target video information, and the first key data includes first target text information. The products in the second key data can be marked according to the first target text information, and marking information can be formed according to the marked data. For example, the production stage of the product in the second key data, the location of the product, the working status of the product production equipment, the work content of the personnel, the temperature of the production equipment, etc. can be marked.

[0087] In an optional embodiment, the information source collection module further includes a data labeling unit, which can be used to automatically label the second key data, or can be manually labeled using a manual input unit.

[0088] Step S4: Construct a basic large model module and connect it to the information source collection module.

[0089] In this embodiment, constructing a basic large model module includes the following steps: constructing a model bearing unit; constructing a basic model unit in the model bearing unit to preliminarily form a basic large model module, wherein the basic model unit includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data operation unit, and an intelligent prediction unit; training the preliminarily established basic large model module according to the preset first sample data and the second sample data to construct the basic large model module. Optionally, the model bearing unit can be, for example, a computer system, and the basic large model module is stored in the computer system.

[0090] In an optional embodiment, an intelligent computing system is provided in the enterprise, and the first sample data can be obtained from the first type of target system, and the second sample data can be obtained from the second type of target system. For example, the information source collection module can be used to obtain the first sample data and the second sample data from the first type of target system and the second type of target system, respectively, or other suitable methods can be used to obtain the first sample data and the second sample data. Based on the obtained first sample data and second sample data, the basic large model module is trained using the intelligent computing system to optimize the basic large model module.

[0091] Step S5: Generate first target data and first algorithm data using the basic large model module according to the first key data, the second key data and the annotation information.

[0092] In this embodiment, the information source collection module is connected to the basic large model module in communication, and the first key data, the second key data, and the annotation information are sent to the basic large model module through the information source collection module, and the basic large model module generates the first target data and the first algorithm data according to the first key data, the second key data, and the annotation information. Specifically, the basic large model module can generate the first target data by processing the first key data, and can generate the first algorithm data by identifying and processing the second key data and the annotation information.

[0093] Step S6: construct a scene large model module according to the first algorithm data, the second key data and the annotation information, and use the scene large model module to generate the second target data.

[0094] In this embodiment, based on the second key data and annotation information in the information source collection module and the first algorithm data generated by the basic big model module, a scene big model module can be built, and the scene big model module can reflect various production scenes in the actual product production process; based on the second key data and annotation information, the scene big model module can be used to generate second target data, and the second target data is used to reflect the current status of the actual product production process.

[0095] Specifically, multiple scene algorithm modules are established according to the first algorithm data, the second key data and the annotation information to construct a scene large model module, and the multiple scene algorithm modules can reflect the actual production conditions in which multiple production links are carried out simultaneously in different working areas in actual production. Optionally, the ship enterprise has an intelligent computing system, and before using the scene large model module to generate the second target data, the following steps are also included: according to the second sample data, the scene large model module is trained using the intelligent computing system; the trained scene large model module can more accurately reflect the actual production scene, thereby improving the accuracy of the second target data.

[0096] Step S7: Create a digital twin model module, and update the digital twin model in the digital twin model module according to the first target data and the second target data.

[0097] In this embodiment, the digital twin model module has a digital twin model. The staff can establish a digital twin model of the corresponding scene around each scene of product production according to the actual task to create a digital twin model module, such as establishing a digital twin model of a production line; connect the digital twin model module to the basic large model module and the scene large model module, and obtain the first target data and the second target data. Specifically, the generated first target data is transmitted to the digital twin model module through the basic large model module, and the generated second target data is transmitted to the digital twin model module through the scene large model module; update the digital twin model according to the first target data and the second target data. By docking the digital twin model module with the basic large model module and the scene large model module, the basic large model module and the scene large model module are both connected to the information source collection module, and the high-frequency real-time update of the first target data and the second target data is realized, so that the digital twin model realizes the transformation from manual update to automatic update, from static difficult update to dynamic real-time update, so that the digital twin model can truly reflect the current physical real-world processing and production status.

[0098] In an optional embodiment, the scene large model module includes multiple scene algorithm models, and after updating the digital twin model, it also includes steps S71 to S75.

[0099] S71: Check and calibrate the scene algorithm model in the scene large model module. Specifically, when multiple scene algorithm models are running at the same time, contradictory data may be generated. By checking and calibrating multiple different scene algorithm models, the contradictory data between different scene algorithm models can be corrected. Optionally, multiple different scene algorithm models can be checked and calibrated manually. For example, according to the OA notification information, it is concluded that the No. 2 section of the ship is located on the No. 8 platform, and according to the second target video information, it is concluded that the No. 2 section of the ship is located on the No. 9 platform. After manual inspection and judgment, the data of the OA notification information is adopted.

[0100] S72: Regenerate the second target data according to the calibrated scene algorithm model, and update the digital twin model again. Specifically, use the calibrated scene large model module to regenerate the second target data according to the second key data, and transmit the regenerated second target data to the digital twin model module to update the digital twin model again.

[0101] S73: Test the updated digital twin model and generate a test result. Optionally, the digital twin model can be tested manually to determine whether the digital twin model has defects, generate a test result, and record the existing defects. For example, after testing the digital twin model, it is found that some data content is missing, and generate a test result including the missing data content.

[0102] S74: Update the first key data and / or the second key data according to the detection result. Specifically, according to the content recorded in the detection result, use the information source collection module to re-extract the missing data content from the first target system to update the first key data, and / or, re-extract the missing data content from the second target system to update the second key data, and / or, update the first target system and / or the second target system, for example, connect the information source collection module to a new information system to update the first target system and / or the second target system, and obtain the missing data content from the updated first target system and / or the second target system to update the first key data and / or the second key data.

[0103] S75: Re-update the digital twin model according to the updated first key data and / or second key data. Specifically, re-perform steps S1 to S7 according to the updated first key data and / or second key data to update or supplement the scene algorithm model to re-update the digital twin model.

[0104] By testing, calibrating and updating the scene large model module and the digital twin model, the reliability of the digital twin model is effectively improved, allowing the digital twin model to more accurately reflect the actual product production scenario.

[0105] The method for constructing a digital twin system provided in this embodiment is used to construct the digital twin system in Embodiment 1, and therefore, has the same beneficial effects as Embodiment 1. The first key data and the second key data are automatically acquired through the information source collection module, which effectively improves the synchronization of the digital twin model and the work efficiency of the construction, and further improves the accuracy and reliability of the digital twin module by testing and calibrating and updating the scene algorithm model and the digital twin model.

[0106] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify, change or combine the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A digital twin system, characterized in that: include: An information source collection module, connected to a first type of target system and a second type of target system of an enterprise, wherein the information source collection module obtains first key data from the first type of target system and obtains second key data from the second type of target system; A basic large model module, communicatively connected to the information source collection module, the basic large model module generates first target data according to the first key data; A scene large model module, communicatively connected to the information source collection module, the scene large model module generates second target data according to the second key data; A digital twin model module is communicatively connected to the basic large model module and the scene large model module, and the digital twin model module is used to update the digital twin model according to the first target data and the second target data.

2. The digital twin system according to claim 1, characterized in that: The information source collection module includes a first communication connection unit, a first data acquisition unit, a second data acquisition unit, a manual input unit and a dictionary acquisition unit; The first communication connection unit connects the information source collection module to the basic large model module and the scene large model module. The dictionary acquisition unit can acquire a data target dictionary or a data dictionary from the first type of target system and / or the second type of target system. The first data acquisition unit is used to acquire first key data from the first type of target system. The second data acquisition unit is used to acquire second key data from the second type of target system. The data in the information source collection module can be manually processed through the manual input unit.

3. The digital twin system according to claim 2, characterized in that: According to the first key data and the manual input unit, the second key data can be labeled and labeling information can be generated. According to the second key data and the labeling information, the basic large model module can generate first algorithm data. The scene large model module is communicated with the basic large model module and generates a scene algorithm model according to the first algorithm data and the second key data. The scene algorithm model can generate second target data according to the second key data.

4. The digital twin system according to claim 1, characterized in that: The first type of target systems include automated office systems, legal systems and financial systems, the second type of target systems include enterprise security systems, the first key data include first target text information, and the second key data include second target image information and second target video information.

5. The digital twin system according to claim 1, characterized in that: The basic large model module includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data calculation unit and an intelligent prediction unit.

6. A method for constructing a digital twin system, used to construct the digital twin system according to any one of claims 1 to 5, characterized in that: The following steps are involved: Establish information source collection module and connect it to the first and second target systems of the enterprise; Using the information source collection module to obtain first key data from the first type of target system, and to obtain second key data from the second type of target system; Annotate the second key data according to the first key data and generate annotation information; Constructing a basic large model module and connecting it to the information source collection module; Generate first target data and first algorithm data using the basic large model module according to the first key data, the second key data and the annotation information; Constructing a scene large model module according to the first algorithm data, the second key data and the annotation information, and using the scene large model module to generate second target data; Create a digital twin model module, and update the digital twin model in the digital twin model module according to the first target data and the second target data.

7. The method for constructing a digital twin system according to claim 6, characterized in that: Connecting the information source collection module to the first type target system and the second type target system comprises the following steps: Acquire target system information, where the target system information includes first-category system information and second-category system information; According to the first type of system information, the information source collection module is connected to the first type of target system, and the information source collection module is connected to the first type of target system having an API interface through an API interface; According to the second type of system information, the information source collection module is connected to the second type of target system, and the information source collection module is connected to the second type of target system having an API interface through an API interface.

8. The method for constructing a digital twin system according to claim 6, characterized in that: Acquiring first key data from the first-type target system includes the following steps: Performing a data target dictionary acquisition operation on the first type of target system to acquire a data target dictionary; If the data target dictionary is successfully obtained, obtaining first key data from the first type of target system according to the data target dictionary; If the data target dictionary cannot be obtained, performing a data dictionary acquisition operation on the first type of target system to acquire a data dictionary, the data dictionary including a data target dictionary; If the data dictionary is successfully obtained, obtaining first key data from the first type of target system according to the data dictionary; If the data dictionary cannot be obtained, the first key data is obtained from the first type of target system using a web crawler technology.

9. The method for constructing a digital twin system according to claim 6, characterized in that: Building a basic large model module includes the following steps: Build the model load-bearing unit; A basic model unit is built in the model carrying unit to preliminarily establish a basic large model module, wherein the basic model unit includes a language recognition processing unit, a visual analysis unit, a machine learning unit, a data calculation unit, and an intelligent prediction unit; The initially established basic large model module is trained according to the preset first sample data and the preset second sample data to construct the basic large model module.

10. The method for constructing a digital twin system according to claim 6, characterized in that: The scene big model module includes multiple scene algorithm models. After updating the digital twin model, the following steps are also included: Testing and calibrating the scene algorithm model in the scene large model module; Regenerate the second target data according to the calibrated scene algorithm model, and update the digital twin model again; Testing the updated digital twin model and generating a test result; According to the detection result, updating the first key data and / or the second key data; The digital twin model is updated again according to the updated first key data and / or the second key data.

Citation Information

Cited By

  • Large-model digital twin system

    CN119885874A

  • Data model system for ship group vertical assembling and welding operation and use method

    CN120705985A