A digital twin construction method driven by geometric model and data model fusion

Abstract

The application discloses a kind of geometric model and data model fusion driven digital twin construction method.The application steps:S1 visual modeling rule, establish the Collada format file of industrial scene;2, the Collada format file is parsed and constructed using data interaction method;3, the Collada format file is parsed and constructed using automatic storage construction method;4, using the automatic binding method of geometric model and data model, the construction of digital twin body is completed, and the display of Web page is realized.The data interaction construction method of the application realizes the automatic implementation deployment of industrial scene data acquisition, control instruction issuing.The automatic storage construction method of the application avoids the need for manual design when digital twin is constructed, creates database table, realizes real-time data automatic storage.The geometric model and data model automatic binding method of the application realizes the flexible binding of model attribute and data.The application realizes the rapid development, deployment of digital twin, reduces time cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of digital twin data interaction, in particular to a geometric model and data model fusion driven digital twin construction method, belonging to the industrial field. BACKGROUND

[0002] With the promotion of intelligent manufacturing and the popularization of Internet of Things technology, digital twin technology (Digital Twin, DT) as a kind of simulation process using sensor updates, physical models, operation history and other data, integrating multi-disciplinary, multi-angle, multi-physical quantity and multi-probability, is constantly used in the industrial field and is considered as a key technology for Internet of Things. Data interaction is an important link in the industrial field, and the data interaction technology includes storage, compression, transmission, etc., which mainly embodies protocols, gateways, and is one of the main contents of the digital twin data model.

[0003] Through the establishment of virtual models and data fusion models, auxiliary geometric models, digital twins can describe the real-time behavior and real-time state of industrial scenes, thereby realizing data monitoring, scheme optimization and historical state analysis of physical objects. Among them, the data model as the basis of digital twin requires collecting and processing real-time data, storing historical data, and issuing control instructions.

[0004] However, the current digital twin technology cannot use a unified data collection method for twinning in a complex multi-device scene. At the same time, it also cannot automatically store data and needs manual database design, so it is difficult to quickly and universally establish an adaptive digital twin. In addition, in the visualization of digital twin, the current method is to use Web to display geometric models and data. However, there is a lack of a general, dynamically displayable and controllable data-driven geometric model solution, making the binding relationship between data and geometric models in many Web implementations of digital twin cases fixed and unchangeable, and unable to adapt to new data or geometric models. SUMMARY

[0005] The present application proposes a geometric model and data model fusion driven digital twin construction method to solve the following problems in the current digital twin implementation scheme: unable to use a unified data collection method for multi-source heterogeneous devices; unable to realize unified automatic data storage; unable to realize automatic binding of geometric models and data.

[0006] In order to solve the above problems in the construction of digital twin of industrial scene, the application adopts the following solutions: the OPC UA protocol is used as the basic protocol for data interaction, which has the characteristics of reliability, cross-platform, security, etc. A visual modeling rule is provided to realize visual driving, and the data interaction method and data storage scheme are recorded in the Collada geometry scene model, and the rule is used to realize the binding of the geometry model and the data model. The corresponding data acquisition scheme, data storage scheme and data binding scheme can be obtained by parsing the XML elements of the model.

[0007] In order to achieve the above purpose, the application adopts the following technical solutions:

[0008] The application provides a visual modeling rule. In order to realize visual driving, the application uses the OPC UA protocol as the basic protocol for data interaction, establishes a geometry scene model based on Blender, 3DMAX and other three-dimensional modeling tools, uses the Collada format based on the XML framework as the data format, and specifies a method for storing acquisition variables and the relationship between acquisition variables and geometry models in the geometry model, that is, a visual modeling rule. The schematic diagram is shown in Figure 2 The rule includes the following contents:

[0009] Some specific element tags are used to represent acquisition variables, and element names are used to record the information of acquisition variables, including the type, variable name, acquisition range, unit, etc. of the acquisition variable; the method of controlling the device is recorded, including the name, input parameter and parameter type corresponding to the device method; the binding relationship between the acquisition variable and the geometry model attribute is recorded, including the coordinate axis position, rotation and size; the device or sensor to which the acquisition variable belongs is recorded; whether the acquisition variable is a fixed value, the update frequency of the acquisition variable, etc. are recorded; the kinematics information of the device, the material information of the model, the light information, the camera, etc. are recorded; and the connection string of the database to be stored is recorded.

