Generating digital twins, methods, systems, and computer program products

Through the combination of digital twin modular system and tuning module, the problem of difficulty in reusing digital twin systems in different environments is solved, the flexibility and security of the system is improved, production and commissioning costs are reduced, and intellectual property protection is enhanced. .

CN116529034BActive Publication Date: 2024-11-08SIEMENS IND SOFTWARE NV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202180078199.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-11-20
Filing Date
2021-10-21
Publication Date
2024-11-08
Estimated Expiration
2041-10-21

AI Technical Summary

Technical Problem

The existing digital twin systems are difficult to reuse flexibly in different environments, and their production and debugging processes are complex, resulting in high costs and waste of resources, increasing security risks and difficulty in protecting intellectual property rights.

Method used

Provide a digital twin modular system, which generates digital twins based on feature parameters through tuning modules, including computing characteristics, simulation range, model architecture, communication security, accessibility, data storage, encryption functions and hardware requirements, to improve the digital twin’s Adaptability and flexibility, and optimize computational performance and accuracy through prediction modules.

Benefits of technology

It realizes the ease of reusability and flexibility of digital twins in different environments, reduces production and commissioning costs, enhances security and intellectual property protection, and improves the reliability and efficiency of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116529034B_ABST
    Figure CN116529034B_ABST
Patent Text Reader

Abstract

The present invention relates to a computer-implemented method for generating a digital twin (DTW). In order to improve the efficiency of processing a digital twin (DTW), the method comprises the following steps: a.) providing a digital twin modular system (DTS), wherein the digital twin modular system (DTS) is characterized in that a digital twin module (DTM) is adapted to or can be adapted to a digital twin feature (DTE), wherein the digital twin feature (DTE) involves a computing characteristic (CPR), a simulation range (SCS), a model architecture (MAT), a communication security (CMS), an accessibility (ACB), a data storage (DST), an encryption function (ECF), a resource allocation (RAL), At least one of the hardware requirements (HRQ), b.) providing a tuning module (TMD), wherein the tuning module (TMD) includes a tuning module parameter set (TMS), wherein the tuning module parameter set (TMS) includes tuning module parameters (TMP) corresponding to at least one digital twin feature (DTF), respectively, c.) generating a digital twin (DTW) by using the tuning module (TMD) to perform the steps of selecting, configuring and combining digital twin modules (DTM) of the digital twin modular system (DTS) according to the tuning module parameters (TMP). In addition, a system, a computer program product and a computer-readable medium for applying the computer-implemented method are provided.
Need to check novelty before this filing date? Find Prior Art

Description

Field of the Invention

[0001] The present invention relates to a computer-implemented method for generating a digital twin and a system for performing the method, as well as a computer program product and a computer-readable medium.

[0002] More specifically, the present invention relates to generating digital twins that include specific features related to computational properties, simulation scope, model architecture, communication security, accessibility, data storage, encryption capabilities, resource allocation or hardware requirements. Background Art

[0003] In computer-physical systems, the digital twin concept includes three elements: a physical product, a digital or virtual product, and a connection between the two products. The digital or virtual product is considered a digital twin in the following. The connection between the physical product and the digital twin can be several sensors and data flowing from the physical product to the digital twin and information available from the digital twin to the physical environment.

[0004] A method for modularly generating digital twins is known from WO 2020 / 190272 A1. Documents WO 2018 / 111368A1, CN 111 880 497A, and US 2020 / 274768A1 disclose additional technical background information in the field of the present invention, and US 2019 / 173109 A1 discloses a digital twin based on a fuel cell management system, including independent digital twins for commissioning and for operation.

[0005] Complex products, or products that are planned to be part of more complex systems, are currently partially combined with digital twins, and may more often need to provide digital twins with products in the future. If the technology to be sold is not equipped with a digital twin, potential customers may not be able to simulate the correct operation of the entire system in advance to ensure that everything works together correctly. Often, providing a digital twin up front can be the default door opener for customers to make any sales move.

[0006] Providing a digital twin requires a lot of investment in research, development and implementation. Due to the high costs, most products are not provided in conjunction with a digital twin. Today, digital twins are often provided as separate work of software, and in case of changes, modifications in the software code require additional investments, which respectively result in many hours of work even for minor adjustments.

[0007] A significant risk arising from this trend is that the digital twin at least partially enables the product to be sold, which is related to the digital twin itself, but also to the specific hardware to be simulated by the digital twin. Third parties can analyze the digital twin, which can have been provided for sales purposes, and ultimately can be able to reconstruct the product at least partially at the time of provision. Additional information can be obtained from other sources to achieve a complete copy of the product. All information combined enables anyone familiar with such practices to save R&D efforts and provide a product with essentially the same characteristics.

[0008] The current rulings' numeric dual types, listed in chronological order, are:

[0009] 1. “Digital twin prototyping” consists of design, analysis and physical product design processes,

[0010] 2. A “Digital Twin Instance” is a digital twin of each individual instance of a manufactured product, and

[0011] 3. Digital Twin Aggregation (DTA) is an aggregation involving digital twins of operational products for interrogation, prediction, and learning about physical products.

[0012] The invention relates to all these types to improve the ease of reusability of a single module in different simulation environments. The invention also relates to a digital twin as a functional simulation interface (FMU or FMI; defining a standardized interface for developing complex computer-physical systems in computer simulations), an executable digital twin or similar concepts.

[0013] A digital twin that meets specific technical standards can be provided, enabling effortless use in a simulation environment and collaboration with other digital twins of components provided by different companies.

[0014] One object of the present invention is to improve the reusability of digital twins or at least their modules. To achieve this object, the present invention proposes to improve the adaptability of digital twins and the flexibility of the generation of digital twins. Summary of the invention

[0015] According to the present invention, a solution to the above problem is provided. In order to increase the flexibility of the digital twin concept and enable the digital twin to be easily reused in different environments, the present invention proposes a method of the type mentioned at the beginning, which method comprises the following additional steps:

[0016] a.) providing a digital twin modular system, wherein the digital twin modular system is characterized in that the digital twin module is applicable or applicable to the digital twin feature,

[0017] The digital twin feature involves at least one of computing characteristics, simulation scope, model architecture, communication security, accessibility, data storage, encryption function, resource allocation, and hardware requirements.

[0018] b.) providing a tuning module, the tuning module comprising a tuning module parameter set,

[0019] The tuning module parameter set includes tuning module parameters corresponding to at least one digital twin feature, respectively.

[0020] c.) generating a digital twin by using the tuning module to perform the steps of selecting, configuring and combining digital twin modules of the digital twin modular system according to the tuning module parameters.

[0021] The objects of the invention are achieved by the independent claims. The dependent claims describe advantageous developments and modifications of the invention.

[0022] Digital twin characteristics related to at least one of computational characteristics, simulation scope, model architecture, communication security, accessibility, data storage, encryption capabilities, resource allocation, hardware requirements are necessary characteristics of the digital twin related to the operational destination of the environment for which the digital twin is designed. In some environments, communication security may be very important, such as communications over public networks in conjunction with critical infrastructure or military technology. In other environments, communication security may be secondary, such as if communications are done in a closed network environment without current / physical connections to the public network.

[0023] The above digital twin features can partly relate to the same digital twin parameters and be interconnected or even interdependent. For example, computation speed can be related to computational behavior and digital twin hardware requirements.

[0024] For internal simulations of providers, it may be desirable to have a very fast digital twin to enable full-scale simulations of complex systems. It may be desirable not to be able to do such a full-scale simulation for each customer, or even to make it available to everyone by downloading it in the Internet. It may even be possible to provide artificial delays in the digital twin to limit specific uses.

[0025] The required computer performance is related to the digital twin hardware requirements and the computational behavior. The computational accuracy is related to the computational behavior and the digital twin hardware requirements. With the accuracy of the implementation of the hardware details in the digital twin, the risk of corruption of such a copy increases. Several properties of a digital twin involve security issues and the risk of losing confidential intellectual property. The customer may not have the same accuracy requirements as the vendor of the hardware to be simulated. Moreover, the accuracy may consume too much computing performance, so the customer needs a faster but less accurate version of the digital twin.

[0026] The operating scope of the digital twin is related to the scope of the simulation. The underlying physical or empirical model of the digital twin is related to the model architecture, computational behavior, and digital twin hardware requirements. It may be reasonable to limit the operating scope of the digital twin to avoid too much complexity and to avoid that the digital twin operates in an operating scope that is not recommended and may consume too much hardware life. It is also very important for the provider to limit the operating scope setting to avoid using the provided digital twin to simulate any third-party products with different operating scopes.

[0027] The complexity of the physics of considering / ignoring effects respectively is related to the model architecture, computational behavior and digital twin hardware requirements. For specific simulation purposes, it is reasonable to restrict or limit or simplify the physics implemented in the digital twin. For example, it may not be necessary to consider a specific temperature distribution in the hardware for a simulation with medium accuracy. On the other hand, for R&D purposes, it may be necessary to simulate the temperature distribution.

[0028] The number of input or output parameters, the choice of input or output parameters is related to the scope of the simulation, the computational behavior and the digital twin hardware requirements, respectively.

[0029] Often security features are also important, such as security features to prevent hacker attacks or security features to prevent manipulation of digital twin tuning parameters.

[0030] In order for the digital twin to be operated to simulate a technically complex environment - such as a nuclear power plant, it may be necessary to implement security features (encrypted communication, firewalls...) that can primarily or exclusively protect the digital twin from being manipulated. It may be necessary to avoid digital twin manipulation, since deviations of the digital twin behavior and the real hardware behavior could lead to an emergency shutdown of the plant or any other system, respectively. This would call into question the digital twin concept if only the digital twin had to be manipulated to cause a problem, while the hardware would be in perfect order. In order to avoid unexpected developments, it may be reasonable, under high safety requirements, to have a second virtual representation of the real system - a second digital twin of the real system that could be provided next to the "normal" digital twin.

[0031] Furthermore, when providing a digital twin to a third party (e.g., a customer), it may not be advisable to make all the functionality and full performance of the digital twin available. On the other hand, it may be too laborious to generate separate digital twins for internal use and digital twins for external use, such as for everyone's downloads, customer information, sales activities, customer simulations, and other purposes.

[0032] The present invention solves this problem by providing a tuning module capable of generating a digital twin that performs the steps of selecting, configuring and combining digital twin modules.

[0033] A preferred embodiment of the present invention provides a method according to the present invention preferably for critical systems, the method having the following additional steps:

[0034] - Provide a second digital twin next to the first digital twin, both of which are virtual representations of the same system,

[0035] - operating said second digital twin in parallel with the first digital twin,

[0036] - receiving a specific categorized or predetermined parameter to be received by the digital twin submitting that parameter only to the first digital twin, while delaying submission of that parameter for the second digital twin to alter the behavior of the digital twins such that, at least in a first instance, only one of the two digital twins will receive that categorized input while the other is without that input for a predetermined time interval,

[0037] - compare the behavior during that time interval,

[0038] - If the first digital twin behaves in a significantly unexpected manner compared to the reference data, isolating the second digital twin from receiving the parameters

[0039] -Compare two digital twins to the corresponding hardware behavior to determine which digital twin behaves more like the hardware and which digital twin may be missing functionality.

[0040] A preferred embodiment of the method preferably for critical systems can include the following additional steps:

[0041] - Identify the digital twin of the correct operation,

[0042] - Synchronize the digital twin to the correct operational digital twin,

[0043] -Optionally: compensate for any delays in the digital twin (which may be caused by delayed input) by adjusting the calculation speed if necessary.

[0044] Here, “synchronizing the two digital twins to the correctly operating digital twin” means that the correctly operating digital twin is the reference for the two digital twins, and after synchronizing the two digital twins, both digital twins have the configuration of the reference digital twin.

[0045] Such safety features may be beneficial for critical systems but may not be required for every virtual representation or digital twin of a system or for any sales version of a digital twin that may be intended only to facilitate the hardware or for any research and development tool that serves as a digital twin.

[0046] The decentralization of the digital twin, the cloudification or distribution of the system, respectively, relates to communication, data storage and resource allocation. The possibility of remote access to the system, respectively, relates to general access, security properties and communication. Remote accessibility of the digital twin can be a preferred application of the digital twin tuning module. This accessibility may only involve the possibility of tuning from a remote location, but can also include influencing the operation of the digital twin and can also involve data monitoring of the input data and output data of the digital twin, which is also useful for machine learning of certain neural networks on the behavior of the digital twin.

[0047] Protected communication or interaction with the digital twin involves communication and encryption of the communication. Protecting the communication or interaction of the digital twin with external modules involves remote access. It may be useful to deactivate this protection under inherently safe conditions to improve the performance of the digital twin. This is possible in isolated systems without any external access.

[0048] Stepwise monitoring involves the possibility for the provider of the digital twin to monitor the user's activities. The monitoring can be limited to the time of use and / or the repetition cycle of the solver and / or the number of outputs generated to any other countable item related to the use of the digital twin. The monitoring activity can be extended to observe each input and each output as well as each intermediate result. Stepwise monitoring parameters make it possible to adjust the monitoring activities of the provider related to the use of the digital twin, wherein the corresponding monitored items as well as the monitoring scope can be defined.

[0049] Similarly, information sharing involves access security, security and communication properties of the system. Sharing of information and data monitoring can be done within the limits defined by, for example, tuning module parameters of the provider of the digital twin. If the digital twin is provided free of charge, for example to perform a quick simulation (e.g., a customer checks whether the hardware is suitable for a complex system), it is possible that the provider reserves the right to monitor such operations and the provider has access to the input / output of the digital twin. If the customer pays for the hardware including the digital twin, this option can be deactivated at the customer's request to achieve a certain degree of total independence for the user.

[0050] The digital twin can be operated on only one CPU or can have a higher degree of distribution respectively. The tuning module can change the distribution settings of individual modules of the digital twin in the preparation phase or even during operation. This can be motivated by the allocation of computing resources on different computers and in the cloud, which can be motivated by the provider's desire to control the digital twin users. The distribution of the modules of the digital twin can enable the provider to keep certain confidential modules only on internal machines.

[0051] According to a preferred embodiment of the present invention, the tuning module can be provided as a module of the digital twin and can be accessed through the digital twin tuning module interface. This embodiment can quickly and easily adjust the digital twin features. Preferably, the tuning module access is protected by at least password authentication.

[0052] Another possibility is to have a tuning module, and the tuning module interface can be a separate module from the digital twin. This embodiment improves the security of the digital twin, preventing manipulation and possible loss of intellectual property or secrets that can be identified from an unprotected digital twin.

[0053] According to a preferred embodiment of the method, the tuning module of the digital twin includes access restrictions such as password protection that only allows specific access to identified users. Preferably, the tuning module of the digital twin can be password protected and / or can be protected differently to prevent manipulation. The tuning module can be a separate module or can be part of the digital twin enabling tuning without the need for an additional tuning module. In any of these variations, the digital twin and the tuning module can be well protected. Access restrictions should be based on known standards.

[0054] According to another preferred embodiment of the method, the digital twin includes at least one input interface and at least one output interface, wherein the digital twin includes at least one functional module that generates an output from an input. The input interface can be used to receive data from a sending module. The output interface can be used to send data to a receiving module, respectively. The input of the input interface can also be provided by a user, and the input interface can be designed as a user interface. The functional module generates at least one output from at least one input, which can include an application algorithm, a recursive function, a neural network or other functional modules or a combination thereof.

[0055] According to another preferred embodiment of the method, the tuning module comprises an information interface for providing information to a user, the information comprising an actual set value of at least two tuning module parameters, the method comprising:

[0056] - providing information to the user through the information interface, the information comprising actual setting values ​​of at least two tuning module parameters.

[0057] Such information displayed to the user via the information interface can be at least one adjustable first tuning parameter (fixed after user input) and a second tuning parameter (defined as free, respectively adjustable by a prediction module of the tuning module), depending on the first tuning parameter and its predicted value upon changing the first tuning parameter or even the second tuning parameter. Such a prediction module can work as described below.

[0058] According to another preferred embodiment of the method, the tuning module comprises a prediction module, and the method comprises:

[0059] - predicting at least one variation trend of at least one second tuning module parameter when changing at least one first tuning module parameter,

[0060] - providing at least the change trend to the information interface as a change trend indicator of at least the second tuning module parameter when changing at least the first tuning module parameter.

[0061] Such a prediction module can use at least one algorithm to derive predicted values ​​from known parameter values, respectively. Such an algorithm can include a neural network trained by a machine learning step, and it can be permanently trained during ongoing operation. The neural network can be trained by comparing the predicted values ​​with the results obtained by the digital twin during operation. A good example of the possibility of adjusting a fixed tuning parameter to a specific value can be the computational accuracy of a digital twin associated with a specific output parameter, which has a strong impact on the computational speed. When the user determines (possibly through a user input interface) the computational accuracy for a specific requirement, the prediction module can predict the computational speed of the digital twin. Additional aspects are described below.

[0062] Preferably, the method comprises the following additional steps:

[0063] - assigning some of said tuning module parameters to a set of fixed tuning module parameters, and

[0064] - assigning other of the tuning module parameters to a set of free tuning module parameters and tuning module parameters, wherein the prediction module is designed to derive estimates of the free tuning module parameters from the fixed tuning module parameters.

[0065] Preferably, the method comprises the following additional steps:

[0066] - receiving an interface input via a user interface, the user interface comprising assigning at least two tuning module parameters to one of the classifiers: a fixed tuning module parameter, a free tuning module parameter,

[0067] - determining the assigned tuning module parameters as free tuning module parameters,

[0068] -Generate a digital twin by using the tuning module to perform the steps of selecting, configuring and combining digital twin modules of the digital twin modular system according to the tuning module parameters.

[0069] The invention also relates to a system for performing the method according to at least some of the aforementioned aspects. According to the invention, the system comprises a digital twin modular system, characterized in that the digital twin modules are adapted or adaptable to digital twin characteristics related to computing characteristics, simulation scope, model architecture, communication security, accessibility, data storage, encryption functions, resource allocation and hardware requirements.

[0070] In addition, the system includes a tuning module, wherein the tuning module includes a tuning module parameter set, wherein the tuning module parameter set includes tuning module parameters corresponding to at least one digital twin feature respectively, and wherein the tuning module is prepared to generate a digital twin by performing steps of selecting, configuring and combining digital twin modules of the digital twin modular system according to the tuning module parameters.

[0071] According to another preferred embodiment of the system, the digital twin characteristics are at least one of the following: calculation speed, required computer performance, calculation accuracy, hardware operating range of the digital twin, underlying physical model of the physics considered, number of input and output parameters, selection of input parameters and output parameters, security features against hacker attacks, security features against manipulation of digital twin tuning parameters, decentralization / cloudification / distributed system remote accessibility, protected communication / interaction, step-by-step monitoring / step-by-step sharing of information.

[0072] According to another preferred embodiment of the system, the tuning module comprises a prediction module, wherein the tuning module comprises an information interface for providing information to a user, wherein the information interface is prepared to provide the user with at least a change trend indicator of at least a second tuning module parameter when changing at least a first tuning module parameter.

[0073] According to another preferred embodiment of the system, the set of tuning module parameters comprises fixed tuning module parameters and free tuning module parameters, wherein the prediction module is designed to derive estimates of at least some of the free tuning module parameter allocations from the fixed tuning module parameters.

[0074] According to another preferred embodiment of the system, said user interface input comprises assigning at least two tuning module parameters to one of the classifiers:

[0075] -Fixed tuning module parameters, in other words: no change in tuning module parameters,

[0076] - Free tuning module parameters, in other words: the change of possible tuning module parameters during the prediction of free parameters by the prediction module and the generation of the digital twin is done accordingly based on the adjusted set of free tuning module parameters.

[0077] In such an embodiment, the tuning module is capable of generating a digital twin including a digital twin feature having at least some of the tuning module parameters classified as varying free tuning module parameters and all of the tuning module parameters classified as unchanged fixed tuning module parameters when compared to set values ​​of the tuning module parameters provided in the tuning module prior to generating the digital twin feature.

[0078] According to another preferred embodiment of the system, said prediction module can comprise a neural network trained to derive estimates of said free tuning module parameters from said given tuning module parameters.

[0079] The present invention also relates to a computer-readable medium encoded with executable instructions which, when executed, cause a system as described above, in particular a computer system, to perform a method according to the present invention or one of its preferred embodiments as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Embodiments of the present invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0081] Figure 1 A flow chart of the method according to the present invention is shown;

[0082] Figure 2 shows a schematic diagram illustrating optional details of a system including a tuning module according to the present invention;

[0083] Figure 3 , Figure 4 Schematic diagrams illustrating embodiments of the present invention are shown respectively, which illustrate possible details of the tuning module;

[0084] The illustrations in the drawings are in schematic form. It is noted that in different drawings, similar or identical elements can have the same reference signs. DETAILED DESCRIPTION

[0085] Figure 1 A simplified flow chart illustrating a computer-implemented method and a system SYS for performing the method of generating a digital twin DTW is shown. The following steps are performed in Figure 1 The simplified illustration is shown in FIG. Step a here involves providing a digital twin modular system DTS. The digital twin modular system DTS is characterized in that the digital twin module DTM is suitable for or can be adapted to the digital twin feature DTF. The digital twin feature DTF involves at least one of the calculation property CPR, simulation scope SCS, model architecture MAT, communication security CMS, accessibility ACB, data storage DST, encryption function ECF, resource allocation RAL, and hardware requirement HRQ.

[0086] Furthermore, the method comprises a step b, which involves providing a tuning module TMD, wherein the tuning module TMD comprises a tuning module parameter set TMS. The tuning module parameter set TMS comprises tuning module parameters TMP respectively corresponding to at least one digital twin feature DTF.

[0087] Furthermore, the method comprises a step c, which involves generating a digital twin DTW by using the tuning module TMD, and performing the steps of selecting, configuring and combining digital twin modules DTM of the digital twin modular system DTS according to the tuning module parameters TMP.

[0088] like Figure 1 As shown, the digital twin DTW can include at least one input interface IPI and at least one output interface OPI, wherein the digital twin DTW includes at least one functional module SLV that generates an output OTP from an input INP. The digital twin feature DTF is part of at least one of the digital twin feature groups DFG: computational properties CPR, simulation scope SCS, model architecture MAT, communication security CMS, accessibility ACB, data storage DST, encryption function ECF, resource allocation RAL and hardware requirements HRQ. The digital twin feature DTF is at least one of the following: calculation speed CCS, required computer performance CPP, calculation accuracy ACC, hardware operating range HOR of digital twin, basic physics / empirical model PHM, complexity of physics considering / ignoring effects CNP, number of input parameters and / or selection of input parameters SIP, number of output parameters and / or selection of output parameters SOP, security feature against hacker attacks SHA, security feature against manipulation of digital twin tuning parameters STP, decentralization / cloudification / distributed system DSD, remote accessibility RAC, protected communication and / or interaction PTC, progressive monitoring and / or progressive information sharing GRM.

[0089] In order to enable a user to interact with the method according to the invention, the tuning module TMD can comprise an information interface IIF for providing information to the user USR, the information comprising actual set values ​​of at least two tuning module parameters TMP. Therefore, the method can comprise the additional step of providing information to the user via the information interface IIF, wherein the information can comprise actual set values ​​of at least two tuning module parameters TMP.

[0090] In order to better and reasonably fix the input of the tuning parameters, the tuning module TMD can include a prediction module PDM. The method includes:

[0091] - predicting at least one variation trend of at least one second tuning module parameter TMP when changing at least one first tuning module parameter TMP,

[0092] - providing at least the change trend to the information interface IIF as a change trend indicator TCI of at least the second tuning module parameter TMP when changing at least the first tuning module parameter TMP. The user can set parameters according to the predicted values ​​of other parameters to obtain the best fit parameterization of the digital twin characteristics as a basis for generating the digital twin DTW.

[0093] To make the process transparent to the user USR, the method comprises the following steps:

[0094] - assigning some of said tuning module parameters TMP to a fixed set of tuning module parameters TPG, and

[0095] - assigning other of the tuning module parameters TMP to a set of free tuning module parameters TPF and tuning module parameters TPF, wherein the prediction module PDM is designed to derive estimates of the free tuning module parameters TPF from the fixed tuning module parameters TMG.

[0096] The user can enter an interface input UIN through the user interface UIF, which input comprises assigning at least two of said tuning module parameters TMP to one of these classifiers:

[0097] - Fixed tuning module parameters TPG,

[0098] - Free tuning of module parameters TPF.

[0099] The tuning module parameter TPG may then be determined as the tuning module parameter TPG for the tuning module parameter TPF assigned as the free tuning module parameter TPF,

[0100] - Generating a digital twin DTW by using the tuning module TMD. Based on this, the steps of selecting, configuring and combining digital twin modules DTM of the digital twin modular system DTS can be performed according to the tuning module parameters TMP.

[0101] According to the present invention, the method as shown in the figure can be stored on a computer readable medium. The computer readable medium can be encoded with executable instructions, which, when executed, cause the system SYS, in particular a computer system, to perform the method according to the present invention or at least one preferred embodiment thereof.

[0102] Figure 3 and Figure 4 A possible implementation of a tuning module for a digital twin is schematically shown.

[0103] Figure 3A digital twin is shown which can be provided as a functional simulation unit FMU which is part of a simulation SIM running on a computer CPU. A tuning module TMD can be part of a digital twin which contains a plurality of tuning module parameters TMP of a tuning module parameter set TMS corresponding respectively to at least one digital twin feature DTF. The tuning module user interface UIN enables influencing the tuning module parameters TMP via a user interface input UIN to change the digital twin behavior. The tuning module parameter set TMS and the tuning module parameters TMP can typically be of a progressive (e.g. all possible values) or binary (e.g. yes / no or 0 / 1 or true / fals...) nature. Depending on the tuning module parameter TMP settings, a digital twin with specific features or characteristics listed and explained above is generated.

[0104] The digital twin DT can be distributed (arranged to be distributed or located) on multiple instances and can, for example, include communication with a cloud CLD to a certain extent specified by one of the tuning module parameters TMP. The settings of the digital twin by the tuning module TMD affect the input INP and output OPT of the digital twin DTW. Furthermore, the exchange of data or information between the digital twin and the provider PRV can be limited to the tuning module parameters TMP.

[0105] Figure 4 An embodiment is shown, in which, before performing the digital twin operation OPR in a second step 2), the digital twin tuning preparation PRP can be performed separately in a separate step 1). The preparation stage also shows that the tuning module can be a module of the digital twin DTW, or can be an external module that can be operated on the same central processing unit CPU and configured by the tuning module TMD based on the user input UIN. As shown in step 2, the operation OPR is performed without the tuning module TMD being configured by the tuning module interface. According to this embodiment, during the operation OPR, the tuning module TMD is not part of the digital twin DTW.

[0106] Although the invention has been described in detail with reference to preferred embodiments, it should be understood that the invention is not limited to the disclosed examples but is only limited by the scope of the independent claims and that those skilled in the art will be able to make additional modifications and variations thereto without departing from the scope of the invention, but these modifications and variations are different from the examples.

[0107] It should be noted that the use of "a" or "an" throughout this application does not exclude a plurality, and "comprising" does not exclude other steps or elements. Furthermore, elements described in connection with different embodiments can be combined. It should also be noted that reference signs in the claims should not be interpreted as limiting the scope of the claims.

Claims

1. A computer-implemented method for generating a digital twin, the method include: Providing a digital twin modular system, The digital twin modular system is characterized in that the digital twin module is suitable for the digital twin feature, The digital twin features relate to at least one of the following: computational properties, simulation scope, model architecture, communication security, accessibility, data storage, encryption capabilities, resource allocation, hardware requirements, Provide tuning module, Wherein, the tuning module includes a tuning module parameter set, The tuning module parameter set includes tuning module parameters corresponding to at least one of the digital twin features, respectively. receiving an interface input through a user interface, the interface input comprising assigning some of the tuning module parameters to a set of fixed tuning module parameters and assigning other of the tuning module parameters to a set of free tuning module parameters, Determine tuning module parameters for free tuning module parameters, and generate a digital twin by selecting, configuring and combining digital twin modules of the digital twin modular system based at least in part on the tuning module parameters and the interface input to increase the adaptability of the digital twin in different environments.

2. The method according to claim 1, in, The digital twin comprises at least one input interface and at least one output interface, wherein the digital twin comprises at least one functional module that generates output from input.

3. The method according to claim 1 or 2, in, The digital twin feature is part of at least one of a group of digital twin features: - computed properties, - simulation range, - Model architecture, - Communication security, - Accessibility, - data storage, - Encryption function, - Resource allocation and -Hardware requirements.

4. The method according to claim 3, in, The digital twin feature is at least one of the following: - calculation speed, - Required computer performance, - calculation accuracy, - the hardware operating scope of the digital twin, - basic physics / empirical models, - consider / ignore the physical complexity of the effects, - the number of input parameters and / or the selection of input parameters, - the number of output parameters and / or the selection of output parameters, - Security features to prevent hacker attacks, - Safety features to prevent manipulation of digital twin tuning parameters, -Decentralization / Cloud / Distributed Systems, - Remote accessibility, - protected communications and / or interactions, -Progressive monitoring and / or progressive information sharing.

5. The method according to claim 1 or 2, in, The tuning module comprises an information interface for providing information to a user, the information comprising actual setting values ​​of at least two tuning module parameters, the method comprising: - providing information comprising said actual set values ​​of at least two tuning module parameters to a user via said information interface.

6. The method according to claim 1 or 2, in, The tuning module includes a prediction module, and the method includes: - predicting at least one variation trend of at least one second tuning module parameter when changing at least one first tuning module parameter, - when changing at least said first tuning module parameter, providing at least said variation trend to an information interface as a variation trend indicator of at least said second tuning module parameter.

7. A method according to at least the preceding claim 6, in, The prediction module is designed to derive estimates of the free tuning module parameters from the fixed tuning module parameters.

8. A system for performing the method according to claim 1, The system includes a digital twin modular system, The digital twin modular system is characterized in that the digital twin modules are adapted to digital twin characteristics related to computational properties, simulation scope, model architecture, communication security, accessibility, data storage, encryption capabilities, resource allocation, and hardware requirements, The system includes a user interface configured to receive an interface input, the interface input including assigning at least two tuning module parameters to one of the classifiers including fixed tuning module parameters, in, determining tuning block parameters for the tuning block parameters assigned as free tuning block parameters, and The system includes a tuning module, Wherein, the tuning module includes a tuning module parameter set, The tuning module parameter set includes tuning module parameters corresponding to at least one digital twin feature, respectively. Among them, the tuning module is prepared to generate a digital twin by executing the steps of selecting, configuring and combining the digital twin modules of the digital twin modular system according to the tuning module parameters and interface inputs to increase the adaptability of the digital twin in different environments.

9. The system according to claim 8, in, The digital twin feature is at least one of the following: - calculation speed, - Required computer performance, - calculation accuracy, - the hardware operating scope of the digital twin, - basic physics / empirical models, - consider / ignore the physical complexity of the effects, - the number of input parameters and / or the selection of input parameters, - the number of output parameters and / or the selection of output parameters, - Security features to prevent hacker attacks, - Safety features to prevent manipulation of digital twin tuning parameters, -Decentralization / Cloud / Distributed Systems, - Remote accessibility, - protected communications and / or interactions, -Progressive monitoring and / or progressive information sharing.

10. A system according to claim 8 or 9, in, The tuning module comprises a prediction module, wherein the tuning module comprises an information interface for providing information to a user, wherein the information interface is configured to provide the user with at least a change trend indicator of at least a second tuning module parameter when changing at least a first tuning module parameter.

11. The system according to claim 8 or 9, in, The prediction module is designed to derive estimates of at least some of the free tuning module parameter assignments from the fixed tuning module parameters.

12. System according to at least the preceding claim 11, in, The prediction module includes a neural network that is trained to derive estimates of the free tuning module parameters given the tuning module parameters.

13. A computer readable medium encoded with executable instructions which, when executed, cause the system according to claim 8 to perform the method according to claim 1.