CN111461338A - Intelligent system updating method and device based on digital twin - Google Patents
Intelligent system updating method and device based on digital twin Download PDFInfo
- Publication number
- CN111461338A CN111461338A CN202010150875.5A CN202010150875A CN111461338A CN 111461338 A CN111461338 A CN 111461338A CN 202010150875 A CN202010150875 A CN 202010150875A CN 111461338 A CN111461338 A CN 111461338A
- Authority
- CN
- China
- Prior art keywords
- data
- virtual
- digital twin
- intelligent
- agent
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012549 training Methods 0.000 claims abstract description 58
- 238000004088 simulation Methods 0.000 claims abstract description 39
- 238000012937 correction Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims description 16
- 238000007726 management method Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 230000007246 mechanism Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 11
- 238000010276 construction Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 7
- 230000009467 reduction Effects 0.000 claims description 7
- 238000013501 data transformation Methods 0.000 claims description 6
- 239000000126 substance Substances 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 18
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000013473 artificial intelligence Methods 0.000 abstract description 6
- 230000004927 fusion Effects 0.000 abstract description 6
- 238000009940 knitting Methods 0.000 abstract description 4
- 238000012795 verification Methods 0.000 abstract description 4
- 239000003795 chemical substances by application Substances 0.000 description 63
- 230000008569 process Effects 0.000 description 14
- 238000005457 optimization Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 8
- 230000006399 behavior Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000001149 cognitive effect Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000004806 packaging method and process Methods 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 230000002787 reinforcement Effects 0.000 description 2
- 108010001267 Protein Subunits Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000007795 chemical reaction product Substances 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000013102 re-test Methods 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention provides an intelligent system updating method and device based on digital twins, firstly, acquiring physical quantity to be measured in a physical system based on training requirements of an intelligent body, digital twins correction requirements and the scalability of the physical quantity to obtain truth value data, then obtaining simulation data according to knitting data, training the intelligent body by taking the truth value data and the simulation data as training samples, continuously updating an initial intelligent body by using the trained intelligent body, and integrating digital twins with an intelligent body design and training technology based on an artificial intelligence technology, thereby providing an intelligent body continuous training method for guaranteeing effectiveness and mobility; meanwhile, virtual-real fusion correction and verification are carried out, so that the effectiveness of digital twins is guaranteed, and an intelligent convergence agent is trained to migrate to an actual physical system.
Description
Technical Field
The invention relates to the technical field of system optimization, in particular to an intelligent system updating method and device based on digital twin.
Background
In recent years, artificial intelligence technology has been rapidly developed, and intelligent systems based on artificial intelligence technology have gradually shown its power and gradually achieve a dramatic increase in performance in a plurality of fields, even in some fields far beyond human level. With the aid of deep learning techniques, computers are trained in massive data, and have reached a very high level in the fields of image recognition, target detection, machine translation, word and sentence prediction, and even in the field of creative design such as poetry creation, drawing creation, cover design, and the like. Based on deep reinforcement learning, massive interactive training of computers with given environments and rule programs has also respectively prevailed top-level human players in simple interactive games such as Atari games, MuJoCo, Gym, etc., round games such as go, chess, general chess, texas poker, etc., and complex instant strategy games such as Dota2, interplanetary contestant II, etc.
The digital twin is to create a virtual model of a physical entity in a digital mode, simulate the behavior of the physical entity in a real environment by means of data, and add or expand new capability to the physical entity through means of virtual-real interaction feedback, data fusion analysis, decision iteration optimization and the like. As a technology which fully utilizes models, data, intelligence and integrates multiple disciplines, the digital twin is oriented to the whole life cycle process of products, the functions of bridges and links connecting a physical world and an information world are exerted, and more real-time, efficient and intelligent services are provided.
For operation, maintenance, perfection and optimization upgrading of the system, including operation, maintenance, perfection and optimization upgrading of an intelligent system, the process is completed by human operation, and the specific form mainly comprises the following steps: (1) releasing a software upgrading patch or upgrading package or a new module, and completing the downloading (or copying), installation and deployment by a user (mainly aiming at software or an operating system and the like); (2) and dispatching deployment personnel by a design developer to perform upgrading deployment (mainly aiming at platforms deployed in an enterprise). The specific process is generally as follows: the new function design development and the design perfection of the existing function are carried out in the design development system, after the test in the design development system is correct, the updated part is deployed to the real physical system in a manual deployment mode, and after a series of physical system tests, the system updating and upgrading are completed after no problem exists. Although some internet companies have realized automatic iterative update of their business platform systems through intelligent module automatic update systems, they are all based on computer systems and do not involve perfect optimization of real physical systems.
Disclosure of Invention
To address at least one of the above-mentioned deficiencies, an embodiment of an aspect of the present invention provides a digital twin-based intelligent system updating method, including:
constructing a pair of digital twin virtual systems based on a preset physical system, wherein the physical system and the virtual system both comprise intelligent agents;
acquiring physical quantity to be measured in the physical system based on training requirements, digital twin correction requirements and the measurability of the physical quantity of the intelligent agent to obtain truth value data;
constructing a plurality of instances based on the digital twin, and performing parallel simulation through the virtual system to obtain virtual simulation data;
randomly extracting at least part of data in the true value data and the virtual simulation data to form training data, and training the agent;
and updating the initial agent in the physical system by using the trained agent.
In a preferred embodiment, the physical quantities to be measured include: system state information, agent awareness information, agent decision information, and control information.
In a preferred embodiment, the constructing multiple instances based on the digital twin and performing parallel simulation through the virtual system to obtain virtual simulation data includes:
building a plurality of instances of the virtual system by means of a Spring Cloud management framework and mechanism;
and performing parallel simulation on the multiple instances to generate the virtual simulation data.
In a preferred embodiment, further comprising:
preprocessing the truth value data to obtain preprocessed truth value data; wherein the content of the first and second substances,
the preprocessing includes at least one of noise reduction filtering, missing value processing, and data transformation.
Another embodiment of the present invention provides an intelligent system updating apparatus based on digital twin, including:
the virtual system construction module is used for constructing a pair of digital twin virtual systems based on a preset physical system, and the physical system and the virtual system both comprise intelligent agents;
the truth value data obtaining module is used for acquiring physical quantity to be measured in the physical system based on the training requirement of the intelligent agent, the digital twin correction requirement and the measurability of the physical quantity to obtain truth value data;
the virtual simulation data acquisition module is used for constructing a plurality of instances based on the digital twin and carrying out parallel simulation through the virtual system to obtain virtual simulation data;
the training module randomly extracts at least part of data in the true value data and the virtual simulation data to form training data and trains the agent;
and the updating module is used for updating the initial agent in the physical system by using the trained agent.
In a preferred embodiment, the physical quantities to be measured include: system state information, agent awareness information, agent decision information, and control information.
In a preferred embodiment, the virtual simulation data obtaining module includes:
the instance construction unit is used for constructing a plurality of instances of the virtual system by means of a Spring Cloud management framework and a Spring Cloud management mechanism;
and the virtual simulation data generation unit is used for performing parallel simulation on the plurality of instances to generate the virtual simulation data.
In a preferred embodiment, further comprising:
the preprocessing module is used for preprocessing the truth value data to obtain the preprocessed truth value data; wherein the preprocessing comprises at least one of noise reduction filtering, missing value processing, and data transformation.
In yet another embodiment of the present invention, an electronic device is provided, which includes a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the digital twin-based intelligent system updating method when executing the program.
A further embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned digital twin-based intelligent system updating method.
The invention has the following beneficial effects:
the invention provides an intelligent system updating method and device based on digital twins, firstly, acquiring physical quantity to be measured in a physical system based on training requirements of an intelligent body, digital twins correction requirements and the scalability of the physical quantity to obtain truth value data, then obtaining simulation data according to knitting data, training the intelligent body by taking the truth value data and the simulation data as training samples, continuously updating an initial intelligent body by using the trained intelligent body, and integrating digital twins with an intelligent body design and training technology based on an artificial intelligence technology, thereby providing an intelligent body continuous training method for guaranteeing effectiveness and mobility; meanwhile, virtual-real fusion correction and verification are carried out, so that the effectiveness of digital twins is guaranteed, and an intelligent convergence agent is trained to migrate to an actual physical system.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart illustrating a digital twin-based intelligent system updating method according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram illustrating an intelligent system updating apparatus based on digital twin according to an embodiment of the present invention.
FIG. 3 shows a schematic structural diagram of an electronic device suitable for use in implementing embodiments of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the 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, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
With the increasing increase of complexity and updating integrity and real-time requirement of some intelligent systems and the problems faced by the intelligent systems, the manual testing, deploying and retesting modes are difficult to deal with. As is well known, the perfect optimization of an intelligent system is mainly based on data driving, and when a system needs to be updated and upgraded in a short time when new data is required to be obtained in some scenes, the existing manual deployment mode is difficult to implement, and particularly when an actual system in the physical world is faced, the system needs to be manually tested and then completed after deployment. Therefore, the invention provides an intelligent system continuous updating method based on a digital twin, and the core concept is that the rapid continuous updating of a complex intelligent system is realized by constructing the digital twin of a pair of embedded intelligent body modules and rapidly training an intelligent body based on the digital twin. The method is based on digital twins, a large amount of simulation data of a virtual environment and actual data of a physical environment are accumulated to serve as historical experience, the intelligent agent serving as a cognitive decision module is driven to perform iterative optimization, and continuous evolution improvement of the cognitive decision capability of the intelligent agent is achieved by means of continuous training and iterative updating. Specifically, firstly, acquiring physical quantity to be measured in the physical system based on training requirements of the intelligent agent, digital twin correction requirements and the scalability of the physical quantity to obtain truth value data, then obtaining simulation data according to knitting data, training the intelligent agent by using the truth value data and the simulation data as training samples, continuously updating an initial intelligent agent by using the trained intelligent agent, and integrating digital twin and an intelligent agent design and training technology based on an artificial intelligence technology to provide an intelligent agent continuous training method for guaranteeing effectiveness and mobility; meanwhile, virtual-real fusion correction and verification are carried out, so that the effectiveness of digital twins is guaranteed, and an intelligent convergence agent is trained to migrate to an actual physical system.
Fig. 1 illustrates an intelligent system updating method based on digital twin according to an embodiment of the present invention, including:
s1: based on a preset physical system, a pair of digital twin virtual systems is constructed, wherein the physical system and the virtual systems both comprise intelligent bodies.
It is understood that a pair of digital twins includes a physical system, and a virtual system (also referred to as a simulation system) which integrates multidisciplinary, multi-physical quantity, multi-scale, multi-probability simulation and maps in a virtual space according to various data in the physical system, such as sensor data, operation data and the like.
The intelligent agent is embedded into the controller as a cognitive decision module, and the cognition of the system state and the autonomous decision based on the cognition are completed by controlling the interaction of the controlled object and the physical environment. The process will produce a certain amount of control process data including system state information, agent awareness information, agent decision information, and control information, among others. And a part of the data is observed by the sensor, and is uploaded to the cloud after being preprocessed.
In a preferred embodiment, since physical space data is generally small, this portion of data is used for both virtual space simulation system modification in digital twins in subsequent steps, and for agent training.
In some embodiments, this step is mainly performed with reference to an existing digital twin construction method, and a virtual system or a computer system forming a pair of digital twin with the existing physical system is constructed based on the existing physical system. The construction process can be based on mechanism modeling, and a mathematical model is constructed through mechanism analysis of a real physical system, so that the guarantee of the state consistency and the input and output consistency of a virtual-real system is realized. Meanwhile, the construction of the process can be realized based on non-mechanism modeling, and the virtual-real consistency is realized on the behavior by fitting the state and the input-output relation of the physical system by adopting methods such as deep neural network training and the like.
S2: and acquiring physical quantities to be measured in the physical system based on the training requirements, the digital twin correction requirements and the measurability of the physical quantities of the intelligent agent to obtain truth value data.
In some embodiments, the physical quantity to be measured includes: system state information, agent awareness information, agent decision information, and control information.
S3: and constructing a plurality of examples based on the digital twin, and performing parallel simulation through the virtual system to obtain virtual simulation data.
Step S3 specifically includes:
s31: with the Spring Cloud management framework and mechanism, multiple instances of the virtual system are built.
Specifically, in some embodiments, the step mainly involves, by means of a management framework and a mechanism such as Spring Cloud, and by means of large-scale multi-instance construction and parallel simulation of the virtual system, fast generation of massive virtual simulation data is achieved, so as to support efficient training of the agent. Specifically, in some embodiments, the simulation template packaging may be performed first, where the simulation template packaging packages start, operation, and end of the simulation engine and its environment configuration into one script, and packages the simulation engine, the simulation model, and the like into a mirror image, so as to implement deployment, operation, and management of the simulation engine and its model on any computing node through mirror image replication and script operation.
And then, managing the simulation examples by the Spring Cloud framework, building the Spring Cloud management framework, and fusing the application management of the framework with the encapsulated simulation template, namely, managing the multiple simulation examples by means of a mechanism of managing Cloud application by the Spring Cloud.
S32: and performing parallel simulation on the multiple instances to generate the virtual simulation data.
In the embodiment, by means of supporting the parallel read-write database, parallel write-in of massive simulation data generated by parallel simulation and parallel read-out of a plurality of pieces of data in a subsequent training process are realized.
S4: and randomly extracting at least part of the true value data and the virtual simulation data to form training data, and training the intelligent agent.
The method comprises the following steps of obtaining virtual data and physical data, performing optimization on the virtual data and the physical data, and performing iterative updating on an intelligent agent in a virtual environment and a physical environment by means of intelligent agent model synchronization. The whole method is based on digital twins, real-time consistency of behaviors of both twins guarantees the consistency of the virtual data and the physical data on the whole, and effectiveness of migration application of an intelligent agent obtained by virtual data training to a real physical system, so that the intelligent agent model obtained by training can be directly migrated to the physical system. At the same time, given that the nature of the agent model is actually a parametrical function, the agent behavior is completely determined by both the parameter and the functional form. Based on this, only need when the intelligent agent model migrates to duplicate function form and corresponding parameter can, can accomplish automatically.
The intelligent agent training algorithm mainly comprises three categories of supervised learning, unsupervised learning and reinforcement learning according to different specific task scenes, the specific design needs to be determined according to the specific task scenes, and the intelligent agent training algorithm generally comprises intelligent agent structural design and training design. The structure design of the intelligent agent is to select and design the mathematical expression form of the intelligent agent (such as a neural network, a decision tree, a Bayes classifier and the like); the training design of the intelligent agent mainly designs how the intelligent agent finishes the correction of the self parameters through self prediction data and truth value data by means of an optimization algorithm based on rules given by people. The designed structure and training algorithm of the intelligent agent can be used for training to form the intelligent agent, and the training data is extracted from the database and the training algorithm of the intelligent agent is filled.
S5: and updating the initial agent in the physical system by using the trained agent.
In particular, the nature of the agent model is actually a parametrical function, and the agent behavior is completely determined by both the parametric and functional forms. Therefore, the storage/update of the intelligent agent model only needs to save/update the function form and the corresponding parameters. For updating the model to the physical system, the process can also be accomplished by functional form and parameter update thanks to the characteristics of the digital twin.
In some preferred embodiments, the intelligent system updating method further includes:
s01: preprocessing the truth value data to obtain preprocessed truth value data; wherein the content of the first and second substances,
the preprocessing includes at least one of noise reduction filtering, missing value processing, and data transformation.
In particular, the acquired data is preprocessed to obtain usable, valid measurement data. The preprocessing process includes noise reduction filtering, missing value processing, necessary transformation and the like. In addition, the processed data splits the training data from the correction data. For training data, based on the range characteristics of the measured physical quantity, normalization processing is performed on the data, and all dimension data are reduced to a space for subsequent training. And finally, uploading the processed data to a cloud.
The core of the invention is to provide a continuous updating method of an intelligent system based on digital twins, which is mainly oriented to continuous optimization upgrading of the intelligent system related to a real physical system, completes quick training of an intelligent body by means of fusion data of virtual data and real physical data generated by massive multi-instance parallel simulation, and realizes formation and migration of a perfect optimization model by ensuring consistency of virtual and real behaviors through the digital twins.
Compared with the current system perfection optimization method of manual test, deployment and retest (physical system test), the method provided by the invention can realize real-time and full automation of the upgrading process, and can automatically and quickly complete the optimization and updating of the intelligent system when the newly generated data reaches a certain amount after the set time or the physical system is changed.
Based on the same inventive concept, another embodiment of the present invention provides a digital twin-based intelligent system updating apparatus, as shown in fig. 2, including:
the virtual system construction module 1 is used for constructing a pair of digital twin virtual systems based on a preset physical system, wherein the physical system and the virtual system both comprise intelligent agents;
a truth value data obtaining module 2, which is used for acquiring physical quantity to be measured in the physical system based on the training requirement of the intelligent agent, the digital twin correction requirement and the measurability of the physical quantity to obtain truth value data;
the virtual simulation data obtaining module 3 is used for constructing a plurality of examples based on the digital twin and carrying out parallel simulation through the virtual system to obtain virtual simulation data;
the training module 4 randomly extracts at least part of the true value data and the virtual simulation data to form training data, and trains the agent;
and the updating module 5 is used for updating the initial agent in the physical system by using the trained agent.
According to the intelligent system updating device based on the digital twin, firstly, physical quantity to be measured in the physical system is collected based on training requirements of the intelligent body, digital twin correction requirements and the measurability of the physical quantity to obtain truth value data, then simulation data is obtained according to knitting data, the truth value data and the simulation data are used as training samples to train the intelligent body, the initial intelligent body is continuously updated by the trained intelligent body, the intelligent body design and training technology based on the digital twin and artificial intelligence technology is fused, and an intelligent body continuous training method for guaranteeing effectiveness and mobility is provided; meanwhile, virtual-real fusion correction and verification are carried out, so that the effectiveness of digital twins is guaranteed, and an intelligent convergence agent is trained to migrate to an actual physical system.
In a preferred embodiment, the physical quantities to be measured include: system state information, agent awareness information, agent decision information, and control information.
In a preferred embodiment, the virtual simulation data obtaining module includes:
the instance construction unit is used for constructing a plurality of instances of the virtual system by means of a Spring Cloud management framework and a Spring Cloud management mechanism;
and the virtual simulation data generation unit is used for performing parallel simulation on the plurality of instances to generate the virtual simulation data.
In a preferred embodiment, the updating means further comprises:
the preprocessing module is used for preprocessing the truth value data to obtain the preprocessed truth value data; wherein the preprocessing comprises at least one of noise reduction filtering, missing value processing, and data transformation.
An embodiment of the present invention further provides a specific implementation manner of an electronic device, which is capable of implementing all steps in the method in the foregoing embodiment, and referring to fig. 3, the electronic device specifically includes the following contents:
a processor (processor)601, a memory (memory)602, a communication interface (communications interface)603, and a bus 604;
the processor 601, the memory 602 and the communication interface 603 complete mutual communication through the bus 604;
the processor 601 is used to call the computer program in the memory 602, and when the processor executes the computer program, the processor implements all the steps of the method in the above embodiments.
Embodiments of the present invention also provide a computer-readable storage medium capable of implementing all the steps of the method in the above embodiments, the computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements all the steps of the method in the above embodiments.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the hardware + program class embodiment, since it is substantially similar to the method embodiment, the description is simple, and the relevant points can be referred to the partial description of the method embodiment. Although embodiments of the present description provide method steps as described in embodiments or flowcharts, more or fewer steps may be included based on conventional or non-inventive means. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an actual apparatus or end product executes, it may execute sequentially or in parallel (e.g., parallel processors or multi-threaded environments, or even distributed data processing environments) according to the method shown in the embodiment or the figures. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, article, or apparatus that comprises the recited elements is not excluded. For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, in implementing the embodiments of the present description, the functions of each module may be implemented in one or more software and/or hardware, or a module implementing the same function may be implemented by a combination of multiple sub-modules or sub-units, and the like. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form. The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein. The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of an embodiment of the specification. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction. The above description is only an example of the embodiments of the present disclosure, and is not intended to limit the embodiments of the present disclosure. Various modifications and variations to the embodiments described herein will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present specification should be included in the scope of the claims of the embodiments of the present specification.
Claims (10)
1. A digital twin-based intelligent system updating method is characterized by comprising the following steps:
constructing a pair of digital twin virtual systems based on a preset physical system, wherein the physical system and the virtual system both comprise intelligent agents;
acquiring physical quantity to be measured in the physical system based on training requirements, digital twin correction requirements and the measurability of the physical quantity of the intelligent agent to obtain truth value data;
constructing a plurality of instances based on the digital twin, and performing parallel simulation through the virtual system to obtain virtual simulation data;
randomly extracting at least part of data in the true value data and the virtual simulation data to form training data, and training the agent;
and updating the initial agent in the physical system by using the trained agent.
2. The digital twin-based intelligent system updating method according to claim 1, wherein the physical quantity to be measured includes: system state information, agent awareness information, agent decision information, and control information.
3. The intelligent system updating method based on digital twin as claimed in claim 2, wherein the building of multiple instances based on digital twin and the parallel simulation by the virtual system to obtain virtual simulation data comprises:
building a plurality of instances of the virtual system by means of a Spring Cloud management framework and mechanism;
and performing parallel simulation on the multiple instances to generate the virtual simulation data.
4. The digital twin based intelligent system updating method according to claim 2, further comprising:
preprocessing the truth value data to obtain preprocessed truth value data; wherein the content of the first and second substances,
the preprocessing includes at least one of noise reduction filtering, missing value processing, and data transformation.
5. An intelligent system updating device based on digital twinning, comprising:
the virtual system construction module is used for constructing a pair of digital twin virtual systems based on a preset physical system, and the physical system and the virtual system both comprise intelligent agents;
the truth value data obtaining module is used for acquiring physical quantity to be measured in the physical system based on the training requirement of the intelligent agent, the digital twin correction requirement and the measurability of the physical quantity to obtain truth value data;
the virtual simulation data acquisition module is used for constructing a plurality of instances based on the digital twin and carrying out parallel simulation through the virtual system to obtain virtual simulation data;
the training module randomly extracts at least part of data in the true value data and the virtual simulation data to form training data and trains the agent;
and the updating module is used for updating the initial agent in the physical system by using the trained agent.
6. The digital twin-based intelligent system updating apparatus according to claim 5, wherein the physical quantity to be measured includes: system state information, agent awareness information, agent decision information, and control information.
7. The digital twin based intelligent system updating apparatus according to claim 6, wherein the virtual simulation data obtaining module comprises:
the instance construction unit is used for constructing a plurality of instances of the virtual system by means of a Spring Cloud management framework and a Spring Cloud management mechanism;
and the virtual simulation data generation unit is used for performing parallel simulation on the plurality of instances to generate the virtual simulation data.
8. The digital twin based intelligent system updating apparatus according to claim 6, further comprising:
the preprocessing module is used for preprocessing the truth value data to obtain the preprocessed truth value data; wherein the content of the first and second substances,
the preprocessing includes at least one of noise reduction filtering, missing value processing, and data transformation.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the digital twin based intelligent system updating method of any of claims 1 to 4 are implemented when the program is executed by the processor.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the digital twin based intelligent system updating method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010150875.5A CN111461338A (en) | 2020-03-06 | 2020-03-06 | Intelligent system updating method and device based on digital twin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010150875.5A CN111461338A (en) | 2020-03-06 | 2020-03-06 | Intelligent system updating method and device based on digital twin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111461338A true CN111461338A (en) | 2020-07-28 |
Family
ID=71682692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010150875.5A Pending CN111461338A (en) | 2020-03-06 | 2020-03-06 | Intelligent system updating method and device based on digital twin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111461338A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111858642A (en) * | 2020-07-31 | 2020-10-30 | 科大讯飞股份有限公司 | Data set updating method, related device and readable storage medium |
| CN112138396A (en) * | 2020-09-23 | 2020-12-29 | 中国电子科技集团公司第十五研究所 | Intelligent training method and system for unmanned system simulation confrontation |
| CN112417619A (en) * | 2020-11-23 | 2021-02-26 | 江苏大学 | Pump unit optimal operation adjusting system and method based on digital twinning |
| CN112711631A (en) * | 2020-11-27 | 2021-04-27 | 国网山东省电力公司电力科学研究院 | Digital twin information synchronization method, system, readable storage medium and device |
| CN112925291A (en) * | 2021-01-22 | 2021-06-08 | 大连理工大学 | Digital twin automatic driving test method based on camera dark box |
| CN113379123A (en) * | 2021-06-09 | 2021-09-10 | 中德（珠海）人工智能研究院有限公司 | Fault prediction method, device, server and storage medium based on digital twin |
| CN113852515A (en) * | 2021-08-26 | 2021-12-28 | 西安电子科技大学广州研究院 | Node state control method and system of digital twin network |
| CN113868898A (en) * | 2021-11-29 | 2021-12-31 | 联想新视界（北京）科技有限公司 | Data processing method and device based on digital twin model |
| CN114035425A (en) * | 2021-11-05 | 2022-02-11 | 广东工业大学 | Virtual and real state synchronization method and system for unmanned aerial vehicle digital twin system |
| CN114187728A (en) * | 2021-12-09 | 2022-03-15 | 香港理工大学 | Fire monitoring method and system based on artificial intelligence and digital twin technology |
| CN114398175A (en) * | 2021-12-30 | 2022-04-26 | 中国电信股份有限公司 | Intelligent interaction system and method, electronic device and storage medium |
| CN114926583A (en) * | 2022-05-09 | 2022-08-19 | 北京工业大学 | Site simulation management method and device, electronic equipment and storage medium |
| CN115169855A (en) * | 2022-06-29 | 2022-10-11 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115577641A (en) * | 2022-11-14 | 2023-01-06 | 成都飞机工业（集团）有限责任公司 | Training method, device, equipment and medium for digital twin model |
| CN115994458A (en) * | 2023-03-23 | 2023-04-21 | 华南理工大学 | Virtual-real integrated multi-agent cluster system simulation method |
| WO2023082949A1 (en) * | 2021-11-10 | 2023-05-19 | 达闼科技（北京）有限公司 | Agent control method and apparatus, electronic device, program, and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109471377A (en) * | 2018-11-22 | 2019-03-15 | 浙江科技学院 | Based on two twin axis X-Y leather and fur products Cutting machine control methods of number |
| CN109947567A (en) * | 2019-03-14 | 2019-06-28 | 深圳先进技术研究院 | A kind of multiple agent intensified learning dispatching method, system and electronic equipment |
| CN110456635A (en) * | 2019-07-02 | 2019-11-15 | 北京航空航天大学 | The control method of power system of electric automobile based on the twin technology of number |
| CN110488629A (en) * | 2019-07-02 | 2019-11-22 | 北京航空航天大学 | A kind of management-control method of the hybrid vehicle based on the twin technology of number |
| CN110765589A (en) * | 2019-09-10 | 2020-02-07 | 上海大学 | Intelligent workshop virtual and real synchronous monitoring system and method based on digital twins |
| CN110824954A (en) * | 2019-10-24 | 2020-02-21 | 北京仿真中心 | Intelligent agent training method and system, computer equipment and readable storage medium |
-
2020
- 2020-03-06 CN CN202010150875.5A patent/CN111461338A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109471377A (en) * | 2018-11-22 | 2019-03-15 | 浙江科技学院 | Based on two twin axis X-Y leather and fur products Cutting machine control methods of number |
| CN109947567A (en) * | 2019-03-14 | 2019-06-28 | 深圳先进技术研究院 | A kind of multiple agent intensified learning dispatching method, system and electronic equipment |
| CN110456635A (en) * | 2019-07-02 | 2019-11-15 | 北京航空航天大学 | The control method of power system of electric automobile based on the twin technology of number |
| CN110488629A (en) * | 2019-07-02 | 2019-11-22 | 北京航空航天大学 | A kind of management-control method of the hybrid vehicle based on the twin technology of number |
| CN110765589A (en) * | 2019-09-10 | 2020-02-07 | 上海大学 | Intelligent workshop virtual and real synchronous monitoring system and method based on digital twins |
| CN110824954A (en) * | 2019-10-24 | 2020-02-21 | 北京仿真中心 | Intelligent agent training method and system, computer equipment and readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 李伯虎;林廷宇;贾政轩;施国强;肖莹莹;张迎曦;曾贲;李鹤宇;郭丽琴;: "智能工业系统智慧云设计技术" * |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111858642B (en) * | 2020-07-31 | 2022-12-06 | 科大讯飞股份有限公司 | Data set updating method, related device and readable storage medium |
| CN111858642A (en) * | 2020-07-31 | 2020-10-30 | 科大讯飞股份有限公司 | Data set updating method, related device and readable storage medium |
| CN112138396A (en) * | 2020-09-23 | 2020-12-29 | 中国电子科技集团公司第十五研究所 | Intelligent training method and system for unmanned system simulation confrontation |
| CN112138396B (en) * | 2020-09-23 | 2024-04-12 | 中国电子科技集团公司第十五研究所 | Unmanned system simulation countermeasure-oriented intelligent body training method and system |
| CN112417619A (en) * | 2020-11-23 | 2021-02-26 | 江苏大学 | Pump unit optimal operation adjusting system and method based on digital twinning |
| CN112711631B (en) * | 2020-11-27 | 2022-07-08 | 国网山东省电力公司营销服务中心(计量中心) | Digital twin information synchronization method, system, readable storage medium and device |
| CN112711631A (en) * | 2020-11-27 | 2021-04-27 | 国网山东省电力公司电力科学研究院 | Digital twin information synchronization method, system, readable storage medium and device |
| CN112925291B (en) * | 2021-01-22 | 2021-12-24 | 大连理工大学 | Digital twin automatic driving test method based on camera dark box |
| CN112925291A (en) * | 2021-01-22 | 2021-06-08 | 大连理工大学 | Digital twin automatic driving test method based on camera dark box |
| CN113379123A (en) * | 2021-06-09 | 2021-09-10 | 中德（珠海）人工智能研究院有限公司 | Fault prediction method, device, server and storage medium based on digital twin |
| CN113852515A (en) * | 2021-08-26 | 2021-12-28 | 西安电子科技大学广州研究院 | Node state control method and system of digital twin network |
| CN114035425B (en) * | 2021-11-05 | 2022-07-26 | 广东工业大学 | Virtual and real state synchronization method and system for unmanned aerial vehicle digital twin system |
| CN114035425A (en) * | 2021-11-05 | 2022-02-11 | 广东工业大学 | Virtual and real state synchronization method and system for unmanned aerial vehicle digital twin system |
| WO2023082949A1 (en) * | 2021-11-10 | 2023-05-19 | 达闼科技（北京）有限公司 | Agent control method and apparatus, electronic device, program, and storage medium |
| CN113868898A (en) * | 2021-11-29 | 2021-12-31 | 联想新视界（北京）科技有限公司 | Data processing method and device based on digital twin model |
| CN114187728A (en) * | 2021-12-09 | 2022-03-15 | 香港理工大学 | Fire monitoring method and system based on artificial intelligence and digital twin technology |
| CN114398175A (en) * | 2021-12-30 | 2022-04-26 | 中国电信股份有限公司 | Intelligent interaction system and method, electronic device and storage medium |
| WO2023124393A1 (en) * | 2021-12-30 | 2023-07-06 | 中国电信股份有限公司 | Intelligent interaction system and method, and electronic device and storage medium |
| CN114926583A (en) * | 2022-05-09 | 2022-08-19 | 北京工业大学 | Site simulation management method and device, electronic equipment and storage medium |
| CN114926583B (en) * | 2022-05-09 | 2023-06-27 | 北京工业大学 | Site simulation management method and device, electronic equipment and storage medium |
| CN115169855B (en) * | 2022-06-29 | 2023-05-26 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115169855A (en) * | 2022-06-29 | 2022-10-11 | 郑州轻工业大学 | Unsafe state detection method based on digital twin workshop mixed data set |
| CN115577641A (en) * | 2022-11-14 | 2023-01-06 | 成都飞机工业（集团）有限责任公司 | Training method, device, equipment and medium for digital twin model |
| CN115994458A (en) * | 2023-03-23 | 2023-04-21 | 华南理工大学 | Virtual-real integrated multi-agent cluster system simulation method |
| CN115994458B (en) * | 2023-03-23 | 2023-07-18 | 华南理工大学 | Virtual-real integrated multi-agent cluster system simulation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111461338A (en) | Intelligent system updating method and device based on digital twin | |
| CN110378463B (en) | Artificial intelligence model standardization training platform and automatic system | |
| Brazier et al. | Principles of component-based design of intelligent agents | |
| Railsback et al. | Improving execution speed of models implemented in NetLogo | |
| Santos et al. | Quantitatively assessing the benefits of model-driven development in agent-based modeling and simulation | |
| Park et al. | A semi-automated filtering technique for software process tailoring using neural network | |
| CN114048024A (en) | Task deployment method, device, equipment, storage medium and product | |
| Wang et al. | Automl from software engineering perspective: Landscapes and challenges | |
| Acuña et al. | A process model applicable to software engineering and knowledge engineering | |
| CN115053209A (en) | Method and system for generating engineering drawings in engineering system | |
| Dick et al. | Computational Intelligence in Software Quality Assurance | |
| Elstermann et al. | Subject-oriented reference model for virtual factory operations commissioning | |
| Kirikova et al. | Enterprise modelling: improving the quality of requirements specifications | |
| Rossiter | Simulation design: Trans-paradigm best-practice from software engineering | |
| CN114399804A (en) | Virtual laboratory teaching realization method, system, medium and equipment based on 3D simulation | |
| Quartuccio et al. | A model-based approach to investigate emergent behaviors in systems of systems | |
| Ören et al. | M&S Bok Core areas and the big picture | |
| Sadli | Using the python library to create simple game animations | |
| Johnston et al. | A Generic Framework for Approximate Simulation in Commonsense Reasoning Systems. | |
| Liu et al. | The management of simulation validation | |
| CN115545213B (en) | Modeling method and device based on graphical behavior tree and reinforcement learning | |
| Wang et al. | Model-based simulation testing for embedded software | |
| Fritz et al. | Systems Engineering Methods and Tools | |
| Laatabi et al. | Formalizing data to agent model mapping using MOF: application to a model of residential mobility in Marrakesh | |
| Persson | A DevOps Approach to the EA Blueprint Architectural Pattern |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200728 |
|
| RJ01 | Rejection of invention patent application after publication |