Digital twin management system and method

Abstract

To provide a digital twin control system in which usability of a user can be enhanced.SOLUTION: A digital twin control system 1 controls a virtual model which reproduces a real system in a physical space in real time in a virtual space. When a processor of the digital twin control system generates an integrated virtual model by adding a second virtual model to a first virtual model (10), the processor extracts a plurality of sections that can be made common from among the first virtual model and the second virtual model (11), generates a plurality of integrated virtual models being candidates of the integrated virtual models by changing the extracted sections that can be made common (12), calculates an evaluation of each of the generated candidates of the integrated virtual models (13), and outputs configuration information of each of the candidates of the integrated virtual models and evaluation of each of the integrated virtual models (15).SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a digital twin management system and method. [Background technology]

[0002] CPS (Cyber ​​Physical System) is being considered, which represents real-world equipment, processes, plants, systems, etc. using digital twins in the cyber (digital) world and utilizes them for understanding phenomena, prediction, optimization, etc. Digital twins predict and verify the behavior of equipment, etc. in the physical world using simulations and various analytical methods in a digital space. In order to facilitate the configuration of this digital twin, methods have been proposed for efficiently configuring it according to the target equipment, etc. (Patent Documents 1 and 2).

[0003] Although not related to digital twins, a technology is known that, in order to improve the efficiency of machine learning processing, executes processing that reduces overlapping portions when multiple patterns of processing that differ only in parameters are given (Patent Document 3). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] U.S. Patent No. 10,274,915 [Patent Document 2] US Patent Application Publication No. 2014 / 0019104 [Patent Document 3] JP 2012-160014 A Summary of the Invention [Problem to be solved by the invention]

[0005] When digital twins use simulations and various analytical methods in a digital space to predict performance, identify faults, and verify the behavior of equipment and other things in the physical world, models, analysis models, etc. (hereinafter referred to as models) may be added to or updated by applying new analytical methods or adding information from new sensors, etc.

[0006] For example, in an environment where there is a digital twin environment (sensors, configuration and data management, models) that estimates the remaining lifespan of a machine tool motor, a new function to estimate the wear degree of a drill, which is a consumable item of the same machine tool, is added.

[0007] In this case, it takes time for users such as system administrators to determine whether the additions or updates are compatible with the existing digital twin environment and to understand the scope of the changes or impacts for each site.

[0008] Conventional technology discloses a method for dynamically constructing new digital twins, but does not disclose a configuration method that takes into account the impact on existing digital twin environments, making it difficult for users to use.

[0009] The present invention has been made in consideration of the above-mentioned problems, and its purpose is to provide a digital twin management system and method that can improve usability for users. [Means for solving the problem]

[0010] In order to solve the above problems, according to one aspect of the present invention, there is provided a digital twin management system for managing a digital twin that reproduces a real system in a physical space in a virtual space, the digital twin management system having a processor and a memory for storing a predetermined computer program executed by the processor, and when generating an integrated scenario by adding a second scenario to a first scenario that constitutes the digital twin, the processor extracts multiple common parts from the first scenario and the second scenario, generates multiple integrated scenarios that are candidates for the integrated scenario by changing the extracted common parts, calculates an evaluation of each of the generated integrated scenario candidates, and outputs configuration information of each integrated scenario candidate in association with the evaluation of each integrated scenario. Effect of the Invention

[0011] According to the present invention, it is possible to output configuration information and an evaluation of a candidate integrated scenario that adds a second scenario to a first scenario that constitutes a digital twin, thereby allowing a user to determine the impact of updating the digital twin based on the output configuration information and evaluation, thereby improving usability for the user. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram of a digital twin management unit according to this embodiment. [Diagram 2] A diagram of the digital twin management system and IoT management system. [Diagram 3] Hardware configuration diagram of the computers used in the digital twin management system. [Figure 4] An example of an equipment management database. [Diagram 5] An example of a digital twin configuration database. [Figure 6] An example of an analysis model database. [Figure 7] 1 is an example of a grouping policy database. [Figure 8] Example of a Datastore Management Database. [Figure 9]An example scenario template database. [Figure 10] An example of a scenario application plan database. [Figure 11] An example of a scenario instance database. [Figure 12] Flowchart of digital twin management process. [Figure 13] A templated process flow included in a scenario for estimating the remaining life of a motor. [Figure 14] A templated process flow included in a scenario for estimating the remaining life of a drill. [Figure 15] FIG. 1 is an explanatory diagram showing an example of an integrated scenario candidate (updated digital twin candidate). [Figure 16] FIG. 13 is an explanatory diagram showing another example of an integrated scenario candidate. [Figure 17] 13 is an example of a screen that provides information on candidate integration scenarios to the user. [Figure 18] 18 is a detailed example of a portion of the screen in FIG. 17. [Figure 19] 13 is a flowchart of a digital twin management process in the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The digital twin management system according to this embodiment has a configuration, data, data processing, behavior model, fault model, and processing flow as scenario templates in a system that predicts or analyzes life spans, failures, etc.

[0014] The digital twin management system of this embodiment can also be called a system that manages digital twins. Digital twins are a technology that uses sensors to collect the operating status of devices or equipment, which are real systems in the physical space, and reproduce them as a model in a virtual space based on the collected information. The digital twins generated in the virtual space are linked to the real systems, and can be used, for example, to calculate remaining life, diagnose failures, determine faults, and analyze performance.

[0015] When creating an integrated scenario by adding a new scenario template to a running digital twin environment, the digital twin management system of this embodiment compares the configuration, data, data processing, behavior model, fault model, and analysis flow of the existing environment with the scenario template. As a result, the digital twin management system creates multiple candidate plans that allow for partial sharing of the configuration, calculates the impact (evaluation) of these candidate plans, and presents them to the user.

[0016] As a result, according to this embodiment, the user can objectively select a candidate plan that the user judges to be optimal based on the information and evaluation of each candidate plan, improving usability for the user. Furthermore, according to this embodiment, even a user with little experience can appropriately manage digital twins. EXAMPLES

[0017] The first embodiment will be described with reference to Figures 1 to 18. Figure 1 is a schematic diagram of a digital twin management unit 120, which is one component of a digital twin management system 100.

[0018] The digital twin management unit 120 includes, for example, a scenario adding unit 10, a common part extracting unit 11, a candidate generating unit 12, an evaluating unit 13, a narrowing down unit 14, and an information presenting unit 15. Furthermore, the digital twin management unit 120 can also include a configuration change implementing unit 16. Furthermore, the digital twin management unit 120 may include a system notification unit 17.

[0019] In the digital twin management unit 120, when the scenario addition unit 10 adds a new scenario to an existing scenario, the common part extraction unit 11 compares the processing flow of the existing scenario with the processing flow of the new scenario to extract the parts that can be made common.

[0020] The candidate generator 12 generates multiple candidates for an integrated scenario (also called an integrated scenario or an updated scenario) by modifying the commonizable parts.

[0021] The evaluation unit 13 evaluates each integrated scenario candidate based on a predetermined evaluation index. Examples of the predetermined evaluation index include indexes that are directly related to the integrated scenario candidate, which is a component of the digital twin, such as the amount of computer resources used and the stability of performance.

[0022] In addition to or instead of these evaluation indices, indices that affect other systems that are linked to the real system (for example, industrial equipment) may be used. Examples of other systems that are linked to the real system include production management systems, maintenance management systems, and inventory management systems. Examples of cases in which updating a digital twin that is linked to a real system can affect other systems include changes in the number of products manufactured by the real system, changes in the inventory of maintenance parts required to maintain the real system, and fluctuations in the maintenance plan for the real system.

[0023] The narrowing down unit 14 selects one or more candidates to present to the user from among the multiple integrated scenario candidates generated by the candidate generation unit 12, based on the evaluation calculated by the evaluation unit 13. The narrowing down unit 14 may group the integrated scenario candidates based on a preset grouping policy, and select the most highly evaluated candidate from the group of grouped candidates. For example, multiple integrated scenario candidates may also be grouped based on policies such as resource sharing rate, rate of change in model accuracy, or amount of change in processing time.

[0024] The information presenter 15 presents the integrated scenario candidates selected by the narrowing down unit 14 to the user via a screen G1 displayed on the monitor display. The information presenter screen G1 includes, for example, a candidate display section GP11 and an evaluation display section GP12. As in the example described later in Fig. 17, the screen G1 may include an area for displaying an update summary.

[0025] The candidate display section GP11 displays a configuration overview of one or more integrated scenario candidates. The configuration overview includes, for example, a processing flow of the first scenario, a processing flow of the second scenario, and a processing flow (common layer processing flow) commonly used in both the first and second scenarios. The processing flow does not need to be displayed in detail, but only needs to be displayed to the extent that the user can understand the overview. Furthermore, a button for knowing the details may be provided on the screen G1, and the details of the processing flow may be displayed when the user operates the button. Here, the first scenario is a target scenario to be added to which another scenario (second scenario) is added. The second scenario is a target scenario to be added to another scenario (first scenario). In this embodiment, a case where one second scenario is added to one first scenario will be described as an example, but this is not limited to this, and the present invention can also be applied to a case where multiple second scenarios are added to one first scenario.

[0026] The evaluation display section GP12 displays an evaluation based on a predetermined evaluation index for each integrated scenario candidate. For example, the name that identifies each integrated scenario candidate and the evaluation value (e.g., the impact on performance, resource sharing rate, time required for update, etc.) are displayed in the evaluation display section GP12.

[0027] When a candidate integrated scenario is selected by the user, the configuration change implementation unit 16 updates the existing digital twin to the selected candidate integrated scenario.

[0028] When updating to an integrated scenario candidate, the system notification unit 17 transmits a predetermined notification to the real system (e.g., industrial equipment) and / or other systems that cooperate with the real system. The predetermined notification is, for example, information indicating the effect on the real system and information indicating the effect on other systems that cooperate with the real system.

[0029] The overall configuration of an IoT (Internet of Things) management system 1 according to this embodiment will be described with reference to Fig. 2. The IoT management system 1 includes, for example, a digital twin management system 100, a digital twin processing execution unit 200, industrial equipment 300, a MES (Manufacturing Execution System) 410, and an IoT gateway 420. The digital twin management system 100 and the digital twin processing execution unit 200 can be configured using a computer CS, which will be described later with reference to Fig. 3.

[0030] The digital twin management system 100 manages the status of each piece of industrial equipment 300 in a factory and the components within the industrial equipment 300. Details of the digital twin management system 100 will be described later.

[0031] The digital twin processing execution unit 200 executes digital twin processing. The digital twin processing execution unit 200 includes, for example, a data collection and processing unit 210, a scenario execution unit 220, and a data store 230. Hereinafter, the data collection and processing unit 210 may be abbreviated as the data collection unit 210.

