A System and Method for Generating a Holistic Digital Twin

Abstract

A system and method for generating a holistic digital twin of an industrial facility that includes a plurality of assets, wherein the said system includes conversion units configured to convert asset related data collected from different tools utilized to plan and/or operate the industrial facility in a tool specific data format and asset related data provided by data sources of the industrial facility in a data source specific data format into a common graphical representation, a matching unit configured to match the common graphical representations of the converted asset related data to provide a mapping between the assets of the industrial facility, and a merging unit configured to merge the mapped assets of the industrial facility into a unified graph to provide the holistic digital twin of the industrial facility.

Claims

1.-14. (Cancelled)

15. A computer-implemented method for providing a holistic digital twin of an industrial facility comprising a plurality of assets, the method comprising: (a) converting asset related data collected from different tools utilized to at least one of plan and operate said industrial facility in a tool specific data format and asset related data provided by data sources of the industrial facility in a data source specific data format into a common graphical representation; (b) matching common graphical representations of the converted asset related data to provide a mapping between assets of said industrial facility; and (c) merging mapped assets of said industrial facility into a unified graph to provide the holistic digital twin of said industrial facility; wherein the common graphical representation of converted asset related data comprises for each asset a node connected via edges to other nodes representing other assets of said industrial facility having a physical or logical relation with the respective asset.

16. The computer-implemented method according to claim 15, wherein the assets of said industrial facility comprise hardware components and software components installed in said industrial facility.

17. The computer-implemented method according to claim 15, wherein each asset of said industrial facility comprises an associated unique asset identifier.

18. The computer-implemented method according to claim 16, wherein each asset of said industrial facility comprises an associated unique asset identifier.

19. The computer-implemented method according to claim 15, wherein the common graphical representations generated by the conversions of the asset related data are stored as unified data in a central storage.

20. The computer-implemented method according to claim 15, wherein the matching of the common graphical representations is performed by a graph matching algorithm.

21. The computer-implemented method according to claim 15, wherein at least one of ambiguities and mismatches occurring during the matching of the common graphical representations are resolved in response to a user input.

22. The computer-implemented method according to claim 15, wherein at least one of ambiguities and mismatches occurring during matching of the common graphical representations are resolved automatically based on received asset related data concerning assets affected by observed at least one of ambiguities and mismatches triggered by the graph matching algorithm in response to detected at least one of ambiguities and mismatches.

23. The computer-implemented method according to claim 20, wherein the graph matching algorithm is machine learned.

24. The computer-implemented method according to claim 15, wherein the different tools providing the asset related data comprise tools of different lifecycle stages of said industrial facility.

25. The computer implemented method according to claim 24, wherein the tools of the different lifecycle stages of said industrial facility comprise at least one of engineering tools, operation management tools and service and maintenance tools.

26. The computer-implemented method according to claim 15, wherein the generated holistic digital twin is fed back to the tools utilized to at least one of plan and operate said industrial facility to at least one of upgrade the respective tools and automatically establish crosslinks between the different tools.

27. The computer-implemented method according to claim 15, wherein the generated holistic digital twin of the industrial facility is processed to at least one of simulate and predict an operational behavior of said industrial facility.

28. The computer-implemented method according to claim 15, wherein data sources of the industrial facility providing asset related data in a data source specific data format comprise at least one of sensor assets and memory assets of said industrial facility.

29. The computer-implemented method according to claim 15, wherein asset related data provided in a tool specific data format of a tool comprise at least one of image data representing the assets, acoustic data of sounds generated by the assets, text data describing the assets, graphical data, topological data of an asset topology and location data of asset locations.

30. A system for generating a holistic digital twin of an industrial facility which comprises a plurality of assets, said system comprising: (a) conversion units configured to convert asset related data collected from different tools utilized to at least one plan and operate said industrial facility in a tool specific data format and asset related data provided by data sources of the industrial facility in a data source specific data format into a common graphical representation; (b) a matching unit configured to match the graph representations of the converted asset related data to provide a mapping between assets of said industrial facility; and (c) a merging unit configured to merge mapped assets of said industrial facility into a unified graph to provide the holistic digital twin of said industrial facility; wherein the common graphical representation of converted asset related data comprises for each asset a node connected via edges to other nodes representing other assets of said industrial facility having a physical or logical relation with the respective asset.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In the following, possible embodiments of the different aspects of the present invention are described in more detail with reference to the enclosed figures, in which:

[0027] FIG. 1 shows an exemplary embodiment of a system used for generation of a holistic digital twin of an industrial facility in accordance with the present invention;

[0028] FIG. 2 shows an exemplary flowchart of a computer-implemented method for providing a holistic digital twin of an industrial facility in accordance with the present invention;

[0029] FIGS. 3, 4, 5 show a diagram for illustrating an exemplary use case where a computer-implemented method has been performed to provide a holistic digital twin of an industrial facility in accordance with.

DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS

[0030] FIG. 1 schematically shows a possible architecture of a system 1 in accordance withthe present invention. The technical system 1 illustrated in FIG. 1 is used to generate a holistic digital twin of an industrial facility IF that comprises a plurality of assets 4. These assets 4 can include hardware components, in particular machine components and software components installed in the controllers of the industrial facility IF. In a preferred embodiment, each asset of the industrial facility IE comprises an associated unique asset identifier. This unique asset identifier can be used across different software tools 2 used in different lifecycle stages of the industrial facility IE. In the illustrated exemplary embodiment, different software tools 2-1, 2-2, 2-3 . . . 2-N can be used to perform automated functions related to the planning, operation and maintenance of an industrial facility IF. These software tools 2 can, for instance, comprise engineering tools such as layout tools (e.g., EPLAN), programmable lofic controller (PLC) programming tools (such as TIA) or a computer-aided design tool such as NX. A number N of different software tools 2 may be available for performing automated functions during the different lifecycle stages of the industrial facility IF. In a possible embodiment, I different software tools 2 can be stored in a database of the system 1. A software tool 2 can be loaded and supplied to an associated conversion unit 3-i, as illustrated in FIG. 1. The conversion units 3-i are configured to convert asset related data provided by different tools 2-i in a tool specific data format and used to plan and/or operate the industrial facility into a common graph representation. The industrial facility can be a complex industrial facility IF comprising a plurality of components or assets 4 generating asset related data. In the illustrated schematic diagram of FIG. 1, a physical industrial facility IF can comprise machines or other hardware components 4 including data acquisition units 5 that provide collected asset related data. The data acquisition units 5 can comprise sensors integrated or attached to a physical asset 4 of the industrial facility. The sensors can collect sensor data from physical components of the industrial facility IF including different production lines and/or manufacturing machines. The sensors can also form hardware assets of the industrial facility IF. The collected data provided by the data acquisition units 5 can comprise an asset identifier of a physical asset or component. This asset identifier can comprise, for instance, a scanned QR code or an RFID. The asset identifier can be connected or related to a semantic description of the respective asset. For instance, the asset identifier can indicate a type of the respective asset. The collected data can further comprise location information concerning the location of the physical asset 4. The location information can be provided in a possible embodiment by an indoor localization system. Further, the location data can be provided by AGVs that can report their position in a shop floor of the industrial facility IF. As can be seen in FIG. 1, the sensors provide, as data sources of the industrial facility IF asset, related data in a data source specific data format. The system 1 in accordance with the present invention comprises conversion units 6 to convert asset related data provided by the data sources or sensors of the industrial facility IF having a data source specific data format into a common graph representation. Accordingly, in the system 1 of the present invention, there are conversion units 3 for conversion of asset related data originating from software tools 2 and conversion units 6 used to convert asset related data provided by physical data sources of the physical industrial facility IF. The different conversion units 3, 6 convert the received asset related data into common graph representations. In a possible embodiment, the common graphical representations of converted asset related data can comprise, for each asset, a node connected via edges to other nodes representing other assets of the same industrial facility IF having a physical relation (e.g., distance between assets) or logical relation (e.g., forms part of) with the respective asset. Common graphical representations generated by the conversions of the asset related data are stored as unified data in a central storage 7 of the system 1, as illustrated in FIG. 1. In a possible embodiment, the central storage 7 can comprise a knowledge graph database. Each asset of the industrial facility IF including physical assets (such as devices) and logical assets (such as information or datasets) can be represented by a node in the common graph representation. Each node of the common graph representation can also contain properties describing, for instance, the type of the respective asset and its potential configuration.

[0031] The system 1 comprises a processing unit 8 configured to process graphical representations stored in the central storage 7 of the system 1. The processing unit 8 can include a matching unit configured to match the graphical representations of the converted asset related data to provide a mapping between the assets of the industrial facility. The processing unit 8 can further comprise a merging unit configured to merge the mapped assets of the industrial facility into a unified graph to provide the holistic digital twin of the industrial facility IF. The holistic digital twin of the industrial facility can then be stored in the central storage 7 of the system 1 for further processing. In a possible embodiment, the matching of the graph representations is performed by a processor of the processing unit 8 using a graph matching algorithm. The graph matching algorithm can be used to connect different views or graphical representations collected and stored in the central data storage 7 of the system 1. The matching algorithm can use both topological information encoded in the graph as well as type information of the nodes and edges. In a possible embodiment, to facilitate the matching, type information can follow a hierarchy that can be exploited. For example, a SIMATIC S7-15PLC with a SIMATIC PLC which is a PLC which, in turn, is an asset of the industrial facility IF.

[0032] FIGS. 3, 4, 5 illustrate different steps performed by the processing unit 8. Information from different tools 2 such as EPLAN, TIA and/or partitional information from performed scans are illustrated in FIG. 3. The system 1 is aware that an S7-1515 is a PLC and that an ET 200SP E/O is an I/O and that a switch 5811 is a switch and an input. In a possible embodiment, the matching algorithm may check that the network topology is compatible. From this, the mapping result as illustrated in FIG. 4 can be inferred. The unified graph can be generated by merging as illustrated in FIG. 5.

[0033] The merging unit can be configured to merge the mapped assets as shown in FIG. 4 into a unified graph to provide the holistic digital twin HDT of the industrial facility IF as illustrated in FIG. 5. In a possible embodiment, ambiguities and/or mismatches occurring during the matching of the graph representations can be resolved in response to a user input. In a possible embodiment, the system 1 can comprise a user interface 9 to receive a user input from a user U processed to resolve ambiguities and/or mismatches occurring during the matching of the graphical representations by the graph matching algorithm. In a possible embodiment, the system 1 can ask a user U to identify assets and/or to verify a specific connection between assets of the industrial facility. With this information, the matching algorithm can improve the matching process. The user interface 9 can also be used to receive user input processed to resolve conflicts detected by the matching algorithm. For example, the data from an EPLAN software tool may be contradicting the data from a TIA Portal and the user input is processed to resolve the detected mismatch. Machine learning (ML) techniques can be used to improve a search heuristic of the matching algorithm by incorporating successful data merges in the past. In a possible embodiment, the holistic view or holistic digital twin HDT can be fed back to the original data sources, in particular to the different software tools 2 (as illustrated by a dashed line in FIG. 1. This feedback allows the respective software tools 2 to be upgraded. Further, the feedback can be used to automatically establish crosslinks between the different software tools 2 such as EPLAN, TIA or NX. This in turn can improve the parallel handling of different software tools 2-i by an operator or user exploiting automatically established crosslinks between the different software tools 2-i. Accordingly, a user U can jump from a certain function of a first software tool directly via an established crosslink to a second function of the other software tool. As can be seen in FIG. 1, asset related data is collected from the different tools 2-i and converted automatically by associated converters 3 to provide a common graphical representation. With the system 1 in accordance with the present invention, not only data from software tools 2-i is collected but also data from physical data sources or sensors installed in the industrial facility IF. Data provided by the data sources of the industrial facility IF in a data source specific format are also converted automatically into the common graphical representation of the system 1. The converted data output by the converters 3-i and 6 are stored in the central storage 7. Then, a graph matching algorithm is run on a processor of the processing unit 8. In cases that an error occurs due to conflicts or ambiguity, a user U can be asked for more information via the user interface 9 of the system 1. Finally, the matched and merged data can be fed back to the original data sources, in particular software tools 2. If the process is repeated many times, this data of multiple matchings can be used for deriving an improved search heuristic by machine learning ML. The common graphical representation is used to automate the merging of several different heterogeneous data sources including both software data sources (software tools) but also physical data sources installed in the industrial facility IF itself. The system 1 unifies and completes otherwise fragmented information of the industrial facility IF into a holistic digital twin HDT of the industrial facility IF representing the industrial facility IF as a whole over different lifecycle stages. The holistic digital twin HDT provides a unified view of the industrial facility IF that can be processed to plan and/or execute maintenance, perform a re-engineering of components of the industrial facility and/or to provide extensions of the industrial facility IF. Further, the holistic digital twin HDT can be processed for detection of faults occurring within the industrial facility IF. The conversion mechanisms can provide a translation of different data formats into a unified graphical format of the system 1.

[0034] The generated holistic digital twin HDT of the industrial facility IF can be further processed to simulate and/or to predict an operational behavior of the industrial facility. The data sources of the industrial facility IF providing asset related data in a data source specific data format can comprise both sensor assets of the industrial facility IF but also memory assets of the industrial facility IF, e.g., local memories of controllers installed in the industrial facility IF. Asset related data provided by a tool specific data format from a software tool 2 can comprise different kinds of data in heterogeneous data formats. Asset related data provided by tool specific data format by a software tool 2 can include image data representing assets of the industrial facility IF. Asset related data can further comprise acoustic data of sounds generated by the assets 4. Asset related data can further comprise text data describing assets 4 of the industrial facility IF. Moreover, the asset related data can comprise graphical data and/or topological data of an asset topology. Further, the asset related data can comprise location data of asset locations.

[0035] FIG. 2 shows a flowchart of a possible exemplary embodiment of a computer-implemented method for providing a holistic digital twin HDT of an industrial facility IF including a plurality of different software and/or hardware components or assets. In the illustrated exemplary embodiment, the computer-implemented method comprises three main steps.

[0036] In a first step S1, asset related data collected from different tools used for planning and/or operation of an industrial facility IF in a tool specific data format as well as asset related data provided by data sources of the industrial facility IF in a data source specific data format are converted into a common graphical representation.

[0037] In a further step S2, the common graphical representations of the converted asset related data are matched to provide a mapping between the assets of the industrial facility as also illustrated in FIG. 4.

[0038] In a further step S3, the mapped assets of the industrial facility are merged into a unified graph as illustrated in FIG. 5 to provide the holistic digital twin HDT of the industrial facility. The holistic digital twin HDT can be stored and used for further processing. Further, the holistic digital twin HDT can be fed back to the tools 2 to upgrade the respective tools 2 and to automatically establish crosslinks between the different tools 2. Accordingly, the feedback can be exploited to improve the quality of the different software tools 2 and to improve the parallel handling of different software tools 2 by a user.

[0039] In a possible embodiment, the computer-implemented method as illustrated in FIG. 2 can be implemented on a web server of a cloud platform that can be connected via a data network, such as an Internet, to several industrial facilities IFs at different customer sites. The conversion of the asset related data by the converters 6 can either occur at the customer site and/or at the cloud platform. In a possible embodiment, the created holistic digital twin HDT, such as illustrated in FIG. 5, can be displayed on a display unit of the graphical user interface 9 of the system 1 as illustrated in FIG. 1 but also on mobile user terminals at the site of the industrial facility IF where hardware components and/or machines 4 are located to assist operators, for instance, in performing maintenance activities at the location of the industrial facility IF. The transferred current holistic digital twin HDT may be provided to different users U performing different functions concerning the same industrial facility IF in parallel, i.e., at the same time. For instance, an application executed on a processor of a mobile handheld user equipment may process the holistic digital twin HDT to assist a user in performing a maintenance or a repair activity at the physical customer site of the industrial facility. At the same time, the holistic digital twin HDT generated by the computer-implemented method in accordance with the disclosed embodiments of the present invention can be processed to re-engineer the industrial facility IF and/or to simulate and/or to predict an operation behavior of the industrial facility IF. Accordingly, different software tools and/or applications can have access to the holistic digital twin HDT of the industrial facility to assist different users in performing different functions over the lifecycle stages of the same industrial facility.

[0040] The holistic digital twin HDT can be further processed to automatically derive lifecycle specific digital twins used for different lifecycle stages of the industrial facility IF, such as a digital product twin, a digital production twin and/or a digital performance twin. The digital product twin can be used during product development of a product. The digital production twin can be used during manufacturing engineering and/or during production operations. The digital performance twin can be used for product simulation and during the utilization life of the manufactured product. Accordingly, the generated complex holistic digital twin HDT of the industrial facility IFcan be processed to derive automatically partial views for different lifecycle stages of the industrial facility IF. The derived digital twins can be distributed to associated user S participating in the system 1 according to the present invention.

[0041] A derived digital twin requires less memory space than the complex holistic digital twin HDT of the industrial facility IF used for all life cycle stages of the industrial facility IF. Consequently, user terminals with limited memory resources may operate on a received derived digital twin instead on the HDT.

[0042] Thus, while there have been shown, described and pointed out fundamental novel features of the invention as applied to a preferred embodiment thereof, it will be understood that various omissions and substitutions and changes in the form and details of the methods described and the devices illustrated, and in their operation, may be made by those skilled in the art without departing from the spirit of the invention. For example, it is expressly intended that all combinations of those elements and/or method steps which perform substantially the same function in substantially the same way to achieve the same results are within the scope of the invention. Moreover, it should be recognized that structures and/or elements and/or method steps shown and/or described in connection with any disclosed form or embodiment of the invention may be incorporated in any other disclosed or described or suggested form or embodiment as a general matter of design choice. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto.