An industrial digital twin system and method

Abstract

The present invention discloses an industrial digital twin system and method, which relates to the field of industrial technologies. The present invention includes an industrial assembly process simulation module, an industrial assembly process evaluation module, a Web integrated display terminal, and a web data warehouse. The present invention determines the component interference group label, the component interference environment group label, and the component interference execution group label, laying a foundation for subsequent simulation of the industrial assembly process. Based on the component interference group label, the component interference environment group label, and the component interference execution group label, combined with the three-dimensional real-scene data of the industrial assembly, the present invention simulates the industrial assembly process, reduces the training difficulty of the three-dimensional model, and improves the training accuracy and training efficiency of the three-dimensional model. Through the simulation data of the industrial assembly, the present invention determines the optimization type of the industrial assembly, ensures the accuracy of the optimization type of the industrial assembly, and improves the value of the industrial assembly process simulation.

Description

Technical Field

[0001] The present invention relates to the field of industrial technologies, and particularly to an industrial digital twin system and method. Background Art

[0002] The application of industrial digital twins is becoming increasingly widespread, and industrial assembly is an essential process in industrial production. The importance of the industrial assembly process is self-evident, and simulation training is gradually becoming the secret weapon for enterprises to enhance their competitiveness. With the rapid development of technology and the continuous change of market demands, the traditional assembly training methods are difficult to meet the requirements of efficient and precise production. Therefore, it is extremely necessary to combine industrial assembly with simulation training. The simulation training of the industrial assembly process can not only improve production efficiency, reduce costs, but also ensure product quality, laying a solid foundation for the sustainable development of enterprises.

[0003] The prior art, such as an invention patent application with the publication number CN114260893B, discloses a method for constructing a digital twin model for the industrial robot assembly picking and placing process, including the following steps: constructing a digital twin virtual model of the industrial robot. Specifying data interaction instructions to achieve data interaction between the digital twin virtual model of the industrial robot and the industrial robot, and completing the actual assembly work. Collecting the actions of the industrial robot under each data interaction instruction, constructing a digital twin model for the industrial robot assembly picking and placing process, and simulating the assembly process. This invention constructs a digital twin virtual model for the industrial robot assembly picking and placing process, which can be used to simulate the real industrial robot assembly process, and is helpful for high-fidelity simulation of the real industrial robot assembly picking and placing process in a virtual environment, providing support for the efficient design of the industrial robot assembly process.

[0004] The prior art, such as an invention patent application with the publication number CN115052258B, discloses an industrial monitoring system based on big data and digital twins, including a digital twin module, an Internet of Things module, and a big data processing module. The digital twin module is used to generate a three-dimensional model of the production workshop. The Internet of Things module is used to obtain the environmental data in the production workshop. The big data processing module is used to calibrate the environmental data to obtain the calibrated environmental data. The digital twin module is also used to import the calibrated environmental data into the three-dimensional model for display. When monitoring industrial production, this invention effectively improves the accuracy of using digital twin technology to monitor industrial production by calibrating the environmental data and then performing industrial monitoring based on the calibrated environmental data.

[0005] Referring to the above solution, it is found that there are still some deficiencies in the prior art, specifically reflected in: most of the prior art directly simulates the 3D model, and there is little prior production data for industrial assembly to confirm the interference data of industrial assembly. The prior production data of industrial assembly reflects the interference probability and interference reasons of components to a certain extent. The neglect of this aspect in the prior art is difficult to provide support for the subsequent optimization of industrial assembly, increases the training difficulty of the 3D model, reduces the training accuracy and training efficiency of the 3D model. At the same time, the lack of verification of the training results, and the verification of the training results can ensure the accuracy of the optimization type of industrial assembly, resulting in a reduction in the value of the simulation of the industrial assembly process. Summary of the Invention

[0006] The purpose of the present invention is to provide an industrial digital twin system and method, which solves the problems existing in the background technology.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions: In the first aspect of the present invention, an industrial digital twin system is provided, including: an industrial assembly process simulation module for simulating the industrial assembly process and generating simulation data of industrial assembly.

[0008] The industrial assembly process simulation module includes an industrial assembly real-scene acquisition unit, an industrial assembly production data acquisition unit, an industrial assembly interference data confirmation unit, and an industrial assembly simulation unit.

[0009] The industrial assembly real-scene acquisition unit is used to acquire the 3D real-scene data of industrial assembly.

[0010] The industrial assembly production data acquisition unit is used to monitor the prior production data of industrial assembly through an online monitoring platform.

[0011] The industrial assembly interference data confirmation unit is used to obtain the test volume of industrial assembly and confirm the interference data of industrial assembly based on the prior production data and 3D real-scene data of industrial assembly.

[0012] The industrial assembly simulation unit is used to simulate the industrial assembly process based on the interference data of industrial assembly and generate simulation data of industrial assembly.

[0013] An industrial assembly process evaluation module for generating an optimization type of industrial assembly based on the simulation data of industrial assembly.

[0014] A Web integrated display terminal for displaying the optimization type of industrial assembly.

[0015] In the second aspect of the present invention, a method for executing the industrial digital twin system of the present invention is provided, including: ST1. Acquire the 3D real-scene data of industrial assembly and monitor the prior production data of industrial assembly through an online monitoring platform.

[0016] ST2. Obtain the test volume of industrial assembly, and confirm the interference data of the industrial assembly based on the prior production data and three-dimensional real-scene data of the industrial assembly.

[0017] ST3. Simulate the industrial assembly process based on the interference data of the industrial assembly, and generate industrial assembly simulation data.

[0018] ST4. Generate the optimization type of the industrial assembly based on the simulation data of the industrial assembly.

[0019] ST5. Display the optimization type of the industrial assembly.

[0020] The beneficial effects of the present invention are as follows: (1) The present invention extracts the assembly data set, environment data set, and execution data set from the prior production data of industrial assembly. First, determine the interference group label of components, that is, the interference categories and interference risks that components are prone to generate. Second, determine the interference environment group label of components, that is, the example environment data set of various interference categories at each level of risk, and determine the interference execution group label of components, that is, the actual execution data set of various interference categories at each level of risk, laying a foundation for subsequent simulation of the industrial assembly process.

[0021] (2) Based on the interference group label of components, the interference environment group label of components, and the interference execution group label of components, combined with the three-dimensional real-scene data of industrial assembly, the present invention simulates the industrial assembly process, reduces the training difficulty of the three-dimensional model, and improves the training accuracy and training efficiency of the three-dimensional model.

[0022] (3) Through the simulation data of industrial assembly, the present invention determines the optimization type of industrial assembly, ensures the accuracy of the optimization type of industrial assembly, and improves the value of the simulation of the industrial assembly process. Description of the Drawings

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0024] Figure 1 It is a schematic diagram of the system structure connection of the present invention.

[0025] Figure 2 It is a schematic diagram of the implementation steps flow of the method of the present invention. Detailed Embodiments

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] Referring to Figure 1 As shown, the first aspect of the present invention provides an industrial digital twin system, including: an industrial assembly process simulation module, an industrial assembly process evaluation module, a Web integrated display terminal, and a web data warehouse.

[0028] It should be noted that the industrial assembly process simulation module is connected to the industrial assembly process evaluation module, the industrial assembly process evaluation module is connected to the Web integrated display terminal, and the web data warehouse is respectively connected to the industrial assembly process simulation module and the industrial assembly process evaluation module.

[0029] The industrial assembly process simulation module is used to simulate the industrial assembly process and generate simulation data of the industrial assembly.

[0030] The industrial assembly process simulation module includes an industrial assembly real-scene acquisition unit, an industrial assembly production data acquisition unit, an industrial assembly interference data confirmation unit, and an industrial assembly simulation unit.

[0031] The industrial assembly real-scene acquisition unit is used to acquire three-dimensional real-scene data of the industrial assembly.

[0032] It should be noted that the acquisition method of the three-dimensional real-scene data includes but is not limited to lidar scanning, oblique photography scanning, structured light scanning, etc., and is relatively mature in the prior art, so it will not be elaborated here.

[0033] The industrial assembly production data acquisition unit is used to monitor the prior production data of the industrial assembly through an online monitoring platform.

[0034] It should be noted that the online monitoring platform is equipped with a monitoring system, an environmental data acquisition system, and an end effector data acquisition system. Among them, the monitoring system is acquired by video acquisition equipment, the environmental data acquisition system includes but is not limited to environmental monitoring sensors such as infrared cameras, humidity sensors, and wind speed sensors for acquisition, and the end effector data acquisition system includes but is not limited to end effector monitoring sensors such as positioners and speed sensors for acquisition.

[0035] As a preferred solution, the prior production data includes: an assembly data set, an environmental data set, and an execution data set.

[0036] The assembly data set includes the assembly dynamic images of all components.

[0037] The environmental data set includes the assembly environment data set of all components.

[0038] The execution data set includes the end effector execution data set of all components.

[0039] The industrial assembly interference data confirmation unit is used to obtain the test volume of industrial assembly, and confirm the interference data of industrial assembly based on the prior production data and three-dimensional real scene data of industrial assembly.

[0040] It should be noted that the test volume of the industrial assembly is specifically uploaded by the staff.

[0041] The present invention extracts an assembly data set, an environmental data set, and an execution data set from the prior production data of industrial assembly. First, it determines the interference group labels of components, that is, the interference categories and interference risks that components are prone to generate. Secondly, it determines the interference environment group labels of components, that is, the example environment data sets of various interference categories at all levels of risk, and determines the interference execution group labels of components, that is, the actual execution data sets of various interference categories at all levels of risk, laying a foundation for subsequent simulation of the industrial assembly process.

[0042] As a preferred solution, the method for specifically confirming the interference data of industrial assembly is as follows: Extract the assembly data set from the prior production data of industrial assembly, obtain the assembly dynamic images of all components, and identify the interference types and interference risk characteristic parameters of several components through action recognition technology. The interference types include component interference, end effector interference, and other interference. The interference risk characteristic parameter is a value between 0 and 1, reflecting the interference degree of components. The greater the interference degree, the greater the interference risk characteristic parameter. Thus, the interference group labels of components are determined.

[0043] Specifically, the component interference is specifically the mutual interference between components, the end effector interference is specifically the mutual interference between components and the end effector, and the other interference is specifically the mutual interference between components and other equipment.

[0044] It should be noted that the method for specifically obtaining the interference risk characteristic parameter is as follows: The embedding length of several components identified through action recognition technology is obtained, and it is compared with the total length of several components to obtain the embedding ratio of several components, which is used as the interference risk characteristic parameter of several components.

[0045] Extract the assembly environment data set of all components from the prior production data of industrial assembly, and combine the interference group labels of components to determine the interference environment group labels of components.

[0046] Extract the end effector execution data set of all components from the prior production data of industrial assembly, and combine the component interference group labels to determine the component interference execution group labels.

[0047] Summarize the component interference group labels, component interference environment group labels, and component interference execution group labels to obtain the interference data of industrial assembly.

[0048] As a preferred solution, the method for determining the component interference group labels is as follows: Based on the interference types and interference risk characteristic parameters of several components, map to obtain several components corresponding to each interference type and their corresponding interference risk characteristic parameters , where is the number of each interference type, .

[0049] If , then mark the component of this interference type as a first-level risk component, is the first-level risk convergence value in the web data warehouse.

[0050] If , then mark the component of this interference type as a second-level risk component, is the second-level risk convergence value in the web data warehouse.

[0051] It should be noted that the first-level risk convergence value and the second-level risk convergence value are specifically the basis for component risk classification, which are specifically set by the staff. For example, in order to reduce the interference risk of components, the first-level risk convergence value and the second-level risk convergence value are set to 0.3 and 0.5 respectively.

[0052] If , then mark the component of this interference type as a third-level risk component.

[0053] Summarize several first-level risk components with the interference type of component interference to generate the first-level sub-label of the component first-class interference group label, summarize several second-level risk components with the interference type of component interference to generate the second-level sub-label of the component first-class interference group label, summarize several third-level risk components with the interference type of component interference to generate the third-level sub-label of the component first-class interference group label, and so on, to generate the first-level, second-level, and third-level sub-labels of the component second-class interference group label and the first-level, second-level, and third-level sub-labels of the component third-class interference group label.

[0054] Specifically, the component second-class interference group label is reflected as the interference type of end effector interference, and the component third-class interference group label is reflected as the interference type of other interference.

[0055] As a preferred solution, the method for determining the interference environment group label of components is as follows: Based on the assembly environment data set of all components, combined with the component interference group label, extract the assembly environment data set of each component of the first-level sub-label of the component's first-class interference group label. After data processing, obtain the environment judgment value of each component. The assembly environment data set includes characteristic parameters of several environment data. The environment judgment value includes numerical values of -1 and 1. When the environment judgment value is -1, it indicates that the assembly environment of the component is not suitable. Therefore, the assembly environment data set of the component can be used as an example. When the environment judgment value is 1, it indicates that the assembly environment of the component is suitable.

[0056] It should be noted that the characteristic parameters of the several environment data include but are not limited to temperature distribution maps, humidity distribution maps, wind speed distribution maps, etc. The temperature distribution map refers to the temperature distribution map within the assembly range of the component during the assembly process of the component. The same applies to the humidity distribution map and the wind speed distribution map. The specific assembly range of the component refers to the assembly range generated with the center point of the component as the origin and a set diameter. The set diameter is specifically set by the staff. For example, in order to improve the accuracy of the environmental assessment of component assembly, the set diameter is set slightly larger.

[0057] It should also be noted that the specific processing method for obtaining the environment judgment value of each component through data processing is as follows: Import the temperature distribution map of each component and the suitable temperature distribution map stored in the web data warehouse into a graphic analysis tool, and output the similarity between the temperature distribution map of each component and the suitable temperature distribution map, which is recorded as the first similarity of each component. Output the similarity between the humidity distribution map of each component and the suitable humidity distribution map, which is recorded as the second similarity of each component. Similarly, obtain several similarities of each component.

[0058] Compare the several similarities of each component with the similarity threshold in the web data warehouse. If there is a similarity less than the similarity threshold, record the environment judgment value of the component as -1. If there is no similarity less than the similarity threshold, record the environment judgment value of the component as 1.

[0059] The graphic analysis tool is specifically such as Adobe Photoshop and ImageMagick in professional image processing software, such as MATLAB and Python in scientific computing and data analysis software, and such as Pixlr and Online Image Comparer in online image processing platforms. They are relatively mature in the prior art and will not be elaborated here.

[0060] If the environmental judgment value of a certain component is -1, then the assembly environment data set of this component is used as the example environment data set, and several example environment data sets are summarized. After summarization processing, the component interference environment group label is obtained.

[0061] It should be noted that the specific processing method for obtaining the component interference environment group label through summarization processing is as follows: Summarize several example environment data sets of the first-level sub-labels of the component first-class interference group label to generate the environment label of the first-level sub-label of the component first-class interference group label, and so on, to obtain the environment labels of the second-level and third-level sub-labels of the component first-class interference group label, the environment labels of the first-level, second-level, and third-level sub-labels of the component second-class interference group label, and the environment labels of the first-level, second-level, and third-level sub-labels of the component third-class interference group label.

[0062] As a preferred solution, the method for determining the component interference execution group label is as follows: Based on the end effector execution data set of all components, combined with the component interference group label, extract the end effector execution data set of each component of the first-level sub-label of the component first-class interference group label. After data processing, the execution judgment value of each component is obtained. The end effector execution data set includes several characteristic parameters of the execution data. The execution judgment value includes values of -1 and 1. When the execution judgment value is -1, it indicates that the execution of the component is not suitable. Therefore, the end effector execution data set of the component can be used as an example. When the execution judgment value is 1, it indicates that the execution of the component is suitable.

