CN115826937B - Digital twin industrial software platform makeTain - Google Patents
Digital twin industrial software platform makeTain Download PDFInfo
- Publication number
- CN115826937B CN115826937B CN202310139311.5A CN202310139311A CN115826937B CN 115826937 B CN115826937 B CN 115826937B CN 202310139311 A CN202310139311 A CN 202310139311A CN 115826937 B CN115826937 B CN 115826937B
- Authority
- CN
- China
- Prior art keywords
- module
- model
- algorithm
- data
- digital twin
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Stored Programmes (AREA)
Abstract
The invention discloses a digital twin industrial software platform makeTain, which comprises a twin model generator module, wherein the quick creation and configuration of a digital twin model are supported; the communication connector module supports creation and configuration of a physical entity communication protocol; the virtual-real interaction configurator module supports virtual-real space synchronization and interaction; the twin data processor module supports a user to configure, mine and analyze data in the database; the algorithm solver module supports a user to create, modify and configure an algorithm according to specific problem solving requirements; and the service template provider module supports the rapid development of digital twin service function development of users. The invention can reduce the period of implementing and deploying digital twin projects for enterprise users and developers, and provides a public test platform for digital twin academic researchers.
Description
Technical Field
The invention belongs to the field of electric digital data processing, and particularly relates to a digital twin industrial software platform makeTain.
Background
With the gradual penetration of theoretical research and practical exploration of digital twinning, digital twinning is widely accepted and tracked, and the U.S. and the United kingdom, french and Germany respectively form the national organizations of digital twinning alliance, national digital twinning plan, digital twinning city plan and industrial digital twinning association, aiming at enlarging the breakthrough of digital twinning theory and key technology, standard formulation and practical floor application. However, research and analysis of the current digital twin research and exploration state, found that:
(1) In the field of academic research, digital twin academic research is developed in world famous universities such as Stanford university, cambridge university and oxford university, and an experimental verification system is built by taking traditional modeling simulation software as a basis. As a college, it is more focused on the innovation of digital twin theory, but generally lacks the systematic tool support platform required for academic research;
(2) In the field of enterprise application, according to statistics, the current digital twinning has been applied and explored in more than sixty fields of smart logistics, intelligent production, smart cities and the like, however, through analysis, almost all applications are developed in a customized way, and the support of a universal digital twinning development platform is lacking;
in summary, the current theoretical research innovation of digital twinning lacks a research support platform of a system, and the industrial landing application of digital twinning lacks a general software development platform. Therefore, the invention proposes and designs a digital twin industrial software platform makeTain to quench the thirst of academic innovation and relieve the pain of industrial transformation and the development of China.
Disclosure of Invention
Aiming at the difficult problems of lack of universal digital twin industrial software support in the current digital twin academic research and floor application, the invention provides and designs a digital twin industrial software platform makeTain, provides uniform system tool support for digital twin theoretical researchers, and provides a universal software platform for quick development and deployment for enterprise application, thereby reducing the digital twin theoretical test period and the project development and deployment period.
The invention solves the technical problems by adopting the following technical scheme:
a digital twin industrial software platform makeTwin comprising: the system comprises a twin model generator module, a communication connector module, a virtual-real interaction configurator module, a twin data processor module, an algorithm solver module and a service template provider module;
the twin model generator module supports rapid creation and configuration of a digital twin model;
the communication connector module supports creation and configuration of a physical entity communication protocol;
the virtual-real interaction configurator module supports virtual-real space synchronization and interaction;
the twin data processor module supports a user to configure, mine and analyze data in the database;
the algorithm solver module supports a user to create, modify and configure an algorithm according to specific problem solving requirements;
the service template provider module supports a user to rapidly develop digital twin service function development.
Further, the twin model generator module comprises a model library, a model creation and parameter configuration module, a model calling module, a model assembling module, a model fusion module, a model verification and correction module and a model visualization module; the model library comprises digital twin models of various industrial general equipment, and has an extensible function, namely, after a user obtains rights, the digital twin models in the model library can be checked, modified and deleted, or user-defined digital twin models can be added; the model creation and parameter configuration module is used for creating a digital twin model and setting parameters of the digital twin model; the model calling module calls a digital twin model from a model library and sets related parameters; the model assembly module is used for assembling different digital twin models into a complex system according to a spatial relationship, a time sequence relationship, a mapping relationship and a logic relationship; the model fusion module fuses models of different subjects in different fields; the model verification and correction module verifies the accuracy of the constructed digital twin model and supports modification and correction of the digital twin model; the model visualization module supports three-dimensional visual rendering of the digital twin model.
Further, the communication connector module comprises a protocol library, a communication protocol creation module, a communication protocol calling module and a communication protocol appointment configuration module; the protocol library comprises a plurality of industrial communication protocols, and has the extensible functions, including the functions of adding, viewing, deleting and modifying the communication protocols; the communication protocol creation module is used for creating a communication protocol which does not exist in the protocol library and storing the communication protocol into the protocol library; the communication protocol calling module supports calling a communication protocol from a protocol library; the communication protocol contract configuration module supports configuration of a data frame contract mode in a protocol.