[0010] The application provides a data interaction construction method. The data interaction construction method requires that in the industrial scene, the sensor device uses the OPC UA protocol to collect data of the variables that need to be interacted, and models according to the visual modeling rule to obtain the Collada file. The method includes the following contents:

[0011] Based on the XML parser, the information of the acquisition variable (type, variable name, acquisition range, unit), the method of controlling the device (corresponding device method name, input parameter, parameter type), the device or sensor to which the acquisition variable belongs, whether the variable is a fixed value, the update frequency of the acquisition variable, the kinematics information of the device, etc. are parsed.

[0012] The parsed kinematics information is used for automatically limiting variables; based on the acquisition variables and the slave relationship of the equipment, a tree structure based on the OPC UA protocol is constructed, and the acquisition scheme of the whole scene is realized. The OPC UA server is established based on the tree structure, and is automatically started. The server accepts the client data from the sensor, and simultaneously realizes the issuing of the control instruction through the server.

[0013] The slave relationship refers to the equipment or sensor to which the acquisition variable belongs.

[0014] The application provides an automatic storage construction method. A Collada format file constructed according to a visual modeling rule is parsed, and the automatic storage construction method comprises the following contents: based on an XML parser, acquisition variable information of the Collada format file is parsed, and the acquisition variable information specifically comprises the type, variable name, update frequency, belonging equipment information and unit of the acquisition variable; database information to be stored is parsed, and a relational database is generally used as a data storage scheme; according to the parsed variable information, a corresponding table is automatically established in the database, and a GUID is used as a unique primary key, an index is automatically established, and a foreign key is automatically established; and a binding data interaction acquisition scheme is obtained, that is, whenever the OPC UA server receives data, data storage is performed to ensure that historical data will not be lost; control request data of a web front end is accepted, that is, when a user issues a control instruction, control data storage is performed.

[0015] As a preferred technical solution, the table automatically established in the database should at least comprise: an equipment table, a sensor table and a sensor data table. The equipment table should at least comprise the following fields: ID, equipment name, equipment type, equipment position and the like. The sensor table should at least comprise the following fields: ID, sensor name, equipment ID, acquisition variable type, acquisition variable upper limit, acquisition variable lower limit, unit and the like. The sensor data table should at least comprise the following fields: ID, time stamp, value, sensor ID and the like.

[0016] This invention provides an automatic binding method for geometric models and data models. The method includes the following steps: parsing a Collada format file using the Collada Loader interface to extract geometric models, cameras, lights, etc., and converting the extracted geometric models, lighting information, and camera coordinate positions. Model material information is then extracted for scene construction. Simultaneously, an OPC UA protocol client is established, connecting to an OPC UA protocol server obtained through a data interaction scheme and receiving real-time scene data from the server. A GUI interface is created to display scene data and for users to send control information. Based on the aforementioned visual modeling rules, the relationships between geometric model attributes and variables are parsed, and the geometric model and data model are bound together to achieve motion changes in the geometric model. The geometric model attributes that can be bound to the data mainly include: coordinate axis positions, rotation attributes, and size attributes. Finally, the scene is rendered using a front-end 3D renderer.

[0017] As a preferred technical solution, the Web implementation can use Three.js to display and render the 3D scene, select node-opcua to build the OPC UA client, and select dat.gui to build the user interface.

[0018] This invention provides a digital twin construction method driven by the fusion of geometric and data models, the overall flowchart of which is shown below. Figure 3 As shown, it includes the following steps:

[0019] S1. Use visual modeling rules to create Collada format files for industrial scenarios;

[0020] S2. Use data interaction methods to parse and construct Collada format files;

[0021] S3. Use the automatic storage construction method to parse and construct the Collada format file;

[0022] S4. Using an automatic binding method of geometric and data models, the digital twin is constructed and displayed on the web page.

[0023] The beneficial effects of this invention are as follows:

[0024] The visual modeling rules of this invention enable collaborative design by multiple personnel. The data interaction construction method of this invention enables automated deployment of data acquisition and control command issuance in industrial scenarios. The automatic storage construction method of this invention avoids the need for manual design and database table creation during digital twin construction, achieving real-time automated data storage. The automatic binding method of this invention between geometric and data models enables flexible binding of model attributes and data. Attached Figure Description

[0025] In order to more clearly illustrate the technical solutions in the present application, specific embodiments and specific drawings will be combined below.