[0063] It should be noted that the several characteristic parameters of the execution data include relevant data reflecting the execution accuracy of the end effector such as the position change diagram, attitude change diagram, and speed change diagram. The position can be monitored using a positioner, and the speed can be monitored using a speed sensor.

[0064] It should be noted that the attitude change diagram specifically refers to the change diagram of the Euler angles of the end effector. The Euler angles are used to describe the rotation direction and angle of the end effector. For the position change and speed change, the monitoring of the Euler angles can be carried out using an inertial measurement model (IMU), a vision sensor, an encoder, etc., and it belongs to the prior art and will not be specifically limited here.

[0065] It also should be noted that based on the reference characteristic parameters of the several execution data of each component in the web data warehouse, the reference characteristic parameters of the several execution data include the position change diagram, attitude change diagram, speed change diagram, etc. The end effector generally executes work according to a preset operation program, and the reference characteristic parameters of the several execution data can be obtained through the preset operation program.

[0066] It should be noted again that the processing method for obtaining the execution judgment values of each component through data processing is the same as that of the environmental judgment values of each component.

[0067] If the execution judgment value of a certain component is -1, then the end effector execution data set of this component is used as the example execution data set, and several example execution data sets are aggregated to obtain the component interference execution group label through aggregation processing.

[0068] It should be noted that the specific processing method for obtaining the component interference execution group label through aggregation processing is as follows: Aggregate several example execution data sets of the first-level sub-labels of the component's first-class interference group label to generate the execution label of the first-level sub-label of the component's first-class interference group label, and so on, to obtain the execution labels of the second-level and third-level sub-labels of the component's first-class interference group label, the execution labels of the first-level, second-level, and third-level sub-labels of the component's second-class interference group label, and the execution labels of the first-level, second-level, and third-level sub-labels of the component's third-class interference group label.

[0069] The industrial assembly simulation unit is used to simulate the industrial assembly process based on the interference data of industrial assembly and generate industrial assembly simulation data.

[0070] Based on the component interference group label, the component interference environment group label, and the component interference execution group label, combined with the three-dimensional real-scene data of industrial assembly, the present invention simulates the industrial assembly process, reduces the training difficulty of the three-dimensional model, and improves the training accuracy and training efficiency of the three-dimensional model.

[0071] As a preferred solution, the specific method for simulating the industrial assembly process is as follows: Based on the test volume of industrial assembly, count the number of components of the first-level, second-level, and third-level sub-labels of each type of component interference group label from the component interference group label, divide them by the total number of components respectively, and then multiply by the test volume of industrial assembly to obtain the training times of the first-level, second-level, and third-level sub-labels of each type of component interference group label.

[0072] Randomly extract the corresponding number of assembly environment data sets and end effector execution data sets from the corresponding environment labels and execution labels based on the training times of the first-level, second-level, and third-level sub-labels of each type of component interference group label, and use them in pairs as the training data of the first-level, second-level, and third-level sub-labels of each type of component interference group label. Based on the three-dimensional real-scene data of industrial assembly, construct an industrial assembly three-dimensional model, and combine all the components of the first-level, second-level, and third-level sub-labels of each type of component interference group label to simulate the industrial assembly process.

[0073] It should be noted that the assembly environment data set and the end effector execution data set are in a one-to-one correspondence relationship, that is, one assembly environment data set and only one end effector execution data set form a group.

[0074] It should also be noted that the current three-dimensional simulation technology has been widely used. According to the specific information in the environment data set and the end effector set, an accurate three-dimensional model can be created to realize the simulation of the assembly process, and the existing technology is relatively mature.

[0075] It should be noted again that all the components of the first-level, second-level, and third-level sub-labels of various interference group labels of the components are respectively placed in the corresponding training data to simulate the assembly of the components.

[0076] As a preferred solution, the simulation data includes the interference types of each training of the first-level, second-level, and third-level sub-labels of various interference group labels.

[0077] It should be noted that based on the interference types of all the components in each training of the first-level, second-level, and third-level sub-labels of various interference group labels, all the components of each interference type in each training are mapped, the number of components is counted, and the interference type with the largest number of components is recorded as the interference type of this training, so as to obtain the interference types of each training of the first-level, second-level, and third-level sub-labels of various interference group labels. If the number of components is tied for the most, the corresponding interference types are all recorded as the interference type of this training.

[0078] The industrial assembly process evaluation module is used to generate the optimization type of industrial assembly based on the simulation data of industrial assembly.

[0079] The present invention determines the optimization type of industrial assembly through the simulation data of industrial assembly, ensures the accuracy of the optimization type of industrial assembly, and improves the value of the simulation of the industrial assembly process.

[0080] As a preferred solution, the specific generation method of the generated optimization type of industrial assembly is: based on various interference group labels, determine the reference types of various interference group labels, compare the interference types of each training of the first-level, second-level, and third-level sub-labels of various interference group labels with the reference types, and determine the accurate values of each training , the accurate values include the numerical values of -1, 0, and 1. When the accurate value is 1, it indicates that the interference type is the same as the reference type. When the accurate value is -1, it indicates that there is no interference type. When the accurate value is 0, it indicates that the interference type is different from the reference type, where is the number of various interference group labels, , is the number of each level of sub-labels, , is the number for each training, , is an integer greater than 2.

[0081] It should be noted that the reference type of the first type of interference group label is part interference, the reference type of the second type of interference group label is end effector interference, and the reference type of the third type of interference group label is other interference.

[0082] The accuracy of interference relationship determination is obtained through numerical processing

[0083] , where and are respectively the weight influence factor of the h-level sub-label and the accurate threshold of the m-th type of interference group label in the web data warehouse.

[0084] It should be noted that the weight influence factor of each level of sub-label specifically reflects the importance of the determination accuracy of each level of sub-label. Specifically, if the interference risk of the third-level sub-label is greater than that of the second-level sub-label, then the importance of the determination accuracy of the third-level sub-label is greater than that of the second-level sub-label; if the interference risk of the determination accuracy of the second-level sub-label is greater than that of the first-level sub-label, then the determination accuracy of the second-level sub-label is greater than that of the first-level sub-label. It is specifically set by the staff. The accurate thresholds of various types of interference group labels are also specifically set by the staff as the basis for judging whether the simulation accuracy of various types of interference group labels meets the requirements.

[0085] Compare the accuracy of interference relationship determination with the interference relationship determination accuracy intervals corresponding to each optimization type in the web data warehouse, and screen out the optimization types of industrial assembly. The optimization types are environment + end effector optimization, training optimization, and production optimization.

[0086] It should be noted that the interference relationship determination accuracy intervals corresponding to each optimization type are specifically set by the staff and stored in the web data warehouse. For example, the interference relationship determination accuracy intervals corresponding to each optimization type are respectively set as (-1, -0.9), (-0.9, 0.5), (0.5, 1).

[0087] It should also be noted that the training optimization is specifically to optimize the industrial assembly process simulation model.

[0088] The Web integrated display terminal is used to display the optimization types of industrial assembly.

[0089] Refer to Figure 2As shown, the second aspect of the present invention provides a method for implementing the industrial digital twin system described in the present invention, including: ST1, collecting three-dimensional real-scene data of industrial assembly and monitoring the prior production data of industrial assembly through an online monitoring platform.

[0090] ST2, obtaining the test volume of industrial assembly and confirming the interference data of industrial assembly based on the prior production data and three-dimensional real-scene data of industrial assembly.

[0091] ST3, simulating the industrial assembly process based on the interference data of industrial assembly and generating industrial assembly simulation data.

[0092] ST4, generating the optimization type of industrial assembly based on the simulation data of industrial assembly.

[0093] ST5, displaying the optimization type of industrial assembly.

[0094] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of this technology can make various modifications, supplements, or use similar methods to replace the specific embodiments described, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should fall within the protection scope of the present invention.

Claims

1. An industrial digital twin system, characterized in that, It includes: An industrial assembly process simulation module, which is used to simulate the industrial assembly process and generate simulation data of the industrial assembly; The industrial assembly process simulation module includes an industrial assembly real-scene acquisition unit, an industrial assembly production data acquisition unit, an industrial assembly interference data confirmation unit, and an industrial assembly simulation unit; The industrial assembly real-scene acquisition unit is used to acquire three-dimensional real-scene data of the industrial assembly; The industrial assembly production data acquisition unit is used to monitor the prior production data of the industrial assembly through an online monitoring platform; The industrial assembly interference data confirmation unit is used to obtain the test volume of the industrial assembly and confirm the interference data of the industrial assembly based on the prior production data and three-dimensional real-scene data of the industrial assembly; The industrial assembly simulation unit is used to simulate the industrial assembly process based on the interference data of the industrial assembly and generate simulation data of the industrial assembly; The simulation data includes the interference types of each training of the first-level, second-level, and third-level sub-labels of various interference group labels; An industrial assembly process evaluation module, which is used to generate an optimization type of the industrial assembly based on the simulation data of the industrial assembly; The method for generating the optimization type of the industrial assembly is as follows: Based on various interference group labels, determine the reference types of various interference group labels. Compare the interference types of each training of the primary, secondary, and tertiary sub-labels of various interference group labels with the reference types, and determine the accurate values of each training through comparison. , the accurate values include the numerical values of -1, 0, and 1. When the accurate value is 1, it indicates that the interference type is the same as the reference type. When the accurate value is -1, it indicates that there is no interference type. When the accurate value is 0, it indicates that the interference type is different from the reference type, where is the number of various interference group labels, , is the number of each level of sub-labels, , is the number of each training, , is an integer greater than 2; The accuracy of determining the interference relationship is obtained through numerical processing ， Among them and are respectively the weight influence factor of the h-level sub-tag and the accurate threshold of the m-type interference group tag in the web data warehouse; Compare the accuracy of determining the interference relationship with the interference relationship determination accuracy interval corresponding to each optimization type in the web data warehouse, and screen to obtain the optimization type of the industrial assembly. The optimization types are environment + end effector optimization, training optimization, and production optimization; A Web integrated display terminal, which is used to display the optimization type of the industrial assembly.

2. An industrial digital twin system according to claim 1, characterized in that, The prior production data includes: an assembly data set, an environment data set, and an execution data set; The assembly data set includes the assembly dynamic images of all components; The environment data set includes the assembly environment data set of all components; The execution data set includes the end effector execution data set of all components.

3. An industrial digital twin system according to claim 2, characterized in that, The method for confirming the interference data of the industrial assembly is as follows: Extract the assembly data set from the prior production data of the industrial assembly, obtain the assembly dynamic images of all components, and identify the interference types and interference risk characteristic parameters of several components through action recognition technology. The interference types include component interference, end effector interference, and other interference. The interference risk characteristic parameter is a value between 0 and 1, which reflects the interference degree of the component. The greater the interference degree, the greater the interference risk characteristic parameter. Thus, the component interference group label is determined; Extract the assembly environment data set of all components from the prior production data of the industrial assembly, and combine it with the component interference group label to determine the component interference environment group label; Extract the end effector execution data set of all components from the prior production data of the industrial assembly, and combine it with the component interference group label to determine the component interference execution group label; Summarize the component interference group label, the component interference environment group label, and the component interference execution group label to obtain the interference data of the industrial assembly.