Further, the virtual-real interaction configurator module comprises a model action configuration module, a virtual-real mapping and synchronizing module and a virtual-real iteration correction module; the model action configuration module supports a user to rapidly configure the association mapping relation between actions, states and performances of the digital twin model and data from actual physical objects in a parameterized and graphical mode; the virtual-real mapping and synchronizing module is used for synchronously updating the actual running state of the physical entity to the digital twin model in real time, and a control instruction generated by the digital twin model is issued to the physical entity in real time; the virtual-real iteration correction module is used for enabling the digital twin model to have self-updating and self-correcting capabilities according to the working condition change of the physical entity.
Further, the twin data processor module comprises a database, a data generation module, a data calling module, a data feature extraction module, a data association module and a data association module; the database is used for storing design data, process data, performance data, task data, fault data and maintenance data, and has an extensible function and is used for adding, checking, deleting and modifying all data in the database; the data generation module supports simulation calculation based on the digital twin model established by the twin model generator module, so that data which are difficult to acquire in practice are generated; the data calling module retrieves and extracts required data from the database according to the category, the scale, the time period and the characteristics; the data feature extraction module supports feature extraction of selected data; the data association module analyzes and calculates association degrees of the features of different data; and the data dimension reduction module is used for reducing the dimension of the data based on the analysis and calculation of the data feature extraction and the association degree.
Further, the algorithm solver module comprises an algorithm library, an algorithm creation module, an algorithm calling module, an algorithm modification module and an algorithm parameter configuration module; the algorithm library is used for storing a plurality of typical algorithms required for solving different problems, and has an extensible function and is used for adding, viewing, deleting and modifying the algorithms in the library; the algorithm creation module supports adding user-defined algorithms, configures algorithm related parameters and stores the algorithm related parameters into an algorithm library; the algorithm calling module supports the algorithm calling from an algorithm library; the algorithm modification module supports adaptive modification of an algorithm called from an algorithm library; the algorithm parameter configuration module supports adjustment and setting of parameters related to an algorithm.
Further, the service template provider module provides a design optimization class template, a scheduling optimization class template, a fault diagnosis class template, a quality analysis class template, a prediction analysis class template and a reproduction tracing class template, and a user invokes the service template provided in the software platform and modifies the service template, so that secondary development of specific problems is rapidly realized.
Further, the twin model generator module adopts a Spring development architecture, wherein a model library of the twin model generator module adopts a Spring container, an XML language is used for describing a digital twin model of an object, and a Java language is used for realizing internal logic of a software module; the object definition descriptions in the model library are
Where Obj represents different objects, i represents the number of objects in the model base, expressed as +.>
Wherein Name represents the Name of the object, type represents the class of the object, VM represents the digitized model of the object, pfm represents the performance of the object, para represents the parameter information of the object, port represents the interface of the object, cus represents the user-defined information; wherein VM is represented as
Wherein->
Geometric appearance model representing an object, +.>
A physical mechanism model representing an object, < >>
Behavior model representing object->
A rule model representing an object;
the model assembly module is based on
The implementation of the assembly association relation of (1), SR represents a spatial relation, TR represents a time sequence relation, MR represents a mapping relation, LR represents a logic relation, and the assembly association relation is based on the instruction of the assembly association relation, so that the assembly association relation from a part level, a component level to an equipment level is realized, and then the assembly association relation is assembled into a unit level, a system level and a complex system level;
the model fusion module firstly establishes different attributes of different models and different physical field attributes including a temperature field, a flow field and vibration; secondly, abstracting different attributes of different models into abstract models with only Input and Output; coupling incidence relations among different models of different attributes are mutually interconnected; finally, the coupled model is converted into an executable model.
Further, the communication connector module adopts a Spring development architecture, wherein a protocol library adopts a Spring container; supporting a user to write a protocol which is temporarily not existed in a protocol library, and storing the protocol into the protocol library; the communication connector module integrates C, C #, C++, python and VB language compatible environments, and is compatible with C, C #, C++, python and VB language-written communication protocol function libraries; the communication connector module supports a user to call an existing communication protocol from a protocol library and configures specific conventions of the communication protocol, and comprises the following steps: (1) field length configuration including frame start length
Arbitration field Length->
Control field Length->
Data field Length->
CRC field Length->
ACK field length +.>
End of frame Length +.>
The method comprises the steps of carrying out a first treatment on the surface of the (2) Bit meaning configuration including frame start bit->
Wherein->
Arbitration field bit->
Wherein, the method comprises the steps of, wherein,
control field bit->
Wherein->
Data field bit->
Wherein
CRC field bits->
Wherein->
ACK field bit->
Wherein, the method comprises the steps of, wherein,
end of frame bit->
Wherein->
。
Further, the virtual-real interaction configurator module comprises the following functions:
(1) the method supports the rapid configuration of the actions of a user on the digital twin model, and for a physical entity, all the actions are defined as follows:
wherein, AName represents the set of action names, AJNum represents the number of action joints, ajrelay represents the joint cascade relation sequence, AStart represents the set of starting positions of each joint action, AEnd represents the set of ending positions of each joint action;
wherein the method comprises the steps of
Spatial coordinates representing the starting position of each action, +.>
Spatial coordinates representing the end position of each action, a=1, 2, …, AJNum;
(2) establishing an association mapping relation between the start and stop of each action of the digital twin model and real-time data from a physical entity, and driving the change of the state of the digital twin model based on the acquired real-time data when the software platform runs;
(3) in order to ensure that the digital twin model and the physical entity still keep high consistency after long-time operation, the self-adaptive control algorithm referenced by the model is used for reference, and the algorithm is used forThe "reference model" is replaced by a digital twin model and the output of the digital twin model is used
And performing differential processing with the output y of the physical entity, and then adjusting the parameters of the controller based on the self-adaptive mechanism, so as to realize the self-adaptive correction of the virtual-real consistency.
Further, the twin data processor module includes the following functions:
(1) the digital twin model constructed by the twin model generator module is supported to be input with parameters under different working conditions by a user, and simulation calculation is carried out, so that state data which cannot be monitored, characteristic data with small sample size and process data with fuzzy rules are obtained;
(2) the database contains design data, performance data, process data, task data, fault data and maintenance data, and supports user addition, modification, deletion and calling of data;
(3) supporting a user to carry out extraction of data characteristics, and extracting frequency domain characteristics and time domain characteristics of actual data and generated data respectively;
(4) supporting analysis of relevance of different data, including numerical accuracy
Feature relevance->
Trend fitness ++>
；
(5) Support the user to reduce the dimension of a large number of data with different characteristics, and analyze the error between the reduced dimension data and the true value, including the mean value
Maximum Max, mean square error S, peak factor PAR, skewness factor SF, and kurtosis factor KF.
Further, the algorithm solver module comprises a data preprocessing class algorithm, a correlation analysis class algorithm, a time sequence class algorithm, a scheduling optimization class algorithm, a neural network class algorithm and a statistical calculation and evaluation class algorithm; the surrounding algorithm is used and managed, and has the following functions:
(2) supporting a user to add a new algorithm written by the user and adding the new algorithm into an algorithm library;
(2) supporting a user to call an existing algorithm from an algorithm library, modifying codes of the algorithm, and debugging the modified algorithm;
(3) while modifying the algorithm, the user modifies the parameters of the algorithm and tests and compares different parameters.
Further, the service template provider module adopts a Spring architecture to package six types of design optimization type templates, scheduling optimization type templates, fault diagnosis type templates, quality analysis type templates, prediction analysis type templates and reproduction tracing type templates into a Spring container, and after selecting a template, a user selects any type of template, the service template provider module comprises the following functions:
(1) the user carries out mathematical modeling on the problem to be solved in an integrated development environment in a mode of creating and calling from a model library, and modifies the model code;
(2) aiming at the actual problem demand, searching and extracting the required data, and preprocessing;
(3) constructing an algorithm required by problem solving in a mode of creating a new structure in an integrated development environment and calling the new structure from an algorithm library, and modifying codes and parameters of the algorithm;
(4) the user selects information to be checked, including an algorithm running process, an algorithm convergence process, a simulation calculation result and different algorithm comparison; the service template provider module adopts visualization tools to provide different types of visualization effects for users.
The invention has the advantages compared with the prior digital twinning-related software tool that:
(1) The invention designs a digital twin industrial software platform with six functions, namely a digital twin model generator, a communication connector, a virtual-real interaction configurator, a twin data processor, an algorithm solver and a service template provider, compared with the current foreign mainstream digital twin tool which focuses on finite element analysis or different Internet of things, the digital twin industrial software platform makeTain has more comprehensive functions;
(2) The six functional modules designed by the invention are realized by adopting a configurable and callable concept, when a user uses the software platform, the creation, the calling and the configuration of a digital twin model are supported, the creation, the calling and the appointed configuration of a communication protocol are carried out, the low-code rapid construction of virtual-real interaction actions is carried out, the retrieval, the extraction and the pretreatment of data in a database are carried out, the creation, the calling and the modification of a solving algorithm are carried out, and a service template suitable for the user requirement is selected for rapid development and use;
(3) Compared with the prior digital twin industrial software tool which only faces one type of user, the digital twin industrial software platform makeTain-oriented user comprises enterprise users, project developers and academic researchers, and based on the digital twin industrial software platform makeTain-oriented user, the enterprise users can quickly build a digital twin management and control system by adopting a low-code development mode; the project developer can multiplex architecture, model, data, algorithm and the like in the platform to rapidly provide services; the academic researcher can quickly build a test scene to verify related theories.
Drawings
FIG. 1 is a block diagram of the overall functional architecture of a digital twin industrial software platform makeTain;
FIG. 2 is a diagram of the functionality and implementation of a twin model creator module;
FIG. 3 is a diagram of a communication connector (TwainConnector) module function and implementation method;
FIG. 4 is a diagram of a virtual-to-real interaction configurator (TwainInteractor) module function and implementation method;
FIG. 5 is a diagram of a twin data processor (TwainDataProcessor) module function and implementation method;
FIG. 6 is a diagram of the algorithm solver (TwailGSolver) module function and implementation;
FIG. 7 is a diagram of the function and implementation of a service template provider (TwoServerTemplite) module.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The invention relates to a digital twin industrial software platform makeTain, which aims at a research support platform lacking a system for the theoretical research innovation of the current digital twin, and the industrial floor application of the digital twin lacks a general software development platform, adopts a Spring development architecture, adopts a model library, a protocol library, an algorithm library and a service template to be packaged by a Spring container, adopts MySQL to build a database, adopts Python integrated development environment for algorithm construction and configuration, adopts Java language to realize the internal logic of a software main body and a module, and is compatible with a plurality of languages such as C, C #, C++, and the like, thereby providing a configurable and callable digital twin testing, developing and deploying tool for users.