[0026] Figure 1 The system architecture diagram of the present application is shown in the figure.

[0027] Figure 2 The Collada architecture diagram of the visual modeling rule in the present application is shown in the figure.

[0028] Figure 3 The overall flowchart of the digital twin construction method in the present application is shown in the figure.

[0029] Figure 4 The digital twin Web visualization interface of the embodiment is shown in the figure. DETAILED DESCRIPTION

[0030] The present application will be further analyzed below in combination with specific embodiments, which are modeling of digital twins for the scene of some small wind-driven generators of a certain wind power station. The specific implementation steps are as follows:

[0031] S1, the Collada format file of the industrial scene is established by using the visual modeling rule.

[0032] Firstly, the variables required by the scene are determined, and the hardware data acquisition scheme is determined by using the device supporting OPC UA. The collected variables for the entire scene are: wind speed (m / s). The main collected variables for each wind-driven generator are: power (kw), blade speed (rpm).

[0033] Secondly, the geometric model is established. Since the wind-driven generator is mainly concerned, the scene can be simplified. The geometric modeling of the wind-driven generator is performed by using Blender, wherein the geometric model of the wind-driven generator is composed of the power station and the blade. The appropriate light, material and camera are established.

[0034] Finally, the rule is used for supplement. The light element is used as the variable, and the name of the element is named by using the visual modeling method of the present application, including the variable name, type, range, wind-driven generator number, etc. For example, the speed (RotorSpeed) belongs to the variable name. Then, by taking this light as the child node of the corresponding wind-driven generator blade model, the binding of the variable and the device is realized. The y-axis rotation value of the light is set to 90°, which indicates that the variable binds the y-axis rotation value of the blade model. The variable of the entire scene such as the wind speed can be directly placed under the root file. The other variables of the scene are set by using the method similar to the above method. The Collada format file is saved.

[0035] S2, the Collada format file is parsed and constructed by using a data interaction method.

[0036] The Collada format file is parsed, and the tree structure of the variables in this scene, the type, range, and collection frequency of each variable are obtained. Based on this, an OPC UA server is created, and the connection address of the server is obtained after running the server. The connection address is informed to the device, and the OPC UA client of the device is connected to the address, so that the data can be uploaded to the server. The control instructions of the OPC UA client are accepted and sent to the device through the OPC UA server, and the control instructions are stored and recorded.

[0037] S3, the Collada format file is parsed and constructed by using an automatic storage construction method.

[0038] The parsed variable information and the database connection information contained in the Collada format file are mapped to the creation statements of the corresponding database tables, and then the database is connected and the statements are executed. The scene information table, device table, sensor table, and sensor data table are obtained. The device table mainly stores the information of the wind turbine, such as number, position, etc.; the sensor table stores the variable information of each wind turbine, such as device, variable type, variable value range, unit, etc.; the sensor value table stores the data information collected by each sensor, such as sensor number, timestamp, value, etc. At the same time, when starting the OPC UA server, real-time data is obtained and stored in the corresponding table in the database.

[0039] S4, the automatic binding method of geometric model and data model is used to complete the construction of digital twin, and the Web page is displayed.

[0040] Firstly, the Collada file is parsed by using the Collada Loader of Three.js, and the parsed geometric model, light, and camera are converted in coordinates to adapt to the coordinate system of Three.js, and finally added to the scene for rendering.

[0041] Secondly, a client program of node-opcua is run to connect the previous server, and the real-time data of the scene can be obtained. The obtained real-time data is displayed by using dat.gui.

[0042] Then, the visual modeling method according to the present application parses the binding relationship between the geometric model and the data, and performs data binding on the attribute of the corresponding geometric model. For example, the y-axis rotation of the blade model in Blender is bound to the RotorSpeed variable, and the parsing to Three.js is the value of the z of the rotation of the blade model bound to the RotorSpeed variable, and at the same time, the RotorSpeed variable will also be displayed on the dat.gui panel, so that the blade model in the front end can be driven to rotate by using the real-time collected RotorSpeed data.

[0043] Finally, the rendering of the scene is performed, and the digital twin scene is established, and the effect is as shown in Figure 4 (a) is the picture of the geometric scene in Blender, and (b) is the front-end scene generated by the method of the present application.

[0044] The above-described embodiments only express one embodiment of the present application, and the description is relatively specific and detailed, but it cannot be understood as a limitation on the scope of the patent of the present application. It should be understood that those skilled in the art can design many other modifications and embodiments, which will fall within the scope and spirit of the principles disclosed in the present application.

Claims

1. A method for constructing a digital twin driven by fusion of geometric models and data models, characterized in that Comprise the following steps: S1, using visual modeling rules, establish the Collada format file of industrial scene; S2, using data interaction method to parse and build the Collada format file; S3, using automatic storage construction method to parse and build the Collada format file; S4, using the automatic binding method of geometric model and data model, complete the construction of digital twin, and realize the display of Web page; The data interaction method is implemented as follows: The data interaction construction method requires that in the industrial scene, the sensor device uses the OPC UA protocol to collect the variables that need to be interacted, and models according to the visual modeling rules to obtain the Collada file, which includes the following contents: Based on XML parser, the information of collected variables is parsed, including type, variable name, collection range, unit, control device method, device or sensor to which the collected variable belongs, whether the variable is a fixed value, update frequency of the collected variable, and kinematics information of the device; The parsed kinematics information is used to automatically limit the variable; Based on the dependent relationship of the collected variables and the device, a tree structure based on the OPC UA protocol is constructed to realize the collection scheme of the whole scene; Based on the tree structure, the OPC UA server is established and automatically started; The server accepts data from the sensor client, and at the same time, the control instruction is issued through the server; The automatic storage construction method is implemented as follows: The Collada format file constructed by the visual modeling rule is parsed, and the automatic storage construction method includes the following contents: based on the XML parser, the collected variable information of the Collada format file is parsed, including the type, variable name, update frequency, device information, and unit of the collected variable; The database information to be stored is parsed, and a relational database is used as the data storage scheme; According to the parsed variable information, the corresponding table is automatically established in the database, and GUID is used as the unique primary key, the index is automatically established, and the foreign key is automatically established; At the same time, the collection scheme obtained by data interaction is bound, that is, when the OPC UA server receives data, data storage is performed to ensure that historical data is not lost; Accept the control request data of the web front end, that is, when the user issues a control instruction, the control data is stored.

2. The geometric model and data model fusion driven digital twin construction method according to claim 1, characterized in that The visual modeling rule is implemented as follows: OPC UA protocol is used as the basic protocol for data interaction, some specific element tags are used to represent collected variables, and element names are used to record the information of collected variables, including the type, variable name, collection range, and unit of the collected variable; The method of controlling the device is recorded, including the name, input parameter, and parameter type of the corresponding device method; The binding relationship between the collected variables and the geometric model attributes is recorded, including the coordinate axis position, rotation, and size of the geometric model attributes; Record the device or sensor to which the acquisition variable belongs; record whether the acquisition variable is a fixed value, the update frequency of the acquisition variable; record the kinematics information of the device, the model material information, the light information, the camera; record the connection string of the database to be stored.

3. The geometric model and data model fusion driven digital twin construction method according to claim 2, characterized in that The automatically established table in the database should at least include: a device table, a sensor table, and a sensor data table; wherein the device table should at least include the following fields: ID, device name, device type, device location; wherein the sensor table should at least include the following fields: ID, sensor name, device ID, acquisition variable type, upper limit of acquisition variable, lower limit of acquisition variable, unit; wherein the sensor data table should at least include the following fields: ID, timestamp, value, sensor ID.

4. The geometric model and data model fusion driven digital twin construction method according to claim 3, characterized in that The automatic binding method of the geometric model and the data model includes the following contents: The Collada format file is parsed through the Collada Loader interface to parse out the geometric model, camera, and light, and to convert the coordinate positions of the parsed geometric model, light information, and camera; the model material information is parsed out for constructing a scene; at the same time, an OPC UA protocol client is established, which connects the OPC UA protocol server obtained by the data interaction scheme and accepts the real-time scene data of the server; a GUI interface is established for displaying scene data and for user to issue control information; based on the above-mentioned visual modeling rules, the relationship between the geometric model attributes and variables is parsed out to bind the geometric model and the data model, and to realize the motion change of the geometric model, wherein the geometric model attributes that can be bound with data mainly include: coordinate axis position, rotation attribute, and size attribute.

Images