4. An industrial digital twin system according to claim 3, characterized in that, The method for determining the component interference group label is as follows: Based on the interference types and interference risk characteristic parameters of several components, several components corresponding to each interference type and their corresponding interference risk characteristic parameters are mapped , where is the number of each interference type, ; If , then mark the component of this interference type as a first-level risk component, which is the first-level risk convergence value in the web data warehouse; If , then mark the component of this interference type as a secondary risk component, which is the secondary risk convergence value in the web data warehouse; If , then mark the component of this interference type as a third-level risk component; Summarize several first-level risk components with the interference type of component interference, generate the first-level sub-tags of the component first-class interference group label, summarize several second-level risk components with the interference type of component interference, generate the second-level sub-tags of the component first-class interference group label, summarize several third-level risk components with the interference type of component interference, generate the third-level sub-tags of the component first-class interference group label, and so on, to generate the first-level, second-level, and third-level sub-tags of the component second-class interference group label and the first-level, second-level, and third-level sub-tags of the component third-class interference group label.

5. An industrial digital twin system according to claim 4, characterized in that, The method for determining the component interference environment group label is as follows: Based on the assembly environment data set of all components, combined with the component interference group label, extract the assembly environment data sets of each component of the first-level sub-tags of the component first-class interference group label, and obtain the environment judgment value of each component through data processing. The assembly environment data set includes several characteristic parameters of environmental data, and the environment judgment value includes values of -1 and 1. When the environment judgment value is -1, it indicates that the assembly environment of the component is not suitable. Therefore, the assembly environment data set of the component can be used as an example. When the environment judgment value is 1, it indicates that the assembly environment of the component is suitable; If the environment judgment value of a certain component is -1, then use the assembly environment data set of this component as the example environment data set, summarize to obtain several example environment data sets, and obtain the component interference environment group label through summary processing.

6. An industrial digital twin system according to claim 4, characterized in that, The method for determining the component interference execution group label is as follows: Based on the end effector execution data set of all components, combined with the component interference group label, extract the end effector execution data sets of each component of the first-level sub-tags of the component first-class interference group label, and obtain the execution judgment value of each component through data processing. The end effector execution data set includes several characteristic parameters of execution data, and the execution judgment value includes values of -1 and 1. When the execution judgment value is -1, it indicates that the execution of the component is not suitable. Therefore, the end effector execution data set of the component can be used as an example. When the execution judgment value is 1, it indicates that the execution of the component is suitable; If the execution judgment value of a certain component is -1, then use the end effector execution data set of this component as the example execution data set, summarize to obtain several example execution data sets, and obtain the component interference execution group label through summary processing.

7. An industrial digital twin system according to claim 1, characterized in that, The specific method for simulating the industrial assembly process is as follows: Based on the test volume of industrial assembly, count the number of components of the first-level, second-level, and third-level sub-tags of each type of component interference group label from the component interference group label, divide them by the total number of components respectively, and then multiply by the test volume of industrial assembly to obtain the training times of the first-level, second-level, and third-level sub-tags of each type of component interference group label. The number of training times for the first-level, second-level, and third-level sub-tags based on various interference group tags of components is randomly selected from the corresponding environment tags and execution tags to obtain the corresponding number of assembly environment data sets and end effector execution data sets. Each pair is used as the training data for the first-level, second-level, and third-level sub-tags of various interference group tags of components. An industrial assembly three-dimensional model is constructed based on the three-dimensional real-scene data of industrial assembly. Combining all components with the first-level, second-level, and third-level sub-tags of various interference group tags of components, the industrial assembly process is simulated.

8. A method for implementing the industrial digital twin system according to any one of claims 1-7, characterized in that, Including: ST1. Collect the three-dimensional real-scene data of industrial assembly and monitor the prior production data of industrial assembly through an online monitoring platform; ST2. Obtain the test volume of industrial assembly and confirm the interference data of industrial assembly based on the prior production data and three-dimensional real-scene data of industrial assembly; ST3. Simulate the industrial assembly process based on the interference data of industrial assembly and generate industrial assembly simulation data; ST4. Generate the optimization type of industrial assembly based on the simulation data of industrial assembly; ST5. Display the optimization type of industrial assembly.

Images