As shown in fig. 1, the digital twin industrial software platform makeTwin of the present invention comprises:
(1) The twin model generator (twinModelCreater) module 101 in FIG. 1 includes:
1) The model library comprises a plurality of digital twin models of industrial general equipment, such as a machine tool, a robot and an AGV, and also comprises digital twin models of standard components commonly used in industry, such as a standard model bearing, a standard model gear, a standard model motor and the like, and has an extensible function, namely after a user obtains rights, the digital twin models in the model library can be checked, modified and deleted, and the user-defined digital twin models can be added;
2) The model creation and parameter configuration module is used for creating a digital twin model and setting parameters of the digital twin model;
3) The model calling module can call a digital twin model from a model library and set related parameters;
4) The model assembly module is used for assembling different digital twin models into a complex system according to a certain spatial relationship, a certain time sequence relationship, a certain mapping relationship and a certain logic relationship;
5) The model fusion module fuses models of different disciplines in different fields, such as the fields of machinery, electricity, hydraulic pressure, control and the like, and discipline mechanisms of temperature field, flow field, stress and the like;
6) The model verification and correction module is used for verifying the accuracy of the constructed digital twin model and supporting the modification and correction of the digital twin model;
7) And the model visualization module supports three-dimensional visual rendering of the digital twin model.
Specifically, the function and implementation of the twin model generator (twinModelcreator) module 101 is shown in FIG. 2. The development adopts a Spring development architecture, wherein a Spring container is adopted in a model library, an XML language is used for describing a digital twin model of an object, and a Java language is used for realizing internal logic of a software module.
(1) The object definition descriptions in the model library are
Wherein
The different objects are represented by a number of different objects,
represented as
Wherein Name represents the Name of an object, and comprises a C52 double-column vertical lathe, a six-axis rotary spraying robot and the like; type represents the class of objects, including machine tools, robots, AGVs, etc.; VM represents a digitized model of an object; pfm the performance of the object, including accuracy capability, operating speed, energy consumption, etc.; para represents parameter information of an object, including parameters required to be set for running, control instructions required for running and the like; port represents the interface of the object, including Input, output, etc.; cus represents user-defined information. Wherein VM is represented as
. Wherein->
Geometric appearance model representing an object, +.>
A physical mechanism model representing an object, < >>
Behavior model representing object->
A rule model representing an object.
(2) Model assembly refers to the combination of different models on different time-space scales according to
Specifically, SR represents a spatial relationship, TR represents a temporal relationship, MR represents a mapping relationship, LR represents a logical relationship, and based on the instruction of the assembly relationship, the assembly from the part level, the component level to the equipment level is realized, and the assembly is performed into a unit level, a system level and a complex system level.
(3) Model fusion refers to the coupling of different models on different disciplines and different physical fields, firstly, the mechanical, electrical and hydraulic properties of the different models and the control of the different discipline properties, as well as the temperature field, the flow field, the vibration and other different physical field properties are established; secondly, abstracting different attributes of the model into an abstract model with only Input and Output; coupling incidence relations among different attribute models are mutually interconnected; finally, the coupled model is converted into an executable model.
(2) The communication connector (TwinConnector) module 102 in fig. 1 includes:
1) The protocol library comprises a plurality of typical industrial communication protocols, such as Profinet, FOCAS, modeBus-TCP, CAN and the like, and also has an extensible function, namely, the communication protocols CAN be added, checked, deleted and modified;
2) The communication protocol creation module is used for creating a new communication protocol which does not exist in the protocol library and can be stored in the protocol library;
3) The communication protocol calling module supports calling a communication protocol from a protocol library;
4) The communication protocol appointing configuration module supports the configuration of the data frame appointing mode in the protocol no matter the user creates or calls the communication protocol from the protocol library.
Specifically, as shown in fig. 3, the function and implementation method of the communication connector (TwinConnector) module 102 are shown, and the development of the communication connector adopts a Spring development architecture, wherein a Spring container is adopted in a protocol library. The user is supported to write the protocols which are not existed in the protocol library, such as Profinet, modBus, CAN, FOCAS, and the like, and the protocols are stored in the protocol library. The TwainConnector module integrates the language compatible environments such as C, C #, C++, python and VB, and can be compatible with a communication protocol function library written by various types of computer programming languages. In addition, the TwainConnector module supports a user to call an existing communication protocol from a protocol library and configures specific conventions of the communication protocol:
(1) field length configuration including frame start length
Arbitration field Length->
Control field Length->
Data field Length->
CRC field Length->
ACK field length +.>
End of frame Length +.>
；
(2) Bit meaning configuration including frame start bit
(/>
) Arbitration field bit->
(
) Control field bit->
(/>
) Data field bit->
(
) CRC field bits->
(/>
) ACK field bit->
(
) End of frame bit->
(/>
)
(3) The virtual-real interaction configurator (twin configurator) module 103 in fig. 1 includes:
1) The model action configuration module supports a user to rapidly configure the association mapping relation between actions, states and performances of the digital twin model and data from actual physical objects in a parameterized and graphical mode;
2) The virtual-real mapping and synchronizing module is used for synchronously updating the actual running state of the physical entity to a digital twin model in real time, and a control instruction generated by the digital twin model is issued to the physical entity in real time;
3) And the virtual-real iteration correction module is used for enabling the digital twin model to have self-updating and self-correction capabilities according to the working condition change of the physical entity.
Specifically, fig. 4 shows the functions and implementation method of the virtual-real interaction configurator (twin configurator) module 103. To enable interaction between the digital twin model and the physical entity:
(1) the method supports the rapid configuration of the actions of a user on a digital twin model, and takes a physical entity as an example, all the actions can be defined as follows:
wherein AName represents a set of action names, AJNU represents the number of action joints, AJRela represents a sequence of joint cascade relationships, AStart represents a set of starting positions of each joint action, end represents a set of ending positions of each joint action,
wherein the method comprises the steps of
(a=1, 2, …, AJNum) represents the spatial coordinates of the starting position of each motion, +.>
Representing the spatial coordinates of the termination location of each action.
(2) And establishing an association mapping relation between the start and stop of each action of the digital twin model and real-time data from a physical entity, and driving the change of the state of the digital twin model based on the acquired real-time data when the system operates.
(3) In order to ensure that the digital twin model and the physical entity can still keep high consistency after long-time operation, the self-adaptive control algorithm (MRAC) referenced by the model is used, the reference model in the algorithm is replaced by the digital twin model, and the output of the digital twin model is output
And performing differential processing with the output y of the physical entity, and then adjusting the parameters of the controller based on the self-adaptive mechanism, so as to realize the self-adaptive correction of the virtual-real consistency.
(4) The twin data processor (TwainDataProcessor) module 104 in FIG. 1 includes:
1) The database is used for storing design data, process data, performance data, task data, fault data, maintenance data and the like, and has an extensible function, namely, all data in the database can be added, checked, deleted and modified;
2) The data generation module supports simulation calculation of the constructed digital twin model, so that data which is difficult to acquire in practice is generated;
3) The data calling module can retrieve and extract required data from the database according to the category, the scale, the time period and the characteristics;
4) The data feature extraction module supports feature extraction of selected data according to a certain rule;
5) The data association module can analyze and calculate the association degree of the features of different data according to the definition of the specific association degree;
6) The data dimension reduction module can reduce dimension of the data according to a certain rule based on data feature extraction and association degree calculation.
Specifically, as shown in fig. 5, the functions and implementation method of the twin data processor (twin data processor) module 104 include:
(1) the method comprises the steps of supporting a user to input parameters under different working conditions based on a constructed digital twin model and carrying out simulation calculation, so as to obtain three types of data, namely state data which cannot be monitored, characteristic data with small sample size and process data with fuzzy rules;
(2) the database adopts a MySQL relational database, and comprises design data, performance data, process data, task data, fault data and maintenance data, so that the user can be supported to add, modify, delete and call the data;
(3) the method supports the user to carry out the extraction of the data characteristics, and can extract the frequency domain characteristics and the time domain characteristics of the actual data and the generated data respectively;
(4) supporting analysis of relevance of different data, including numerical accuracy
Feature relevance->
Trend fitness ++>
；
(5) The method supports the user to reduce the dimension of a large number of data with different characteristics, and can analyze the error between the dimension reduced data and the true value, including the mean value
Maximum Max, mean square error S, peak factor PAR, skewness factor SF, and kurtosis factor KF.
(5) The algorithm solver (TwailGSolver) module 105 in FIG. 1 includes:
1) The algorithm library is used for storing various typical algorithms required for solving different problems, such as algorithms of data preprocessing class, association analysis class, time sequence class, scheduling optimization class, neural network class, statistical calculation and evaluation class and the like, and has an extensible function, namely, the algorithms in the library can be added, checked, deleted and modified;
2) The algorithm creation module supports adding user-defined algorithms, configures related parameters of the algorithms and can be stored in an algorithm library;
3) The algorithm calling module supports calling the existing algorithm from the algorithm library;
4) The algorithm modification module supports adaptive modification of an algorithm called from an algorithm library;
5) And the algorithm parameter configuration module supports adjustment and setting of various parameters related to the algorithm.
Specifically, as shown in fig. 6, the function and implementation method of the algorithm solver (TwinAlgSolver) module 105 include an algorithm library based on a Python development environment, and may be classified into six types of algorithms, namely, a data preprocessing algorithm, an association analysis algorithm, a time sequence algorithm, a scheduling optimization algorithm, a neural network algorithm, and a statistical calculation and evaluation algorithm according to different algorithm types.
The surrounding algorithm is used and managed, and has three functions:
(1) supporting a user to add a new algorithm written by the user and adding the new algorithm into an algorithm library;
(2) supporting a user to call an existing algorithm from an algorithm library, opening a Liclipse Python integrated development environment, modifying codes of the algorithm in the development environment by the user, and debugging the modified algorithm;
(3) while modifying the algorithm, the user can modify the parameters of the algorithm in the Liclipse Python development environment, and can test and compare different parameters.
(6) In fig. 1, a service template provider (twin service template) module 106, a user rapidly develops development of a digital twin service function, provides a design optimization class template, a scheduling optimization class template, a fault diagnosis class template, a quality analysis class template, a prediction analysis class template, a reproduction trace back class template, and the like, and can call a service template provided in a software platform and adaptively modify the service template, thereby rapidly realizing secondary development of a specific problem.
Specifically, the function and implementation method of the service template provider (TwainServerTemplite) module 106 are shown in FIG. 7. The method adopts a Spring framework, uses a UDDI engine and a Web service semantic description engine to provide services, packages six types of design optimization type templates, scheduling optimization type templates, fault diagnosis type templates, quality analysis type templates, prediction analysis type templates and reproduction traceability type templates into a Spring container, and enables a user to select any type of templates from the six types of templates, wherein the steps of selecting the templates include:
(1) the user can carry out mathematical modeling on the problem to be solved in the integrated development environment of Python in a mode of creating and calling from a model library, and modify the model code;
(2) aiming at the actual problem demand, retrieving and extracting required data from a MySQL database, and preprocessing;
(3) creating an algorithm required by problem solving in an integrated Python development environment in a mode of creating and calling from an algorithm library, and modifying codes and parameters of the algorithm;
(4) the user can select information to be checked, such as an algorithm running process, an algorithm convergence process, a simulation calculation result, different algorithm comparison and the like, and the TwoServiceTemplite module adopts Echart visualization tools to provide different types of visualization effects for the user, including the algorithm running process and visualization, simulation result analysis and visualization, result comparison analysis and visualization and the like.
In summary, the invention discloses a digital twin industrial software platform makeTlin, which comprises six functions of TwoModelcreator (twin model generator), twoDataProcessor (twin data processor), twoAlgSolver (algorithm solver), twoInteractor (virtual-real interaction configurator), twoConnector (communication connector) and TwoServiceTemplite (service template provider). The invention can reduce the period of implementing and deploying digital twin projects for enterprise users and developers, and provides a public test platform for digital twin academic researchers.
What is not described in detail in the present specification belongs to the prior art known to those skilled in the art.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (7)
1. A digital twin industrial software platform makeTwin comprising: the system comprises a twin model generator module, a communication connector module, a virtual-real interaction configurator module, a twin data processor module, an algorithm solver module and a service template provider module;
the twin model generator module supports rapid creation and configuration of a digital twin model;
the communication connector module supports creation and configuration of a physical entity communication protocol;
the virtual-real interaction configurator module supports virtual-real space synchronization and interaction;
the twin data processor module supports a user to configure, mine and analyze data in the database;
the algorithm solver module supports a user to create, modify and configure an algorithm according to specific problem solving requirements;
the service template provider module supports a user to rapidly develop digital twin service function development;
the twin model generator module comprises a model library, a model creation and parameter configuration module, a model calling module, a model assembling module, a model fusion module, a model verification and correction module and a model visualization module; the model library comprises digital twin models of various industrial general equipment, and has an extensible function, namely, after a user obtains rights, the digital twin models in the model library can be checked, modified and deleted, or user-defined digital twin models can be added; the model creation and parameter configuration module is used for creating a digital twin model and setting parameters of the digital twin model; the model calling module calls a digital twin model from a model library and sets related parameters; the model assembly module is used for assembling different digital twin models into a complex system according to a spatial relationship, a time sequence relationship, a mapping relationship and a logic relationship; the model fusion module fuses models of different subjects in different fields; the model verification and correction module verifies the accuracy of the constructed digital twin model and supports modification and correction of the digital twin model; the model visualization module supports three-dimensional visual rendering of the digital twin model;
the communication connector module comprises a protocol library, a communication protocol creation module, a communication protocol calling module and a communication protocol appointment configuration module; the protocol library comprises a plurality of industrial communication protocols, and has the extensible functions, including the functions of adding, viewing, deleting and modifying the communication protocols; the communication protocol creation module is used for creating a communication protocol which does not exist in the protocol library and storing the communication protocol into the protocol library; the communication protocol calling module supports calling a communication protocol from a protocol library; the communication protocol appointment configuration module supports configuration of a data frame appointment mode in a protocol;
the virtual-real interaction configurator module comprises a model action configuration module, a virtual-real mapping and synchronizing module and a virtual-real iteration correction module; the model action configuration module supports a user to rapidly configure the association mapping relation between actions, states and performances of the digital twin model and data from actual physical objects in a parameterized and graphical mode; the virtual-real mapping and synchronizing module is used for synchronously updating the actual running state of the physical entity to the digital twin model in real time, and a control instruction generated by the digital twin model is issued to the physical entity in real time; the virtual-real iteration correction module is used for enabling the digital twin model to have self-updating and self-correcting capabilities according to the working condition change of the physical entity;
the twin data processor module comprises a database, a data generation module, a data calling module, a data feature extraction module, a data association module and a data dimension reduction module; the database is used for storing design data, process data, performance data, task data, fault data and maintenance data, and has an extensible function and is used for adding, checking, deleting and modifying all data in the database; the data generation module supports simulation calculation based on the digital twin model established by the twin model generator module, so that data which are difficult to acquire in practice are generated; the data calling module retrieves and extracts required data from the database according to the category, the scale, the time period and the characteristics; the data feature extraction module supports feature extraction of selected data; the data association module analyzes and calculates association degrees of the features of different data; the data dimension reduction module is used for reducing dimension of the data based on data feature extraction and correlation analysis calculation;
the algorithm solver module comprises an algorithm library, an algorithm creation module, an algorithm calling module, an algorithm modification module and an algorithm parameter configuration module; the algorithm library is used for storing a plurality of typical algorithms required for solving different problems, and has an extensible function and is used for adding, viewing, deleting and modifying the algorithms in the library; the algorithm creation module supports adding user-defined algorithms, configures algorithm related parameters and stores the algorithm related parameters into an algorithm library; the algorithm calling module supports the algorithm calling from an algorithm library; the algorithm modification module supports adaptive modification of an algorithm called from an algorithm library; the algorithm parameter configuration module supports adjustment and setting of parameters related to an algorithm;
the service template provider module provides a design optimization class template, a scheduling optimization class template, a fault diagnosis class template, a quality analysis class template, a prediction analysis class template and a reproduction tracing class template, and a user invokes the service template provided in the software platform and modifies the service template, so that secondary development of specific problems is rapidly realized.
2. The digital twin industrial software platform makeTwin according to claim 1, wherein the twin model generator module is developed by adopting a Spring development architecture, wherein a model library of the twin model generator module adopts a Spring container, XML language is used for describing a digital twin model of an object, and Java language is used for realizing internal logic of a software module; the object definition descriptions in the model library are
Where Obj represents different objects, i represents the number of objects in the model base, expressed as +.>
Wherein Name represents the Name of the object, type represents the class of the object, VM represents the digitized model of the object, pfm represents the performance of the object, para represents the parameter information of the object, port represents the interface of the object, cus represents the user-defined information; wherein VM is denoted->
Wherein->
Geometric appearance model representing an object, +.>
A physical mechanism model representing an object, < >>
Behavior model representing object->
A rule model representing an object;
the model assembly module is based on
The implementation of the assembly association relation of (1), SR represents a spatial relation, TR represents a time sequence relation, MR represents a mapping relation, LR represents a logic relation, and the assembly association relation is based on the instruction of the assembly association relation, so that the assembly association relation from a part level, a component level to an equipment level is realized, and then the assembly association relation is assembled into a unit level, a system level and a complex system level;
the model fusion module firstly establishes different attributes of different models and different physical field attributes including a temperature field, a flow field and vibration; secondly, abstracting different attributes of different models into abstract models with only Input and Output; coupling incidence relations among different models of different attributes are mutually interconnected; finally, the coupled model is converted into an executable model.
3. The digital twin industrial software platform makeTwin according to claim 1, wherein the communication connector module is developed using a Spring development architecture, and wherein the protocol library uses a Spring container; supporting a user to write a protocol which is temporarily not existed in a protocol library, and storing the protocol into the protocol library; the communication connector module integrates C, C #, C++, python and VB language compatible environments, and is compatible with C, C #, C++, python and VB language-written communication protocol function libraries; the communication connector module supports a user to call an existing communication protocol from a protocol library and configures specific conventions of the communication protocol, and comprises the following steps: (1) field length configuration including frame start length
Arbitration field Length->
Control field Length->
Data field Length->
CRC field Length->
ACK field length +.>
End of frame Length +.>
The method comprises the steps of carrying out a first treatment on the surface of the (2) Bit meaning configuration including frame start bit->
Wherein->
Arbitration field bit->
Wherein->
Control field bit->
Wherein
Data field bit->
Wherein->
CRC field bits->
Wherein->
ACK field bit->
Wherein->
End of frame bit->
Wherein
。
4. The digital twin industrial software platform makeTwin according to claim 1, wherein the virtual-real interaction configurator module comprises the following functions:
(1) the method supports the rapid configuration of the actions of a user on the digital twin model, and for a physical entity, all the actions are defined as follows:
wherein, AName represents the set of action names, AJNum represents the number of action joints, ajrelay represents the joint cascade relation sequence, AStart represents the set of starting positions of each joint action, AEnd represents the set of ending positions of each joint action;
wherein the method comprises the steps of
Spatial coordinates representing the starting position of each action, +.>
Spatial coordinates representing the end position of each action, a=1, 2, …, AJNum;
(2) establishing an association mapping relation between the start and stop of each action of the digital twin model and real-time data from a physical entity, and driving the change of the state of the digital twin model based on the acquired real-time data when the software platform runs;
(3) in order to ensure that the digital twin model and the physical entity still keep high consistency after long-time operation, the self-adaptive control algorithm of model reference is used for reference, the reference model in the algorithm is replaced by the digital twin model, and the output of the digital twin model is output
And performing differential processing with the output y of the physical entity, and then adjusting the parameters of the controller based on the self-adaptive mechanism, so as to realize the self-adaptive correction of the virtual-real consistency.
5. A digital twin industrial software platform makeTwin according to claim 1, characterised in that the twin data processor module comprises the following functions:
(1) the digital twin model constructed by the twin model generator module is supported to be input with parameters under different working conditions by a user, and simulation calculation is carried out, so that state data which cannot be monitored, characteristic data with small sample size and process data with fuzzy rules are obtained;
(2) the database contains design data, performance data, process data, task data, fault data and maintenance data, and supports user addition, modification, deletion and calling of data;
(3) supporting a user to carry out extraction of data characteristics, and extracting frequency domain characteristics and time domain characteristics of actual data and generated data respectively;
(4) supporting analysis of relevance of different data, including numerical accuracy
Feature relevance->
Trend fitness ++>
The method comprises the steps of carrying out a first treatment on the surface of the (5) Support the user to reduce the dimension of a large number of data with different characteristics, and analyze the error between the reduced dimension data and the true value, including the mean +.>
Maximum Max, mean square error S, peak factor PAR, skewness factor SF, and kurtosis factor KF.
6. The digital twin industrial software platform makeTwin according to claim 1, wherein the algorithm solver module comprises a data preprocessing class algorithm, a correlation analysis class algorithm, a time series class algorithm, a scheduling optimization class algorithm, a neural network class algorithm, a statistical calculation and evaluation class algorithm; the surrounding algorithm is used and managed, and has the following functions:
(1) supporting a user to add a new algorithm written by the user and adding the new algorithm into an algorithm library;
(2) supporting a user to call an existing algorithm from an algorithm library, modifying codes of the algorithm, and debugging the modified algorithm;
(3) while modifying the algorithm, the user modifies the parameters of the algorithm and tests and compares different parameters.
7. The digital twin industrial software platform makeTwin according to claim 1, wherein the service template provider module adopts Spring architecture to package six types of design optimization type templates, scheduling optimization type templates, fault diagnosis type templates, quality analysis type templates, prediction analysis type templates and recurrence trace-back type templates into a Spring container, and the user selects any type of templates from the six types of templates, and after selecting the templates, the method comprises the following functions:
(1) the user carries out mathematical modeling on the problem to be solved in an integrated development environment in a mode of creating and calling from a model library, and modifies the model code;
(2) aiming at the actual problem demand, searching and extracting the required data, and preprocessing;
(3) constructing an algorithm required by problem solving in a mode of creating a new structure in an integrated development environment and calling the new structure from an algorithm library, and modifying codes and parameters of the algorithm;
(4) the user selects information to be checked, including an algorithm running process, an algorithm convergence process, a simulation calculation result and different algorithm comparison; the service template provider module adopts visualization tools to provide different types of visualization effects for users.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310139311.5A CN115826937B (en) | 2023-02-21 | 2023-02-21 | Digital twin industrial software platform makeTain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310139311.5A CN115826937B (en) | 2023-02-21 | 2023-02-21 | Digital twin industrial software platform makeTain |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115826937A CN115826937A (en) | 2023-03-21 |
| CN115826937B true CN115826937B (en) | 2023-04-25 |
Family
ID=85521948
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310139311.5A Active CN115826937B (en) | 2023-02-21 | 2023-02-21 | Digital twin industrial software platform makeTain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115826937B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10929771B2 (en) * | 2016-11-15 | 2021-02-23 | General Electric Company | Multimodal, small and big data, machine tearing systems and processes |
| US20220237532A1 (en) * | 2019-06-29 | 2022-07-28 | Sameer Phadke | System and Method for Modelling and Monitoring Processes in Organizations Using Digital Twins |
| CN111338300B (en) * | 2020-02-27 | 2020-11-06 | 广东工业大学 | Physical simulation method and system of production line based on digital twins |
| CN112306464B (en) * | 2020-10-14 | 2022-03-29 | 中国科学院沈阳自动化研究所 | Method and system for realizing information physical fusion in industrial scene by using digital twin |
| CN115696368A (en) * | 2021-07-23 | 2023-02-03 | 维沃移动通信有限公司 | Digital twinning method and device |
-
2023
- 2023-02-21 CN CN202310139311.5A patent/CN115826937B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115826937A (en) | 2023-03-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Peng et al. | A desktop virtual reality-based interactive modular fixture configuration design system | |
| US8046735B1 (en) | Transforming graphical objects in a graphical modeling environment | |
| Ma et al. | A hierarchically structured and constraint-based data model for intuitive and precise solid modeling in a virtual reality environment | |
| Xia et al. | A CAD/CAE incorporate software framework using a unified representation architecture | |
| Martin et al. | A VR-CAD Data Model for Immersive Design: The cRea-VR Proof of Concept | |
| Van Emmerik | Interactive design of 3D models with geometric constraints | |
| CN115826937B (en) | Digital twin industrial software platform makeTain | |
| US6608623B1 (en) | Feature-based macro language for definition of customized analytic and customized geometric functionality | |
| Fan et al. | Multidomain model integration for online collaborative system design and detailed design of complex mechatronic systems | |
| St-Aubin et al. | A web based modeling and simulation environment to support the DEVS simulation lifecycle | |
| Flotyński | Visual aspect-oriented modeling of explorable extended reality environments | |
| CN116679974A (en) | Multi-software package operation method, computer program product and terminal equipment | |
| CN115469860A (en) | Method and system for automatically generating demand-to-software field model based on instruction set | |
| Li et al. | Digital twin technology in intelligent manufacturing | |
| KR100261272B1 (en) | Dynamic construction method of 3d virtual world on www | |
| Bredenfeld | Integration and evolution of model-based tool prototypes | |
| Flotynski et al. | Semantic Representation of Domain Knowledge for Professional VR Training. | |
| Deshmukh et al. | Multi-agent design architecture for integrated design systems | |
| Su et al. | An Efficient Modeling Method for CAD Data in Virtual Reality System | |
| Liang | A web-based collaborative design architecture for developing immersive VR driving platform | |
| Cao et al. | A distributed simulation system and its application | |
| Yi et al. | Research Method of Assembly Sequence Planning Based on Assembly Constraint Relationship Analysis | |
| Karsai et al. | Towards Two-Level Formal Modeling of Computer-Based Systems. | |
| Guiling et al. | Virtual human control technology for immersed virtual maintenance system in DELMIA environment | |
| Wang et al. | Virtual Assembly Method Based on UG |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |