CN115356949A - System and method for maintaining consistency of digital twin model - Google Patents
System and method for maintaining consistency of digital twin model Download PDFInfo
- Publication number
- CN115356949A CN115356949A CN202211052645.0A CN202211052645A CN115356949A CN 115356949 A CN115356949 A CN 115356949A CN 202211052645 A CN202211052645 A CN 202211052645A CN 115356949 A CN115356949 A CN 115356949A
- Authority
- CN
- China
- Prior art keywords
- digital twin
- model
- twin model
- data
- consistency
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 89
- 238000004088 simulation Methods 0.000 claims abstract description 33
- 238000012795 verification Methods 0.000 claims abstract description 17
- 230000008859 change Effects 0.000 claims description 57
- 238000012217 deletion Methods 0.000 claims description 20
- 230000037430 deletion Effects 0.000 claims description 20
- 238000007781 pre-processing Methods 0.000 claims description 10
- 230000001502 supplementing effect Effects 0.000 claims description 10
- 230000035945 sensitivity Effects 0.000 claims description 5
- 238000012423 maintenance Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
Abstract
The invention discloses a system and a method for keeping consistency of a digital twin model, which comprises the steps of (1) reading data of a physical entity and a simulation result of the digital twin model, taking the difference value of the data of the physical entity and the simulation result of the digital twin model as a model deviation value, and comparing the model deviation value with a consistency threshold value of the digital twin model to realize consistency judgment of the digital twin model; step (2), analyzing the cause of model deviation according to the model consistency judgment result and the model structure characteristics, and carrying out dynamic evolution on the structure and parameters of the digital twin model; and (3) reading the data of the physical entity and the simulation result of the digital twin model after model evolution, calculating a model deviation value and comparing the model deviation value with a consistency threshold value, and realizing consistency verification of the digital twin model. The method can provide a dynamic maintaining method for the consistency of the digital twin model and lay a foundation for providing effective application and service for the digital twin model to a certain extent.
Description
Technical Field
The invention belongs to the field of electronic engineering and computer science, and particularly relates to a system and a method for maintaining consistency of a digital twin model.
Background
A digital twin model based on real-time data driving can describe the state and characteristics of a physical entity. The feature described by the digital twin model and the feature of the physical entity are consistent in the dynamic operation process of the model, and the key point of the implementation of the digital twin model is. At present, discussion on the consistency problem of the digital twin model mainly centers on a judgment method of the consistency of the digital twin model (patent application number 201910920573.9) and a consistency maintaining method of an electromechanical physical model (patent application number 201910099067.8), but the methods have obvious defects and fail to consider two aspects of structural change and parameter change of a physical entity. Therefore, the invention discloses a system and a method for maintaining the consistency of a digital twin model, which can maintain the consistency of the digital twin model in the dynamic operation process to a certain extent through the consistency check, the dynamic evolution and the consistency verification of the digital twin model.
Disclosure of Invention
In order to solve the technical problems, the invention discloses a digital twin model consistency maintaining system and a digital twin model consistency maintaining method, which are suitable for digital twin models with multi-dimensional characteristics, in particular to digital twin models of complex equipment, so that when a physical entity actually works, high-precision operation of the models can be realized through the digital twin consistency maintaining method, and the digital twin model consistency maintaining system and the digital twin model consistency maintaining method are used for controlling, predicting and optimizing the physical entity subsequently.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
the invention relates to a consistency maintaining system for a digital twin model, which comprises: the consistency judging module, the evolution module and the consistency verifying module;
the consistency judging module is used for determining a consistency threshold of a digital twin model of a certain physical entity, wherein the consistency threshold ensures that the simulation of the digital twin model meets basic requirements; taking the difference value of the data of the physical entity and the simulation result of the digital twin model as a deviation value of the digital twin model, comparing the deviation value of the digital twin model with a consistency threshold value of the digital twin model, and if the deviation value of the digital twin model is lower than the consistency threshold value of the digital twin model, enabling the digital twin model to meet the consistency requirement of the digital twin model, so as to obtain a judgment result of the consistency of the digital twin model; otherwise, if the consistency requirement of the digital twin model is not met, the data need to be sent to an evolution module for carrying out digital twin model evolution;
the evolution module divides the deviation of the digital twin model into two types of structural deviation and parameter deviation according to the structure of the digital twin model of the physical entity; determining the deviation type of the digital twin model according to the consistency judgment result of the digital twin model in the consistency judgment module and the structure of the digital twin model; if the deviation type of the digital twin model is a structural deviation, adopting a structural evolution method to obtain the digital twin model after structural evolution; if the deviation type of the digital twin model is parameter deviation, adopting a parameter evolution method to obtain the digital twin model after parameter evolution;
the consistency verification module is used for judging whether the parameters of the evolved digital twin model meet the structural requirements of the digital twin model or not, and if the parameters do not meet the structural requirements, the digital twin model is evolved again, namely the evolution module is repeatedly evolved; reading the simulation result of the evolved digital twin model and the data of the physical entity, and taking the difference value of the simulation result and the data of the physical entity as the accuracy of the digital twin model; and comparing the accuracy of the digital twin model with the consistency threshold, if the accuracy of the digital twin model is lower than the consistency threshold, the evolved digital twin model meets the consistency requirement, and if the accuracy of the digital twin model is higher than the consistency threshold, the evolved digital twin model does not meet the consistency requirement and needs to be re-evolved, namely the content of the evolution module is repeated, and finally the consistency verification result of the digital twin model is obtained.
In the evolution module, the structure evolution method comprises the following steps:
(1) aiming at a certain physical entity, the physical entity is composed of different components, a digital twin model of the physical entity is composed of component models, and the component models describe the components of the physical entity; the components of the physical entity are composed of function modules, the component model is composed of function module models, and the function module models describe the functions of the components of the physical entity; the digital twin model, the component model and the function module model are all stored in a digital twin model library;
(2) determining the structural change condition of a physical entity, wherein the structural change is divided into an external structural change and an internal structural change; the external structural change comprises addition and deletion of components, and the internal structural change comprises addition and deletion of functional modules;
(3) acquiring data of a physical entity, judging whether a component to which the data acquired from the physical entity belongs is changed or not, if so, belonging to an external structure change, and performing the evolution method of the step (4), and if not, belonging to an internal structure change and belonging to a functional module to which the data belongs, and performing the evolution method of the step (5);
(4) aiming at the component increase condition of data acquired from a physical entity, matching a newly increased component model in a digital twin model library, and increasing an I/O interface of the component model to obtain a digital twin model after structure evolution; according to the component deletion condition of the data acquired from the physical entity, deleting the corresponding digital twin model and the I/O interface to obtain a digital twin model after structure evolution; the components of the physical entity are composed of functional modules, and the change of the components can cause the change of the functional modules, so that the internal structure evolution is needed after the external structure evolution of the digital twin model, namely, the step (5) is executed;
(5) adding an incidence relation between a component model and a function module model aiming at the increase condition of a function module to which data obtained from a physical entity belongs to obtain a digital twin model after parameter evolution; aiming at the deletion condition of the functional module to which the data obtained from the physical entity belongs, the incidence relation between the component model and the functional module model needs to be deleted, and the digital twin model after parameter evolution is obtained.
In the evolution module, the parameter evolution method comprises the following steps:
(1) determining characteristic parameters of a digital twin model according to the parameter sensitivity of the digital twin model, storing data required for constructing the digital twin model into a data set for constructing the digital twin model, and judging an evolution strategy by determining the similarity between the data characteristics of a physical entity and the characteristics of the data set, wherein the judging method comprises the following steps: if the similarity is high, indicating that the current physical entity does not generate a new state, and carrying out the evolution method in the step (2); otherwise, if the similarity is low, the current physical entity generates a new state, and the evolution method of the step (3) is carried out;
(2) acquiring data of a physical entity, selecting sample data, carrying out data preprocessing to remove error/redundant data, supplementing the processed data into a data set, and updating parameters of a digital twin model according to the updated data set to obtain a digital twin model after parameter evolution;
(3) acquiring data of a physical entity, selecting sample data and carrying out data preprocessing, carrying out similarity judgment on the sample data and the data of a data set, selecting and removing the data with the lowest similarity in the data set, finally supplementing the sample data into the data set, and updating parameters of a digital twin model according to the updated data set to obtain the digital twin model after parameter evolution;
the invention discloses a method for keeping consistency of a digital twin model, which comprises the following steps:
step 1, consistency judgment is carried out, and the following concrete implementation is carried out:
(1) aiming at a certain physical entity, determining a consistency threshold of the digital twin model, wherein the consistency threshold needs to ensure that the simulation of the digital twin model meets basic requirements, and the consistency threshold needs to be determined according to simulation requirements;
(2) reading data of a physical entity and a digital twin model simulation result, and taking a difference value of the data and the digital twin model simulation result as a digital twin model deviation value;
(3) comparing the deviation value of the digital twin model with the consistency threshold value of the digital twin model, wherein if the deviation value of the digital twin model is lower than the consistency threshold value of the digital twin model, the digital twin model meets the consistency requirement of the digital twin model, otherwise, the digital twin model does not meet the consistency requirement of the digital twin model and needs to be further evolved, so that a judgment result of the consistency of the digital twin model is obtained;
and 2, carrying out evolution, wherein the evolution is specifically realized as follows:
(1) dividing model deviation into two types of model structure deviation and model parameter deviation according to the structure of the digital twin model;
(2) determining the deviation type of the digital twin model according to the consistency judgment result of the digital twin model and the structure of the digital twin model in the step 1;
(3) if the deviation type is the model structure deviation, adopting a model structure evolution method; if the deviation type is the model parameter deviation, adopting a model parameter evolution method so as to obtain a digital twin model after model evolution;
(1) judging whether the parameters of the digital twin model conform to the structural requirements of the digital twin model based on the digital twin model after model evolution obtained in the step 2, if not, carrying out digital twin model evolution again, namely repeating the step 2;
(2) reading a simulation result of the digital twin model after model evolution and physical entity data, and taking the difference value of the simulation result and the physical entity data as the accuracy of the digital twin model;
(3) comparing the accuracy of the digital twin model with the consistency threshold of the digital twin model in the step 1, if the accuracy of the digital twin model is lower than the consistency threshold of the digital twin model, the digital twin model after model evolution meets the consistency requirement of the digital twin model, and if the accuracy of the digital twin model is higher than the consistency threshold of the digital twin model, the digital twin model does not meet the consistency requirement of the digital twin model and needs to be re-evolved, namely the step 2 is repeated, so that the consistency verification result of the digital twin model is obtained.
In the step 2, the model structure deviation is realized by adopting a model structure evolution method, and the specific evolution method comprises the following steps:
(1) aiming at a certain physical entity, the physical entity is composed of different components, a digital twin model of the physical entity is composed of component models, and the component models need to describe the components of the physical entity; the components of the physical entity are composed of function modules, the component model is composed of function module models, and the function module models need to describe the functions of the components of the physical entity; the digital twin model, the component model and the function module model are all required to be stored in a digital twin model library;
(2) determining the structural change condition of a physical entity, wherein the structural change is divided into an external structural change and an internal structural change; the external structural change comprises addition and deletion of components, and the internal structural change comprises addition and deletion of functional modules;
(3) acquiring data of a physical entity, judging whether a component to which the data belongs changes or not, if so, belonging to external structure change, wherein the evolution method is step (4), and if not, belonging to internal structure change and belonging to functional module change, and the evolution method is step (5);
(4) aiming at the increase condition of the components to which the data belong, matching newly-added component models in a digital twin model library, and increasing I/O interfaces of the component models; according to the component deletion condition of the data, deleting a corresponding digital twin model and an I/O interface; the components of the physical entity are composed of functional modules, and the change of the components can cause the change of the functional modules, so that the internal structure evolution is needed after the external structure evolution of the digital twin model, namely, the step (5) is executed;
(5) aiming at the increase condition of the function module to which the data belongs, the association relation between the component model and the function module model needs to be added; aiming at the deletion condition of the functional module to which the data belongs, the incidence relation between the component model and the functional module model needs to be deleted;
in the step 2, a model parameter evolution method is adopted for the model parameter deviation, and the model parameter evolution method comprises the following steps:
determining model characteristic parameters according to the parameter sensitivity of the digital twin model, and judging a model evolution strategy by determining the similarity between the data characteristics of the physical entity and the data characteristics in the model data set, wherein the judging method comprises the following steps: if the similarity is high, the current physical entity does not generate a new state, and a sample replacement strategy is adopted at the moment; otherwise, if the similarity is low, the current physical entity generates a new state, and a sample addition strategy is adopted at the moment; updating the model data set according to the acquired data of the physical entity, thereby updating the parameters of the digital twin model and obtaining an updated digital twin model;
the model data set is a set of data required by constructing a digital twin model;
the sample appending strategy is as follows: acquiring data of a physical entity, selecting sample data, carrying out data preprocessing to remove error/redundant data, and supplementing the processed data into a model data set;
the sample replacement policy: the method comprises the steps of obtaining data of a physical entity, selecting sample data, carrying out data preprocessing, carrying out similarity judgment on the sample data and the data of a model data set, selecting and removing the data with the lowest similarity in the model data set, and finally supplementing the sample data into the model data set.
The method is suitable for digital twin models with different dimensions, including a physical model, a behavior model and a rule model.
Compared with the prior art, the invention has the advantages that: the method comprises the steps of dividing the deviation reasons of the digital twin model into a model structure and model parameters, and selecting a corresponding model evolution method to update the digital twin model. The consistency between the digital twin model and the physical entity characteristics in the operation process is kept through multiple iterations of the steps of consistency judgment, dynamic evolution, consistency verification and the like of the digital twin model, so that the high-precision operation of the model can be realized through a digital twin consistency keeping method when the physical entity actually works, and the method is used for controlling, predicting and optimizing the subsequent physical entity.
The method comprises the steps of judging the consistency of a digital twin model driven by dynamic data, and realizing the evaluation of the model consistency in the dynamic running process of the digital twin model; analyzing the cause of the model deviation, and selecting an evolution method to carry out dynamic evolution on the model parameters and the model structure so as to realize dynamic update of the digital twin model; and comparing the simulation result of the digital twin model after model evolution with the data of the physical entity to realize model consistency verification. And finally realizing the consistency in the dynamic running process of the digital twin model based on the iterative cycle of consistency judgment, model evolution and consistency verification, and providing technical support for the application of the digital twin model. The method can solve the problem that the dynamic operation of the digital twin model is inaccurate due to the change of the physical entity structure or parameters to a certain extent, so that the digital twin model has the capability of keeping consistency when the physical entity actually works, and is used for controlling, predicting and optimizing the physical entity subsequently.
Drawings
FIG. 1 is a block diagram of a digital twin model consistency maintenance system according to the present invention;
FIG. 2 is a flow chart of a digital twin model structure evolution method of the present invention;
FIG. 3 is a flow chart of a method for parameter evolution of a digital twin model according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments, and all other embodiments obtained by a person skilled in the art based on the embodiments of the present invention belong to the protection scope of the present invention without creative efforts.
The dynamic operation of the digital twin model driven by data is a key characteristic of digital twin, so that the key of the application of the digital twin landing is to keep the digital twin model consistent with the characteristics of physical entities in the operation process. In the embodiment of the invention, the digital twin model has physical, behavior and rule multi-dimensional characteristics, and the consistency of the digital twin model needs to cover all the dimensional characteristics of the model. To this end, according to an embodiment of the present invention, a digital twin model consistency maintaining method is proposed, which is applicable to a digital twin model having a multidimensional feature. The consistency judgment of the digital twin model driven by dynamic data is included, and the evaluation of the model accuracy in the dynamic running process of the digital twin model is realized; analyzing the cause of the model deviation, and selecting an evolution method to carry out dynamic evolution on the model parameters and the model structure so as to realize dynamic updating of the model parameters; and comparing the simulation result of the digital twin model after model evolution with the data of the physical entity to realize model consistency verification. Based on the iterative cycle process of consistency judgment, model evolution and consistency verification, the consistency of the digital twin model in the dynamic operation process is finally realized, and the problem that the dynamic operation of the digital twin model is inaccurate due to the change of the physical entity structure or parameters is solved, so that the digital twin model has the consistency maintaining capability when the physical entity actually works, and is used for subsequent control, prediction and optimization of the physical entity.
FIG. 1 shows a general block diagram of a digital twin model consistency maintenance system according to the present invention, which comprises: the consistency judging module 1, the evolution module 2 and the consistency verifying module 3;
the flow block diagram of the model structure evolution method is shown in fig. 2, the flow block diagram of the model parameter evolution method is shown in fig. 3, and the specific implementation manner is as follows:
(1) And reading the data of a physical entity and the simulation result of the digital twin model aiming at a certain physical entity, taking the difference value of the data of the physical entity and the simulation result of the digital twin model as a model deviation value, and comparing the model deviation value with the consistency threshold value of the digital twin model to judge the consistency of the digital twin model. The specific implementation process is as follows:
(1) aiming at a certain physical entity, determining a consistency threshold of the digital twin model, wherein the consistency threshold needs to ensure that the simulation of the digital twin model meets basic requirements, and the consistency threshold needs to be determined according to simulation requirements;
(2) reading data of a physical entity and a simulation result of the digital twin model, and taking the difference value of the data and the simulation result of the digital twin model as a deviation value of the digital twin model;
(3) comparing the deviation value of the digital twin model with the consistency threshold value of the digital twin model, wherein if the deviation value of the digital twin model is lower than the consistency threshold value of the digital twin model, the digital twin model meets the consistency requirement of the digital twin model, otherwise, the digital twin model does not meet the consistency requirement of the digital twin model and needs to be further evolved, and the evolution method is shown in step (2), so that a judgment result of the consistency of the digital twin model is obtained;
(2) Dividing the model deviation into two types of model structure deviation and model parameter deviation according to the structure of the digital twin model, determining the deviation type of the digital twin model according to the consistency judgment result of the digital twin model in the step (1) and the digital twin model structure, and adopting a model structure evolution method if the deviation type is the model structure deviation; and if the deviation type is the model parameter deviation, adopting a model parameter evolution method so as to obtain the digital twin model after the model evolution.
As shown in fig. 2, the specific implementation flow of the model structure evolution method is as follows:
(1) aiming at a certain physical entity, the physical entity is composed of different components, therefore, a digital twin model of the physical entity is composed of component models, and the component models need to describe the components of the physical entity; the components of the physical entity are composed of functional modules, so that the component model is composed of functional module models, and the functional module models need to describe the functions of the components of the physical entity; the digital twin model, the component model and the function module model are all required to be stored in a digital twin model library;
(2) determining the structural change condition of a physical entity, wherein the structural change is divided into an external structural change and an internal structural change; the external structural change comprises addition and deletion of components, and the internal structural change comprises addition and deletion of functional modules;
(3) acquiring data of a physical entity, judging whether a component to which the data belongs changes or not, if so, belonging to external structure change, wherein the evolution method is step (4), and if not, belonging to internal structure change and belonging to functional module change, and the evolution method is step (5);
(4) aiming at the increase condition of the components to which the data belong, matching newly-added component models in a digital twin model library, and increasing I/O interfaces of the component models; according to the component deletion condition of the data, deleting a corresponding digital twin model and an I/O interface; the components of the physical entity are composed of functional modules, and the change of the components can cause the change of the functional modules, so that the internal structure evolution is needed after the external structure evolution of the digital twin model, namely the step (5) is executed;
(5) aiming at the increase condition of the function module to which the data belongs, the association relation between the component model and the function module model needs to be added; and aiming at the deletion condition of the functional module to which the data belongs, the incidence relation between the component model and the functional module model needs to be deleted.
As shown in fig. 3, the specific implementation flow of the model parameter evolution method is as follows:
determining model characteristic parameters according to the parameter sensitivity of the digital twin model, and judging a model evolution strategy by determining the similarity between the data characteristics of the physical entity and the data characteristics in the model data set, wherein the judging method comprises the following steps: if the similarity is high, the current physical entity does not generate a new state, and a sample replacement strategy is adopted at the moment; otherwise, if the similarity is low, the current physical entity generates a new state, and a sample addition strategy is adopted at the moment; updating the model data set according to the acquired data of the physical entity, thereby updating the parameters of the digital twin model and obtaining an updated digital twin model;
the model data set is a set of data required for constructing the digital twin model;
the sample appending strategy is as follows: acquiring data of a physical entity, selecting sample data, carrying out data preprocessing to remove error/redundant data, and supplementing the processed data into a model data set;
the sample replacement strategy is as follows: and acquiring data of a physical entity, selecting sample data, carrying out data preprocessing, carrying out similarity judgment on the sample data and the data of the model data set, selecting and removing the data with the lowest similarity in the model data set, and finally supplementing the sample data into the model data set.
(3) And (4) carrying out consistency verification on the digital twin model after model evolution. The specific implementation process is as follows:
(1) judging whether the parameters of the digital twin model meet the structural requirements of the digital twin model or not based on the digital twin model after the model evolution obtained in the step (2), if not, carrying out the digital twin model evolution again, namely, repeating the step (2);
(2) reading a simulation result of the digital twin model after model evolution and physical entity data, and taking the difference value of the simulation result and the physical entity data as the accuracy of the digital twin model;
(3) comparing the accuracy of the digital twin model with the consistency threshold of the digital twin model in the step (1), if the accuracy of the digital twin model is lower than the consistency threshold of the digital twin model, the digital twin model after model evolution meets the consistency requirement of the digital twin model, and if the accuracy of the digital twin model is higher than the consistency threshold of the digital twin model, the digital twin model does not meet the consistency requirement of the digital twin model, and re-evolution is needed, namely, the step (2) is repeated, so that the consistency verification result of the digital twin model is obtained.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and amendments can be made without departing from the principle of the present invention, and these modifications and amendments should also be considered as the protection scope of the present invention.
Claims (6)
1. A digital twin model consistency maintenance system, comprising: the consistency judging module, the evolution module and the consistency verifying module;
the consistency judging module is used for determining a consistency threshold of a digital twin model of a certain physical entity, and the consistency threshold ensures that the simulation of the digital twin model meets basic requirements; taking the difference value of the data of the physical entity and the simulation result of the digital twin model as a deviation value of the digital twin model, comparing the deviation value of the digital twin model with a consistency threshold value of the digital twin model, and if the deviation value of the digital twin model is lower than the consistency threshold value of the digital twin model, enabling the digital twin model to meet the consistency requirement of the digital twin model, so as to obtain a judgment result of the consistency of the digital twin model; otherwise, if the consistency requirement of the digital twin model is not met, the data need to be sent to an evolution module for carrying out digital twin model evolution;
the evolution module divides the deviation of the digital twin model into two types of structural deviation and parameter deviation according to the structure of the digital twin model of the physical entity; determining the deviation type of the digital twin model according to the consistency judgment result of the digital twin model in the consistency judgment module and the structure of the digital twin model; if the deviation type of the digital twin model is a structural deviation, adopting a structural evolution method to obtain the digital twin model after structural evolution; if the deviation type of the digital twin model is parameter deviation, adopting a parameter evolution method to obtain the digital twin model after parameter evolution;
the consistency verification module is used for judging whether the parameters of the evolved digital twin model meet the structural requirements of the digital twin model or not, and if the parameters do not meet the structural requirements, the digital twin model is evolved again, namely the evolution module is repeatedly evolved; reading the simulation result of the evolved digital twin model and the data of the physical entity, and taking the difference value of the simulation result and the data of the physical entity as the accuracy of the digital twin model; and comparing the accuracy of the digital twin model with the consistency threshold, if the accuracy of the digital twin model is lower than the consistency threshold, the evolved digital twin model meets the consistency requirement, and if the accuracy of the digital twin model is higher than the consistency threshold, the evolved digital twin model does not meet the consistency requirement and needs to be re-evolved, namely the content of the evolution module is repeated, and finally the consistency verification result of the digital twin model is obtained.
2. The system for maintaining consistency of the digital twin model as claimed in claim 1, wherein in the evolution module, the structural evolution method is implemented to include the following steps:
(1) aiming at a certain physical entity, the physical entity is composed of different components, a digital twin model of the physical entity is composed of component models, and the component models describe the components of the physical entity; the components of the physical entity are composed of functional modules, the component model is composed of functional module models, and the functional module models describe the functions of the components of the physical entity; the digital twin model, the component model and the function module model are all stored in a digital twin model library;
(2) determining the structural change condition of a physical entity, wherein the structural change is divided into an external structural change and an internal structural change; the external structural change comprises addition and deletion of components, and the internal structural change comprises addition and deletion of functional modules;
(3) acquiring data of a physical entity, judging whether a component to which the data acquired from the physical entity belongs is changed or not, if so, belonging to an external structure change, and performing the evolution method of the step (4), and if not, belonging to an internal structure change and belonging to a functional module to which the data belongs, and performing the evolution method of the step (5);
(4) aiming at the component increase condition of data acquired from a physical entity, matching a newly increased component model in a digital twin model library, and increasing an I/O interface of the component model to obtain a digital twin model after structure evolution; aiming at the component deletion condition of data acquired from a physical entity, deleting a corresponding digital twin model and an I/O interface to obtain a digital twin model after structure evolution; the components of the physical entity are composed of functional modules, and the change of the components can cause the change of the functional modules, so that the internal structure evolution is needed after the external structure evolution of the digital twin model, namely the step (5) is executed;
(5) adding an incidence relation between a component model and a function module model aiming at the increase condition of a function module to which data obtained from a physical entity belongs to obtain a digital twin model after parameter evolution; aiming at the deletion condition of the functional module to which the data obtained from the physical entity belongs, the incidence relation between the component model and the functional module model needs to be deleted, and the digital twin model after parameter evolution is obtained.
3. The system for maintaining consistency of a digital twin model as claimed in claim 1, wherein the parameter evolution method in the evolution module comprises the following steps:
(1) determining characteristic parameters of a digital twin model according to the parameter sensitivity of the digital twin model, storing data required for constructing the digital twin model into a data set for constructing the digital twin model, and judging an evolution strategy by determining the similarity between the data characteristics of a physical entity and the characteristics of the data set, wherein the judging method comprises the following steps: if the similarity is high, indicating that the current physical entity does not generate a new state, and carrying out the evolution method in the step (2); otherwise, if the similarity is low, the current physical entity generates a new state, and the evolution method of the step (3) is carried out;
(2) acquiring data of a physical entity, selecting sample data, carrying out data preprocessing to remove error/redundant data, supplementing the processed data into a data set, updating parameters of a digital twin model according to the updated data set, and obtaining the digital twin model after parameter evolution;
(3) acquiring data of a physical entity, selecting sample data, carrying out data preprocessing, carrying out similarity judgment on the sample data and the data of the data set, selecting and removing the data with the lowest similarity in the data set, finally supplementing the sample data into the data set, updating parameters of the digital twin model according to the updated data set, and obtaining the digital twin model after parameter evolution.
4. A consistency maintaining method for a digital twin model is characterized by comprising the following steps:
step 1, consistency judgment is carried out, and the method is specifically realized as follows:
(1) aiming at a certain physical entity, determining a consistency threshold of the digital twin model, wherein the consistency threshold needs to ensure that the simulation of the digital twin model meets basic requirements, and the consistency threshold needs to be determined according to simulation requirements;
(2) reading data of a physical entity and a digital twin model simulation result, and taking a difference value of the data and the digital twin model simulation result as a digital twin model deviation value;
(3) comparing the deviation value of the digital twin model with the consistency threshold value of the digital twin model, wherein if the deviation value of the digital twin model is lower than the consistency threshold value of the digital twin model, the digital twin model meets the consistency requirement of the digital twin model, otherwise, the digital twin model does not meet the consistency requirement of the digital twin model and needs to be further evolved, so that a judgment result of the consistency of the digital twin model is obtained;
and 2, carrying out evolution, specifically realizing the following steps:
(1) dividing model deviation into two types of model structure deviation and model parameter deviation according to the structure of the digital twin model;
(2) determining the deviation type of the digital twin model according to the consistency judgment result of the digital twin model and the structure of the digital twin model in the step 1;
(3) if the deviation type is the model structure deviation, adopting a model structure evolution method; if the deviation type is the model parameter deviation, adopting a model parameter evolution method so as to obtain a digital twin model after model evolution;
step 3, carrying out consistency verification, and concretely realizing the following steps:
(1) judging whether the parameters of the digital twin model conform to the structural requirements of the digital twin model based on the digital twin model after model evolution obtained in the step 2, if not, carrying out digital twin model evolution again, namely, repeating the step 2;
(2) reading a simulation result of the digital twin model after model evolution and physical entity data, and taking the difference value of the simulation result and the physical entity data as the accuracy of the digital twin model;
(3) comparing the accuracy of the digital twin model with the consistency threshold of the digital twin model in the step 1, if the accuracy of the digital twin model is lower than the consistency threshold of the digital twin model, the digital twin model after model evolution meets the consistency requirement of the digital twin model, and if the accuracy of the digital twin model is higher than the consistency threshold of the digital twin model, the digital twin model does not meet the consistency requirement of the digital twin model and needs to be re-evolved, namely the step 2 is repeated, so that the consistency verification result of the digital twin model is obtained.
5. The method for maintaining consistency of the digital twin model as claimed in claim 4, wherein in the step 2, the model structure deviation is realized by using a model structure evolution method, and the specific evolution method comprises the following steps:
(1) aiming at a certain physical entity, the physical entity is composed of different components, a digital twin model of the physical entity is composed of component models, and the component models need to describe the components of the physical entity; the components of the physical entity are composed of function modules, the component model is composed of function module models, and the function module models need to describe the functions of the components of the physical entity; the digital twin model, the component model and the function module model are all required to be stored in a digital twin model library;
(2) determining the structural change condition of a physical entity, wherein the structural change is divided into an external structural change and an internal structural change; the external structural change comprises addition and deletion of components, and the internal structural change comprises addition and deletion of functional modules;
(3) acquiring data of a physical entity, judging whether a component to which the data belongs changes or not, if so, belonging to external structure change, wherein the evolution method is step (4), and if not, belonging to internal structure change and belonging to functional module change, and the evolution method is step (5);
(4) aiming at the increase condition of the components to which the data belong, matching newly-added component models in a digital twin model library, and increasing I/O interfaces of the component models; according to the component deletion condition of the data, deleting a corresponding digital twin model and an I/O interface; the components of the physical entity are composed of functional modules, and the change of the components can cause the change of the functional modules, so that the internal structure evolution is needed after the external structure evolution of the digital twin model, namely the step (5) is executed;
(5) aiming at the increase condition of the function module to which the data belongs, the association relation between the component model and the function module model needs to be added; and aiming at the deletion condition of the function module to which the data belongs, the association relation between the component model and the function module model needs to be deleted.
6. The method for maintaining consistency of the digital twin model as claimed in claim 5, wherein in the step 2, the model parameter deviation adopts a model parameter evolution method, and the model parameter evolution method comprises the following steps:
determining model characteristic parameters according to the parameter sensitivity of the digital twin model, and judging a model evolution strategy by determining the similarity between the data characteristics of the physical entity and the data characteristics in the model data set, wherein the judging method comprises the following steps: if the similarity is high, the current physical entity does not generate a new state, and a sample replacement strategy is adopted at the moment; otherwise, if the similarity is low, the current physical entity generates a new state, and a sample addition strategy is adopted at the moment; updating the model data set according to the acquired data of the physical entity, thereby updating the parameters of the digital twin model and obtaining an updated digital twin model;
the model data set is a set of data required for constructing the digital twin model;
the sample appending strategy is as follows: acquiring data of a physical entity, selecting sample data, carrying out data preprocessing to remove error/redundant data, and supplementing the processed data into a model data set;
the sample replacement policy: the method comprises the steps of obtaining data of a physical entity, selecting sample data, carrying out data preprocessing, carrying out similarity judgment on the sample data and the data of a model data set, selecting and removing the data with the lowest similarity in the model data set, and finally supplementing the sample data into the model data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211052645.0A CN115356949B (en) | 2022-08-30 | 2022-08-30 | Digital twin model consistency maintaining system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211052645.0A CN115356949B (en) | 2022-08-30 | 2022-08-30 | Digital twin model consistency maintaining system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115356949A true CN115356949A (en) | 2022-11-18 |
| CN115356949B CN115356949B (en) | 2024-05-07 |
Family
ID=84004107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211052645.0A Active CN115356949B (en) | 2022-08-30 | 2022-08-30 | Digital twin model consistency maintaining system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115356949B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116751A (en) * | 2018-07-24 | 2019-01-01 | 西安西电电气研究院有限责任公司 | Digitization system and its construction method based on the twin technology of number |
| US20210109837A1 (en) * | 2019-10-09 | 2021-04-15 | International Business Machines Corporation | Digital twin workflow simulation |
| CN112905385A (en) * | 2021-01-27 | 2021-06-04 | 北京航空航天大学 | Digital twin model operation and iterative evolution method based on model backup |
| EP3944034A1 (en) * | 2020-07-21 | 2022-01-26 | Rockwell Automation Technologies, Inc. | Model-based design of linear synchronous motor transport systems |
-
2022
- 2022-08-30 CN CN202211052645.0A patent/CN115356949B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109116751A (en) * | 2018-07-24 | 2019-01-01 | 西安西电电气研究院有限责任公司 | Digitization system and its construction method based on the twin technology of number |
| US20210109837A1 (en) * | 2019-10-09 | 2021-04-15 | International Business Machines Corporation | Digital twin workflow simulation |
| EP3944034A1 (en) * | 2020-07-21 | 2022-01-26 | Rockwell Automation Technologies, Inc. | Model-based design of linear synchronous motor transport systems |
| CN112905385A (en) * | 2021-01-27 | 2021-06-04 | 北京航空航天大学 | Digital twin model operation and iterative evolution method based on model backup |
Non-Patent Citations (1)
| Title |
|---|
| 陶飞;刘蔚然;刘检华;刘晓军;刘强;屈挺;胡天亮;张执南;向峰;徐文君;王军强;张映锋;刘振宇;李浩;程江峰;戚庆林;张萌;张贺;隋芳媛;何立荣;易旺民;程辉;: "数字孪生及其应用探索", 计算机集成制造系统, no. 01, 15 January 2018 (2018-01-15), pages 4 - 21 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115356949B (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112528519A (en) | Method, system, readable medium and electronic device for engine quality early warning service | |
| CN112214617B (en) | Digital file management method and system based on block chain technology | |
| WO2023284443A1 (en) | Method and apparatus for determining working condition of excavator | |
| EP3871120A1 (en) | Apparatus and method for detecting an anomaly among successive events and computer program product therefor | |
| CN115687432A (en) | Method, apparatus, and medium for monitoring anomalous transaction data | |
| CN112163299A (en) | Excavator oil consumption prediction method and system and electronic equipment | |
| CN106557881B (en) | Business process system construction method based on business activity execution sequence | |
| CN115356949A (en) | System and method for maintaining consistency of digital twin model | |
| CN117318033A (en) | Power grid data management method and system combining data twinning | |
| CN116562012A (en) | Dynamic reliability analysis method based on Kriging model | |
| CN114756602A (en) | Real-time streaming process mining method and system and computer readable storage medium | |
| CN111625525B (en) | Environment data repairing/filling method and system | |
| CN112668223B (en) | Electromechanical equipment simulation method and system based on digital twin lightweight model | |
| CN111737319B (en) | User cluster prediction method, device, computer equipment and storage medium | |
| CN113610225A (en) | Quality evaluation model training method and device, electronic equipment and storage medium | |
| CN110990470A (en) | QAR data decoding method, system and storage medium based on distributed computation | |
| CN114117763B (en) | Model screening method based on shift-one cross-validation method | |
| CN115510170A (en) | Real-time index calculation method based on financial data incremental calculation | |
| CN117076141B (en) | High-applicability off-line data processing task issuing method and system | |
| CN117729176B (en) | Method and device for aggregating application program interfaces based on network address and response body | |
| CN117539948B (en) | Service data retrieval method and device based on deep neural network | |
| CN112416928B (en) | Demand prediction model for first-aid repair spare parts of ground complex electronic equipment | |
| US20230237225A1 (en) | Machine learning based reservoir modeling | |
| CN110120982B (en) | Sensor data compression method based on coal mine safety monitoring networking system | |
| CN116185990A (en) | Formatted data processing system with timestamp as version characteristic |
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 |