[0032] The data collection unit 210 acquires data from the sensor 330 of each industrial device 300. The data acquired from the sensor 330 is stored in the data store 230. The scenario execution unit 220 executes predetermined processing such as estimating remaining life or detecting anomalies based on the data acquired from the sensor 330 and the processing flow.

[0033] As described above, the digital twin processing execution unit 200 receives and analyzes data from multiple industrial devices 300 in a factory, and outputs the analysis results. Considering the amount and time of communication between the digital twin processing execution unit 200 and each industrial device 300, it is preferable that the physical distance between the digital twin processing execution unit 200 and each industrial device 300 is short. Therefore, in this embodiment, the digital twin processing execution unit 200 is provided in the factory that has each industrial device 300, or in a building close to the factory.

[0034] Although there are advantages to arranging the digital twin processing execution unit 200 and each industrial device 300 close to each other as described above, this embodiment is not limited to this. The digital twin processing execution unit 200 may be installed physically far away from each industrial device 300. The digital twin processing execution unit 200 may be provided on the same computer as the digital twin management system 100.

[0035] The industrial equipment 300 is an example of a "real system" and is equipment used in a factory. The industrial equipment 300 includes, for example, a cutting machine, a press machine, a rolling machine, an injection molding machine, a conveying device, a drying furnace, a heating furnace, a reaction furnace, a stirring device, a centrifuge, a packaging machine, etc. In this embodiment, a cutting machine will be described as an example of the industrial equipment 300.

[0036] The actual system is not limited to industrial equipment, but may be, for example, a man-machine conveyor device such as an elevator or escalator, an automatic door, a blower, or other equipment used in offices, commercial facilities, or public facilities.

[0037] The industrial device 300 of this embodiment includes, for example, a motor 310, a drill 320, a sensor 330, a power source 340, and a controller (not shown). When power is supplied from the power source 340 as a "power source," the motor 310 as a "driving source" rotates, and the drill 320 as a "consumable" rotates to cut a workpiece.

[0038] The sensor 330 detects, for example, the number of rotations of the motor 310, the current of the motor 310, and the shaft voltage of the motor 310. , MoThe sensor 330 provided in the industrial equipment 300 does not need to be a real sensor that actually measures the physical quantity of the measurement target, but may be a virtual sensor that outputs data based at least in part on measurement data from the real sensor. The output of the virtual sensor is obtained by inputting the measurement data from the real sensor to a predetermined logic. The predetermined logic is a logic (analysis model) that estimates the behavior of the target. For example, the predetermined logic estimates the rotation speed or torque of the motor 310 based on the actual current value of the motor 310. The data from the real sensor and the data from the virtual sensor are examples of "data derived from a real system."

[0039] The MES 410 is a manufacturing execution system. The MES 410 is connected to each of the industrial devices 300 so as to be able to communicate with them. The MES 410 grasps the state of each of the industrial devices 300 and manages the manufacturing process. The MES 410 gives instructions to workers and supports the workers in their work.

[0040] The IoT gateway 420 collects data from the sensors 330 of each industrial device 300 and transmits it to the digital twin processing execution unit 200.

[0041] The other system 500 is a system related to the industrial equipment 300 which is an actual system, and examples thereof include a design support system, a parts management system, an inventory management system, a purchasing system, a maintenance management system, and a worker management system.

[0042] The digital twin management system 100 can provide data to other systems 500 based on the analysis results for each industrial device 300. For example, when the remaining life of the industrial device 300 or its parts changes, it becomes necessary to change the maintenance plan, adjust the number of parts ordered, and adjust inventory. Therefore, the other systems 500 execute the necessary processing based on the information provided by the digital twin management system 100.

[0043] The digital twin management unit 120 stores a digital twin configuration DB T2, an analysis model DB T3, and a grouping policy DB The digital twin management unit 120 manages the scenario application plan DB T4, the scenario application plan DB T7, and the scenario instance DB T8. The digital twin management unit 120 executes the process described in FIG. 1 and the process described later in FIG. 12 and FIG. 19. The data store management unit 140 manages the industrial equipment 300 The meta-information of the data collected from the real sensors and virtual sensors is managed in the data store management DB T5. The scenario template management unit 150 manages the scenario template DB T6. The management user interface 160 teeth This is a function that displays the management screen.

[0044] FIG. 3 shows an example of the hardware configuration of a computer CS that can be used as the digital twin management system 100 or the digital twin processing execution unit 200.

[0045] The computer CS includes, for example, a processor 101, a memory 102, an auxiliary storage device 103, a communication device 104, an output device 105, an input device 106, and a read / write device 107, and these devices 101 to 107 are connected by a bus .

[0046] The processor 101 realizes a predetermined function by reading and executing a predetermined computer program transferred from the auxiliary storage device 103 to the memory 102. When the computer CS is used as the digital twin management system 100, the function as the digital twin management system 100 becomes the predetermined function.

[0047] The communication device 104 communicates with other computers 200, 500, etc. via a communication network CN. The communication network CN may be a public line or a dedicated line.

[0048] The output device 105 provides information to a user of the digital twin management system 100, such as a system administrator. The output device 105 is, for example, a monitor display, a printer, a voice synthesizer, etc.

[0049] The input device 106 acquires instructions or information from a user. The input device 106 is, for example, a keyboard, a touch panel, a pointing device, a voice recognition device, or the like.

[0050] The read / write device 107 reads and writes data from / to an external storage medium MM. The external storage medium MM is a medium capable of retaining data for a long period of time, such as a flash memory, a hard disk, an optical disk, or a magnetic tape.

[0051] Computer programs or data stored in the external storage medium MM can be transferred from the read / write device 107 to the memory 102 or the auxiliary storage device 103. Conversely, computer programs or data stored in the memory 102 or the auxiliary storage device 103 can also be transferred to the external storage medium MM. A part or all of a specific computer program for realizing each function of the digital twin management system 100 of this embodiment can also be stored in the external storage medium MM, and the external storage medium MM can be distributed on the market.

[0052] It is also possible to transfer a specific computer program related to the digital twin management system 100 to another computer or storage device (neither shown) via the communications network CN without using an external storage medium MM. Furthermore, the digital twin management system 100 can receive a computer program or data from an external program distribution server (not shown) and store it in the auxiliary storage device 103.

[0053] Among the components of the computer CS, the output device 105 and the input device 106 used for information exchange with the user may be provided in an operation terminal (not shown) separate from the computer CS. For example, a desktop or laptop personal computer, a tablet terminal, a mobile phone (including a so-called smartphone), a mobile information terminal, a wearable terminal, or the like may be used as the operation terminal.

[0054] 4 is an example of the equipment management database T1. The equipment management database T1 manages information about each piece of equipment (here, each piece of industrial equipment 300) that is subject to management by the digital twin management system 100.

[0055] The device management database T1 includes, for example, an asset ID C11, a device name C12, a model C13, configuration information C14, an installation location C15, and an installation date and time C16.

[0056] The asset ID C11 is identification information that uniquely identifies each industrial device 300 that is managed by the digital twin management system 100. The device name C12 is the name of the industrial device 300, and is information that indicates the type of device, such as "cutting machine 001," "press machine 007," or "injection molding machine 333."

[0057] The model C13 is information for identifying models included in the same model of device 300. Even industrial devices 300 of the same model may have different detailed specifications, in which case they are given different model names.

[0058] The configuration information C14 is information indicating the configuration of the industrial equipment 300. The information indicating the configuration of the industrial equipment 300 is, for example, the type and quantity of main parts constituting the industrial equipment 300. A list of main parts constituting the industrial equipment 300 may be prepared, and information specifying the list may be stored in the field of the configuration information C14.

[0059] The installation location C15 is information for identifying the location where the industrial equipment 300 is installed. The installation location C15 may be identified by, for example, a building number and a line number. A value indicating coordinates within the site of a factory may be stored in the installation location C15. The introduction date and time C16 indicates the date and time when the industrial equipment 300 was introduced into the factory.

[0060] 5 is an example of the digital twin configuration database T2. The digital twin configuration database T2 manages the configuration of a digital twin that reproduces the industrial device 300 in a virtual space.

[0061] The digital twin configuration database T2 includes, for example, a digital twin ID (abbreviated as DTID in the figure) C21, a name C22, a target device ID C23, meta information C24, monitoring information C25, configured scenario instances C26, and digital twin relationship information C27.

[0062] The digital twin ID (DTID in the figure) C21 is identification information that identifies the digital twin. The name C22 is the name of the digital twin. The name of the digital twin is used to identify the industrial machine that corresponds to the digital twin. vessel The name may include at least a portion of the name.

[0063] The target device ID C23 is the corresponding industrial device in the digital twin. vessel 300. The target device ID C23 uses the same value as the asset ID C11 in the device management database T1.

[0064] The meta information C24 is meta information about the digital twin. The meta information includes, for example, the model name, the serial number (S / N), the introduction date and time, etc.

[0065] Monitoring information C25 is the corresponding industrial machine vessel This is the sensor data monitored by the digital twin for 300. DID C51 is stored in order to link it with the information managed in the data store management DB T5.

[0066] The sensor data includes data from the actual sensor 330 (real sensor 330) and data from a virtual sensor. The virtual sensor data is obtained by inputting the data from the real sensor 330 into a predetermined logic (analysis model).

[0067] The configured scenario instance C26 is an instance of a configured scenario. For example, the scenario estimates the remaining life of the motor 310 from the motor current, the motor shaft voltage, and the motor rotation speed. Scenarios with other contents can also be prepared.

[0068] Digital twin relationship information C27 is relationship information with other digital twins. For example, in the digital twin relationship information C27, the relationship between cutting equipment digital twin 001 (DTID: DT-001), which is the digital twin of cutting equipment 001, and industrial motor digital twin 001 (DTID: DT-003), which is the digital twin of internal industrial motor A, one of the parts that make up cutting equipment 001, is expressed as "Child," and this information is stored. Also, in the digital twin relationship information C27 for industrial motor digital twin 001, , cut The relationship with Cutting Machine Digital Twin 001 (DTID: DT-001), which is the digital twin of Cutting Machine 001, is expressed as "Parent," and the relevant information is stored.

[0069] 6 is an example of the analytical model database T3. The analytical model database T3 manages analytical models and the relationships between the analytical models and scenarios.

[0070] The analysis model database T3 includes, for example, a serial number C31, a name C32, a model ID C33, a type C34, a scenario ID C35, and 、 Input parameter C36 and output parameter Meter C37 and setting parameters C38.

[0071] The sequence number C31 is a sequence number assigned to the record. The name C32 is the name of the analysis model. Although omitted in FIG. 6, the name of the analysis model can be given so that the user can see the role (purpose) of the analysis model. 。 Names of the analysis models include, for example, current leveling (NM1), shaft voltage estimation (NM2), torque estimation (NM3), rotation speed estimation (NM4), motor bearing remaining life estimation (NM5), current leveling (NM6), torque estimation (NM7), rotation speed estimation (NM8), and cutting machine drill fault diagnosis (NM9).

[0072] Model ID (MID in the figure) C33 is identification information that uniquely identifies the analysis program that constitutes the analysis model. Here, for example, M-03 in the figure is an identifier of a program for torque estimation, while M-03-ver.2 in the figure represents an identifier of an updated version of a program for torque estimation that is a different version of a program that has improved M-03, for example, in terms of estimation accuracy. Similarly, M-04 in the figure is an identifier of a program for rotation speed estimation, while M-04-ver.2 in the figure represents an identifier of an updated version of a program for rotation speed estimation that is a different version of a program that has improved M-04. Type C34 indicates the type of analysis model. Types of analysis models include, for example, preprocessing (MT1), behavior estimation (MT2), and fault judgment (MT3).

[0073] The scenario ID C35 is information for identifying the scenario in which the analysis model is used. The input parameters C36 are parameters input to the analysis model. The input parameters include, for example, the actual motor current (RS-01), the motor current (after application of a high-pass filter) (VS-01), the shaft voltage (VS-02), the torque (VS-03), the number of rotations (VS-04), the motor current (after application of a high-pass filter) ( V These include: (S-06), material (M-01), usage history (M-02), torque (VS-07), and rotation speed (VS-08).

[0074] The output parameters C37 are parameters output from the analysis model. The output parameters include, for example, the motor current (after high-pass filter application) (VS-01), shaft voltage (VS-02), torque (VS-03), rotation speed (VS-04), remaining life (VS-05), motor current (after high-pass filter application) (VS-06), V S-06), torque (VS-07), rotation speed (VS-08) 、 Drill wear condition (VS-09), etc.

[0075] The setting parameters C38 are parameters that are set in the analysis model, and include, for example, a high-pass filter (HPF).

[0076] 7 is an example of a grouping policy database T4. A grouping policy is a criterion for grouping integrated scenario candidates.

[0077] The grouping policy database T4 includes, for example, policy ID (In the figure, PID) C41, an index C42, a value C43, a priority C44, and a selection policy C45.

[0078] The policy ID C41 is information for identifying the grouping policy. The index C42 is the content of the grouping policy. Examples of the index include resource sharing rate, rate of change in accuracy, and change in processing time. The value C43 is a range of values ​​when grouping multiple integrated scenario candidates based on the index C42. The priority C44 indicates the priority of the grouping policy. The selection policy C45 is a criterion for selecting an integrated scenario.

[0079] 8 is an example of a data store management database T5. The data store management database T5 manages data used in the digital twin management system 100. The data store management database T5 includes, for example, a data store ID C51, a name C52, a type C53, meta information C54, and a field C55.

[0080] The data store ID (DID in the figure) C51 is information that identifies the managed data. The name C52 is the name of the managed data. The type C53 indicates the type of data, such as whether it is time-series data. The meta information C54 is attribute information about the managed data. The meta information includes, for example, the origin of the managed data (whether it is data from a real sensor or data from a virtual sensor) and the data collection interval (in the case of a real sensor).

[0081] Field C55 is an item of the data to be managed. The data field may contain, for example, the data acquisition date and time (time stamp), the industrial machine corresponding to the data, vessel Identification information (asset ID), vesselThese include the digital twin identification information (DTID) corresponding to the digital twin, and items in which data values ​​are stored (current value, voltage value, torque, operation).

[0082] 9 is an example of a scenario template database T6. The scenario template database T6 manages scenario templates. In this embodiment, a predetermined group of information required when adding a model (simulation function) to a digital twin is made into a template as a scenario. The scenario template database T6 includes, for example, a scenario ID (SID in the figure) C61, a name C62, a target device model C63, and premise data C64.

[0083] The scenario ID C61 is information for identifying the scenario template. The name C62 is the name of the scenario template. For example, the name can be a name that indicates the role of the template, such as "Motor remaining life estimation scenario" or "Drill remaining life estimation scenario." The target equipment model C63 stores an identifier that specifies the model of the equipment to which the scenario can be applied. The premise data C64 is data that is a premise for using the scenario template. For example, the motor current is required to use the motor remaining life estimation scenario. The motor current and the operation log are required to use the drill remaining life estimation scenario.

[0084] 10 is an example of a scenario application plan database T7. The scenario application plan database T7 shows the contents of candidates (integrated scenario candidates) when adding a new simulation function to an existing digital twin. In the following, two candidates are illustrated, but three or more candidates may be created.

[0085] The scenario application plan database T7 includes, for example, a plan name (Plan in the drawing) C71, an impact on the existing configuration C72, an impact simulation C73, and a resource consumption C74.

[0086] Plan name C71 is the name of the plan when a new scenario is applied to an existing digital twin. A digital twin in which the functions specified in a new scenario are added to an existing digital twin is sometimes called an updated digital twin. An updated digital twin is a digital twin in which multiple scenarios are integrated.

[0087] The impact on existing configuration C72 indicates the impact on the configuration of the existing digital twin. Information indicating which part of the configuration of the existing digital twin will be changed, such as "changing the pre-processing settings of the current sensor," "changing the rotation speed estimation logic," or "changing the torque estimation logic," is recorded in the impact on existing configuration C72.

[0088] Impact simulation C73 is the result of simulating the impact on the performance of an updated digital twin when a new scenario is applied to an existing digital twin. The digital twin management system 100 calculates the impact on the performance of the updated digital twin by performing a simulation process on candidates for the updated digital twin using data collected in the past. In impact simulation C73, predictions such as "motor remaining life estimation accuracy decreases by -2%" and "motor remaining life estimation processing time 60 seconds" are stored.

[0089] The resource consumption C74 is the resource consumed by the updated digital twin. The resource is, for example, a computer resource such as a processor usage rate, a memory consumption amount, and a usage amount of an auxiliary storage device.

[0090] FIG. 11 shows an example of a scenario instance database T8. The scenario instance database T8 includes, for example, a scenario instance ID C81, an applied digital twin C82, an applied scenario template C83, an applied plan C84, and an applied scenario C85. The scenario instance ID C81 is information for identifying each scenario instance. The applied digital twin C82 is identification information (ID) of the applied digital twin. The applied scenario template C83 is the scenario template to be applied. The applied plan C84 is the plan to be applied (a candidate for an updated digital twin). The applied scenario C85 is the scenario to be applied. The applied scenario C85 stores identification information (ID) of the generated integrated scenario shown in FIG. 15 or FIG. 16. The example of FIG. 11 shows a case where integrated scenario B is selected.

[0091] When the user selects "Apply" from among the plans proposed to the user via the screen, which digital twin (C82), which scenario template (C83), and which plan (C84) was applied are associated with the final processing flow (C85) and managed in the scenario instance database T8.

[0092] The digital twin management process will be explained using the flowchart in Fig. 12. When an instruction is given to add a new scenario template to an existing digital twin (S11), the digital twin management system 100 determines whether the data required for the new scenario template can be acquired from the industrial equipment 300 that is the subject of the digital twin (S12). The required data may be either real sensor data or virtual sensor data.

[0093] If the digital twin management system 100 determines that it cannot acquire the necessary data (S12: NO), it notifies the user (such as a system administrator) of the digital twin management system 100 of an error (S13).

[0094] If the digital twin management system 100 determines that it can obtain the necessary data (S12: YES), it compares the configuration of the existing digital twin with the configuration of the new scenario template to extract multiple common layer candidates, which are common configurations between the existing digital twin and the new scenario template (S14).

[0095] The digital twin management system 100 creates multiple updated digital twin candidates from multiple common layer candidates (S15). The digital twin management system 100 checks whether there is past data available to evaluate each updated digital twin candidate (S16).

[0096] When the digital twin management system 100 determines that there is past data available (S16: YES), it runs each updated digital twin candidate (application scenario) using that past data, compares and evaluates the resulting values ​​with the past resulting values ​​(S17), organizes the evaluation results based on a grouping policy, and narrows down the updated digital twin candidates in order of highest evaluation (S18).

[0097] No historical data available to evaluate candidate updated digital twins It was determined that If so (S16: NO), the digital twin management system 100 skips the evaluation step (S17) and proceeds to step S18.

[0098] The digital twin management system 100 presents information to the user by displaying the contents of the candidates for the updated digital twin and their evaluation on the monitor display (S19). The user selects one of the candidates for the updated digital twin based on the presented information and instructs the digital twin management system 100. The digital twin management system 100 changes the configuration of the digital twin so that it becomes the updated digital twin selected by the user (S20). In addition, various tables related to the digital twin are updated in accordance with the configuration change. Specifically, the scenario instance database T8 is updated by associating with which scenario template was based (C83), which plan (C84) was applied to the digital twin (C82), and the applied scenario (C85), which is the final processing flow. In addition, the configured scenario instance C26 in the digital twin configuration DB T2 is updated to the added or updated scenario instance ID C81. Furthermore, an identifier (DID C51) of the data store management DB T5 regarding storage destination information of data calculated by application of various analytical models described in the analytical model DB T3 that realize each process of the application scenario is stored in the monitoring information C25 of the digital twin configuration DB T2.

[0099] Creation of candidates for an updated digital twin will be described with reference to Fig. 13 to Fig. 16. Fig. 13 shows a templated process flow included in a scenario for estimating the remaining life of the motor 310. Fig. 14 shows a templated process flow included in a scenario for estimating the remaining life of the drill 320.

[0100] Assume that the existing digital twin estimates the remaining lifespan of the motor 310 according to the processing flow shown in Fig. 13. An example will be described in which a scenario for estimating the remaining lifespan of the drill 320 shown in Fig. 14 is added to this existing digital twin to create a candidate for a new integrated digital twin.

[0101] In the process flow of the scenario for estimating the remaining life of a motor shown in Fig. 13, first, data (RS-01) from a current sensor is prepared (S31), an analytical model (NM1) is applied to the prepared sensor data (RS-01) to generate preprocessed data (VS-01) by leveling the sensor data (S32), and the preprocessed data (VS-01) is input to each analytical model (logic) (S33 to S35). The analytical models are a logic for estimating and storing the shaft voltage (NM2, S33), a logic for estimating and storing the torque (NM3, S34), and a logic for estimating and storing the rotation speed (NM4, S35).

[0102] Then, in the processing flow, each data item (torque (VS-02), rotation speed (VS-03), shaft voltage (VS-04)) to be input into the final analysis model is prepared (S36). Each prepared data item is input into an analysis model (NM5) that diagnoses signs of motor bearing failure (S37). The analysis model diagnoses signs of failure in the bearings of the motor 310, estimates the remaining life (VS-05) of the motor 310 from the signs of failure, and stores the results in a data store (D-006) (S37).

[0103] In the process flow of the scenario for estimating the remaining life of a drill shown in Fig. 14, data (RS-01) from a current sensor is prepared (S41), preprocessed data (VS-06) is generated by applying an analytical model (NM6) to the prepared sensor data (RS-01) and smoothing it (S42), and the preprocessed data (VS-06) is input to each analytical model (logic) (S43, S44). Each analytical model is a logic for estimating and storing torque (NM7, S43) and a logic for estimating and storing rotation speed (NM8, S44).

[0104] Then, in the processing flow, each data (torque (VS-07), rotation speed (VS-08), material (M-01), and usage history (M-02)) to be input into the final analysis model is prepared (S45). The material and usage history are the material of the drill 320 and the usage history of that drill 320.

[0105] Each of the prepared data is input (S46) into an analysis model (NM9) that estimates the remaining life of the drill, and the results are stored in a data store (D-008).

[0106] This is an explanatory diagram of an example of a candidate updated digital twin when a scenario for estimating the remaining life of a drill, as described in Figure 14, is added to the digital twin for estimating the remaining life of a motor, as described in Figure 13.

[0107] FIG. 15 is an example of an updated digital twin candidate (Plan A, integrated scenario A). FIG. 15 shows a process flow S50A of the common layer candidate 1, a process flow 60A for estimating the remaining life of a motor incorporating the process flow 50A of the common layer candidate 1, and a process flow 60B for estimating the remaining life of a drill incorporating the process flow 50A of the common layer candidate 1. Processing flow S 70A is shown.

[0108] In this candidate example, steps S31 and S41 for preparing data from the current sensor and steps S32 and S42 for preprocessing the data from the current sensor are standardized as step S51 for preparing data from the current sensor and step S52 for preprocessing the data from the current sensor.

[0109] FIG. 16 is another example of an updated digital twin candidate (Plan B, integrated scenario B). FIG. 16 shows the processing flow of common layer candidate 2. S 50B, a process flow S60B for estimating the remaining motor life incorporating the process flow 50B of common layer candidate 2, and a process flow S70B for estimating the remaining drill life incorporating the process flow S50B of common layer candidate 2 are shown.

[0110] In this candidate example, not only step S51 for preparing data from the current sensor and step S52 for preprocessing the data from the current sensor, but also step S53 for estimating the torque of the motor using an analytical model and step S54 for estimating the rotation speed of the motor using the analytical model are standardized. Step S53 corresponds to step S34 and step S43. Step S54 corresponds to step S35 and step S44. Here, the torque estimation analysis model NM3 used in step S34 and the torque estimation analysis model NM7 used in step S43 use different versions of the torque estimation analysis program, so when standardizing the step for estimating the torque in step S53, it is necessary to decide which program to use. In this embodiment, one method of deciding is to use the program with the newer version, but this is not limited to this.

[0111] In the candidate example (Plan B, integrated scenario B) shown in FIG. 16, the common parts are large compared to the candidate example (Plan A, integrated scenario A) shown in FIG. 15, so the sharing rate of computer resources is high, and for example, the resource consumption C74 is small as shown in FIG. 10. However, in the candidate example (Plan B, integrated scenario B) shown in FIG. 16, the common parts are large compared to the candidate example (Plan A, integrated scenario A) shown in FIG. 15, so the impact C72 on the existing configuration is large (the number of items in the impact C72 on the existing configuration is large). Therefore, for example, there are more steps to apply an analysis model different from the analysis model used up until now, which may affect the estimation accuracy of the analysis and the estimated processing time. The impact of this change is calculated by executing each integrated scenario (integrated scenario A, integrated scenario B) using past data in the step S17, and the calculated motor remaining life estimate value and its processing time, etc. are calculated, and the values ​​are compared with the results obtained in the past applied scenarios, and stored in the impact simulation C73. The administrator can check the degree of the impact by the simulation C73.

[0112] Figure 17 is an example of screen G2 that presents candidates for an updated digital twin to the user. This screen G2 is a detailed example of screen G1 described in Figure 1. This screen G2 provides the user with the results of verifying changes to the digital twin configuration when a new scenario is extended to an existing digital twin.

[0113] This screen G2 includes, for example, an update overview GP21, an update plan GP22, an impact simulation result GP23, an update button, and a cancel button.

[0114] The update summary GP21 displays a summary of the update of the digital twin. The update summary includes, for example, information for identifying the corresponding industrial machines 300 of the digital twin, the number of the identified industrial machines 300, and information for identifying a new scenario to be added to the existing digital twin.

[0115] The update plan GP22 includes information on the configuration of the candidate plan (plan). The information on the configuration of the candidate plan includes, for example, information on the common layer portion, information on the portion for estimating the remaining motor life, and information on the portion for estimating the remaining drill life. The update plan GP22 may be provided with buttons or tabs that display the names of the candidate plans, such as "Plan A" and "Plan B," so that the user can select the buttons or tabs. The update plan GP22 displays the contents of the candidate plan selected by the user.

[0116] The impact simulation result GP23 shows the results of evaluating each candidate using a predetermined evaluation index.

[0117] 18 shows an example of the impact simulation result GP 23. The impact simulation result GP 23 includes, for example, a candidate name GP 231, an impact on an existing configuration (processing flow configuration) of the digital twin GP 232, an impact simulation GP 233, a computer resource consumption amount GP 234, and an update time GP 235.

[0118] The impact GP232 on the existing configuration of the digital twin displays, for example, changes to the settings for pre-processing of sensor data, changes to the analysis model, etc. The impact simulation GP233 displays the results of a simulation of the change in performance when the digital twin is updated according to the selected plan. The impact simulation is performed using past data. The impact simulation displays, for example, the percentage increase (or decrease) in the accuracy of estimating the remaining life of the motor, the change in the time required to estimate the remaining life of the motor, etc.

[0119] The computer resource consumption GP 234 displays the consumption of various resources, such as the processor usage rate, memory consumption, auxiliary storage device consumption, etc. The computer resource consumption also displays the sharing rate of the computer resource.

[0120] Update Time GP235 displays how long it will take to update the digital twin according to the selected plan and whether a reboot is required.

[0121] According to this embodiment configured in this manner, when a new scenario is added to an ongoing digital twin, information about candidates for updated digital twins can be provided to the user, thereby improving usability for the user.

[0122] According to this embodiment, it is possible to provide the user with multiple candidates that share some configurations and simulation results for when these candidates are selected, so that the user can objectively and efficiently determine which candidate is preferable to select. EXAMPLES

[0123] A second embodiment will be described with reference to Fig. 19. In this embodiment, the differences from the first embodiment will be mainly described. Fig. 19 is a flowchart of a digital twin management process.

[0124] In the digital twin management system 100 of this embodiment, when a new scenario is added to the digital twin and updated (S11 to S20), vessel The impact on the system 300 or other system 500 is calculated (S21).

[0125] The digital twin management system 100 determines whether or not notification to the industrial equipment 300 or the other system 500 is necessary based on the calculation result of step S21 (S22). If it is determined that notification is not necessary (S22: NO), this process ends.

[0126] When the digital twin management system 100 determines that notification to the industrial equipment 300 or other system 500 is necessary (S22: YES), it notifies the industrial equipment 300 or other system 500 that the digital twin has been updated (S23).

[0127] This embodiment configured as described above also achieves the same effects as those of the first embodiment. Furthermore, according to this embodiment, by updating the digital twin, for example, when the accuracy of estimating the remaining life of a major part of the industrial equipment 300 decreases, it is possible to increase the inventory of the major parts or to advance the timing of ordering the major parts. Therefore, usability for the user is further improved.

[0128] The present invention is not limited to the above-described embodiment, and various modifications are included. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to those including all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0129] For example, it is possible to evaluate the performance and estimate the remaining life of key components such as control boards and power supply circuits of industrial equipment. [Explanation of symbols]

[0130] 1: IoT management system, 10: Scenario addition unit, 11: Common part extraction unit, 12: Candidate generation unit, 13: Evaluation unit, 14: Narrowing unit, 15: Information presentation unit, 16: Configuration change implementation unit 16, System notification unit 17, 100: Digital twin management system, 200: Digital twin processing execution unit, 300: Industrial equipment, 310: Motor, 320: Drill, 330: Sensor, 340: Power supply, 410: MES, 420: IoT gateway, 500: Other systems, G1.G2: Screen examples

Claims

1. A digital twin management system that manages a digital twin that reproduces a real system in a physical space in a virtual space, A processor and a memory for storing a predetermined computer program to be executed by the processor; The processor, When generating an integrated scenario by adding a second scenario to a first scenario, extracting a plurality of common parts from the first scenario and the second scenario; generating a plurality of candidate integrated scenarios by varying the extracted common parts; Calculating an evaluation of each of the generated integrated scenario candidates; The configuration information of each of the integrated scenario candidates and the evaluation of each of the integrated scenario candidates are output in association with each other. Digital Twin Management System.

2. The first scenario, the second scenario, and the integrated scenario are templated as scenarios, which are digital twin configurations including information indicating the configuration of a target real system, attribute information indicating the purpose of the scenario, data derived from the real system, a processing flow executed based on the acquired data, an analytical model included in the processing flow, and input / output parameters of the analytical model. The digital twin management system of claim 1 .

3. By comparing the first scenario with the second scenario, a plurality of common parts between the first scenario and the second scenario are extracted. The digital twin management system of claim 2.

4. The configuration information of each of the integrated scenario candidates and the evaluation of each of the integrated scenario candidates are provided to a user via a monitor display. The digital twin management system of claim 1 .

5. The evaluation includes at least one of the following aspects: resource consumption and performance stability. The digital twin management system of claim 4.

6. The assessment is calculated using historical data used in the first scenario. The digital twin management system of claim 5.

7. The integrated scenario candidates are selected according to a predefined policy. The digital twin management system of claim 6.

8. The evaluation includes resource consumption, performance stability, and impact on other systems related to the actual system. The digital twin management system of claim 4.

9. Information regarding an impact on the other system regarding the integration scenario candidate selected by the user from among the integration scenario candidates is provided to the other system or to a further system. The digital twin management system of claim 8.

10. The real system is an industrial device used in a factory. The digital twin management system of claim 1 .

11. A virtual model management method for managing a digital twin that reproduces a real system in a physical space in a virtual space using a computer, comprising: The computer includes: When a second scenario is added to a first scenario, a plurality of common parts are extracted from the first scenario and the second scenario; generating a plurality of integrated scenario candidates by varying the extracted common parts; Calculating an evaluation of each of the generated integrated scenario candidates; The configuration information of each of the integrated scenario candidates and the evaluation of each of the integrated scenario candidates are output in association with each other. Digital twin management methods.

Images