Method for operating a system using a digital twin

Abstract

The invention relates to a method for operating a system (100), e.g. an underwater system, having the steps of: providing values of one or more operating variables (220) of the system from the system, in particular from one or more sensors (130) of the system; providing values of one or more actuation variables (230) for the system, said actuation variables having been determined as part of a control or regulating process on the basis of the values of the operating variable or at least one of the plurality of operating variables (220); determining, on the basis of the values of the operating variable or at least one of the plurality of operating variables (220) and on the basis of the values of the actuation variable or at least one of the plurality of actuation variables (230), values of one or more monitoring variables (250) of the system using a model (242) of the system, in particular an integrated model, said one or more monitoring variables (250) of the system corresponding to the operating variable or to one of the plurality of operating variables (220) of the system; if necessary, providing the values of the monitoring variable or at least one of the plurality of monitoring variables (250) for use in the control or regulating process in place of the values of the corresponding particular operating variable (220) in order to output the values of the actuation variable or the plurality of actuation variables (230) to the system (100), in particular to a drive of the system.

Description

[0001] Method for operating a system with a digital twin

[0002] Description

[0003] The present invention relates to a method for operating a system, in particular an electrohydrostatic actuator, a computing unit and a computer program for carrying out the method, as well as such a system.

[0004] Background of the invention

[0005] Systems such as electrohydrostatic actuators can be used in many areas. One application, for example, is the use of an electrohydrostatic actuator underwater, e.g., to operate valves on pipelines or similar systems. In this and other applications, the highest possible system availability is desirable.

[0006] Disclosure of the invention

[0007] According to the invention, a method for operating a system, a computing unit, and a computer program for implementing the method, as well as such a system with the features of the independent patent claims, are proposed. Advantageous embodiments are the subject of the dependent claims and the following description.

[0008] The invention relates to systems in which operating variables or their values ​​are recorded, e.g. by means of sensors, in order to then determine control variables or values ​​therefor based on these, by means of which the system is then controlled. This can be done, for example, within the framework of a control or regulation. In this sense, the recorded values ​​of the operating variables can be actual values, and the values ​​of the control variables can then be manipulated variables. A preferred application example is a system that has an electrohydrostatic actuator or is designed as such. However, other control systems can also be considered as systems, e.g. electrical actuators or generally electrohydraulic actuators; a control system is understood to be a system or device by means of which certain functions can be controlled, e.g. valves can be opened and closed. In the case of an electrohydrostatic actuator, for exampleThe actuator can operate the valve. Instead of a valve, other devices can also be operated using the system.

[0009] The electrohydrostatic actuator mentioned can, for example, comprise a processing unit for control or regulation (control or regulation unit), an electric drive (e.g. electric motor), a hydraulic circuit with control valves, the actual actuator and, if necessary, safety springs or the like. The actuator can then be coupled to the actual valve to be actuated, e.g. via a suitable interface. In addition, several sensors are typically present to measure various operating variables, i.e. to record values ​​of the operating variables. Such operating variables can be, for example, a position of the actuator and one or more pressures in the system, i.e., a position sensor (or angle sensor, depending on the type of actuator) and pressure sensors can be provided.

[0010] Based on the values ​​of the operating variables, values ​​for the control variables can then be determined. Possible control variables include, for example, the speed of the electric drive and the position of one or more control valves (or adjusting valves). In this case, this can be a conventional open-loop or closed-loop control of the actuator.

[0011] Complex and compact systems or control systems, in particular, often require a high degree of integration, reliability, and safety. This is especially true when deployed in remote and inaccessible locations, such as the ocean or underwater in general. To achieve all these factors, redundancy may be implemented in the system's controllers, sensor systems, and actuators.

[0012] Physical redundancy leads to increased system complexity, larger space requirements, and higher costs, especially when expensive materials such as stainless steel must be used to protect sensitive hardware from environmental influences. In the subsea processing and subsea production industries, for example, it is common practice to redundantly implement control units or control boards, sensors, and actuators to ensure reliable operation over a planned lifetime of, for example, at least 25 years in extremely deep waters.

[0013] However, the often required size constraints in such complex systems limit the possibilities for component redundancy and therefore require alternative strategies to increase system reliability when necessary.

[0014] Against this background, the use of a so-called digital twin is proposed for the purposes of redundancy. This digital twin is a realistic model or simulation model which is designed in such a way that it runs in the computing unit or control or regulating unit (also referred to as drive control), i.e. is integrated or implemented there. It is also conceivable that the model is integrated or provided in a different computing unit, i.e. different from the control or regulating unit, whereby this other computing unit is then provided locally with the control or regulating unit, e.g. in a common housing or the like. This can be useful if the computing capacity of a conventional control or regulating unit is insufficient.

[0015] A corresponding method for operating a system such as the electrohydrostatic actuator explained above can therefore be carried out in particular in a computing unit or control or regulating unit or a computing unit different from the control or regulating unit and comprises the following steps.

[0016] Values ​​of one or more operating variables of the system are provided or received by the system, in particular by one or more sensors of the system, e.g. in the executing processing unit. These values ​​can be measured values ​​from the sensors. Within the scope of open or closed-loop control, based on the values ​​of the one or at least one of the multiple operating variables, values ​​of one or more control variables for the system are determined, which are also provided. The values ​​of the one or more control variables can be output to the system, in particular a drive of the system. In this respect, it can be a conventional open or closed-loop control. In addition, based on the values ​​of the one or at least one of the multiple operating variables and the values ​​of the one or at least one of the multiple control variables, using a model of the system (e.g. the so-calledDigital twins are used to determine, in particular estimate, values ​​of one or more monitoring variables of the system. The one or more monitoring variables of the system correspond to one or more of the system's operating variables. In other words, using the system model, comparative values ​​for the measured values ​​of the sensors can be determined digitally.

[0017] If necessary, the values ​​of one or at least one of the several monitoring variables can then be used in the control or regulation process instead of the values ​​of the corresponding operating variables. Possible and preferred criteria for whether and when the values ​​of the monitoring variables are used instead of the operating variables will be explained in more detail later. A simple example would be a detected sensor defect.

[0018] This not only allows the system to be monitored, but also creates digital redundancy for, for example, the sensors present in the system, so that at least one of several sensors can be replaced by the model or the values ​​digitally determined using the model if necessary. This increases the reliability and availability of the system without adding new components.

[0019] At this point, it should be mentioned that the steps explained describe the general operation of the system, i.e. these steps are carried out repeatedly over a longer period of time. It is understood that, for example, the values ​​of the monitoring variables are not initially used in the control or regulation, but are continuously determined based on the control variables or their values ​​obtained in the process. Later, values ​​of one or more monitoring variables can be used instead of corresponding operating variables in the control or regulation. The operating variables then replaced can, but do not have to, be taken into account in the further determination of the monitoring variables. If, for example, a defect in a sensor has been detected, it can be provided that the values ​​of the operating variable in question are no longer used; instead, for example,The values ​​of the relevant monitoring variables are used together with the values ​​of other operating variables to determine the values ​​of all monitoring variables using the model. In this sense, the values ​​of the monitoring variables are used not only in control or regulation, but also in determining the values ​​of the monitoring variables, instead of the values ​​of the corresponding operating variables.

[0020] The model or simulation model of the system can be designed or implemented in various ways. For example, a purely physically motivated model can be used, in which the interrelationships of the system are represented, for example, by differential equations. This is then also referred to as a so-called "first principles model." The system can also be represented using a purely data-based or data-driven model. In this case, so-called artificial intelligence such as a neural network can be considered, which simulates the function of the system. Such a data-based model must then usually be trained in advance, for example using known values ​​of operating and control variables (so-called training data). Such a model can also be referred to as a machine learning model. However, a combination of both types of models, i.e., physically motivated and data-based models, can also be used. For example,Some aspects are represented by differential equations, while additional aspects that are difficult to represent using equations are represented by a data-based model component. This could include frictional forces or similar factors.

[0021] In one embodiment, the model is updated based on the values ​​of the one or at least one of the plurality of operating variables and / or based on the values ​​of the one or at least one of the plurality of control variables. This can also be referred to as learning or adapting the model. For this purpose, for example, state variables (or state parameters) of the system, such as system efficiency, as well as one or more friction factors of the system, can be repeatedly estimated (or determined) in order to update the system model as needed. For example, a value of a control variable that increases over time while the value of the operating variable remains the same can indicate that friction has increased. This does not in itself represent a malfunction of the system, but merely aging.However, this should also be simulated in the model in order to obtain the most accurate values ​​of the monitored variables throughout the entire service life. One or more indicators of wear or aging processes, or other information, should also be considered. The state variables do not necessarily have to describe the system itself, but can also describe, for example, the device operated by the system, such as the aforementioned valve. Changes in the device operated by the system also affect the system itself.

[0022] In one embodiment, information about such variables of the device to be actuated by the system can also be determined and output. This allows the device to be actuated to be further developed or maintained, for example.

[0023] In one embodiment, it is also determined whether a malfunction of the system is present or is to be expected, based on the values ​​of, for example, at least one of the following variables: the one or at least one of the plurality of monitoring variables, the one or at least one of the plurality of operating variables, and the one or at least one of the plurality of control variables. For example, a continuous comparison can be made between the values ​​of an operating variable and the corresponding monitoring variable. If a deviation between these values ​​is greater than a predetermined threshold value or is to be expected, a malfunction can be assumed. A defective sensor, for example, would permanently supply a value of zero, which would lead to a deviation. However, a permanent offset could also indicate a malfunction.If a sensor suddenly fails, the malfunction can be detected immediately. However, if a deviation becomes larger over time, it can be assumed that a failure of the sensor is to be expected.

[0024] If a malfunction of the system exists or is to be expected, it can be determined that the values ​​of the one or at least one of the several monitoring variables are to be used in the control or regulation, and then in particular are also used, i.e. instead of the values ​​of the actual operating variable. This is particularly the case if the malfunction lies in a sensor. Alternatively, it can also be provided that if a malfunction of the system exists or is to be expected, a check is carried out to determine whether the values ​​of the one or at least one of the several monitoring variables are to be transferred to the control or regulation module or the control or regulation according to a test criterion.If the test criterion is met, it is determined whether the values ​​of one or at least one of the several monitoring variables are to be used in the control or regulation, and then, in particular, to maintain the system's services until repairs can be carried out through scheduled maintenance. The test criterion can be used, for example, to check or determine whether the malfunction lies in a sensor, possibly in an actuator, or elsewhere in the system.

[0025] The use of a highly reliable digital twin (or model) of a system allows for the combination of all available sensors to create virtual redundancy without requiring additional space or physical complexity. The digital twin can be configured to automatically learn system behavior during use. It can monitor component behavior, detect or predict malfunctions, and even take over system functions if a component malfunction occurs. This allows for the provision of both affordable (i.e., cost-effective) and reliable systems.

[0026] In the field of renewable energy and CC>2 storage in the subsea industry, for example, cost-effective systems and components are required that offer a high degree of reliability and controllability, as they are installed up to several kilometers below sea level and hundreds of kilometers from the coast. Due to these extreme requirements, a system used there should be able to continue performing tasks for several weeks or years, even after parts of the system malfunction, before it can be serviced.

[0027] Physical redundancy can be avoided and replaced by virtual redundancy via a digital twin, which can be activated, for example, in the event of a relevant event during deployment. The accumulated knowledge of the system prior to failure can enable a user to continue operating the system even after a component fails. This leads to optimized operating costs, as the system can automatically respond to malfunctions and, for example, also provide information about the urgency of maintenance and repair.

[0028] A computing unit according to the invention, e.g., a control or regulating unit of an electrohydrostatic actuator, or a separate computing unit intended for local use with such a control or regulating unit, is configured, particularly in terms of programming, to carry out a method according to the invention. The invention also relates to a system having one or more sensors for detecting values ​​of one or more operating variables of the system, having a drive, and having a computing unit according to the invention. In particular, the system may comprise an electrohydrostatic actuator or be designed as such.

[0029] Implementing a method according to the invention in the form of a computer program or computer program product with program code for performing all method steps is also advantageous, as this entails particularly low costs, especially if an executing control unit is also used for additional tasks and is therefore already present. Suitable data storage media for providing the computer program include, in particular, magnetic, optical, and electrical storage devices, such as hard disks, flash memories, EEPROMs, DVDs, SD cards, and others. Downloading a program via computer networks (Internet, intranet, etc.) is also possible.

[0030] Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

[0031] It is understood that the features mentioned above and those to be explained below can be used not only in the combination specified in each case, but also in other combinations or on their own, without departing from the scope of the present invention.

[0032] The invention is illustrated schematically in the drawing using an embodiment and is described in detail below with reference to the drawing.

[0033] Character description

[0034] Figure 1 shows schematically a system according to the invention in a preferred embodiment.

[0035] Figure 2 shows a schematic flow diagram to illustrate a method according to the invention in a preferred embodiment.

[0036] Detailed Description of the Drawing Figure 1 schematically shows a system 100 according to the invention in a preferred embodiment. System 100 is an electrohydrostatic actuator, which is schematically indicated here with some components. The electrohydrostatic actuator includes, for example, a processing unit 110 designed as a control or regulating unit, an electric drive 120, a hydraulic circuit 122 comprising control or regulating valves (e.g., electromagnetically controllable valves), a safety spring 124, optionally including the control system therefor, the actual actuator 126, and a mechanical interface 128.

[0037] Furthermore, the electrohydrostatic actuator 100 has sensors 130, by means of which, for example, a position of the actuator 126 and one or more pressures in the hydraulic circuit 122 can be detected or measured. The measured values ​​can be transmitted to the computing unit 110.

[0038] In addition, a valve 140 is schematically indicated, which is to be actuated by means of the electrohydrostatic actuator 100. For this purpose, the electrohydrostatic actuator 100 can be coupled to the valve 140 via the mechanical interface 128. The valve 140 can be, for example, a valve in a pipeline, a pipe or the like, which is arranged underwater, for example, and must be able to be actuated when required. For example, the valve 140 is a rotary valve; accordingly, the actuator 126 can also be designed as a rotary actuator, or as a linear actuator. It should be mentioned that the valve 140 to be controlled is to be distinguished from any control valves or control valves integrated in the system 100 as part of the hydraulic system. Instead of the valve 140, other parts or components could also be controlled or otherwise actuated by means of the system 100. Furthermore, if necessary.Communication or data lines must be provided to send control commands from outside to the electrohydrostatic actuator.

[0039] Figure 2 shows a schematic flow diagram to illustrate a method according to the invention in a preferred embodiment. For this purpose, the computing unit 110 and the system 100 from Figure 1 are shown schematically, wherein the system 100 can then, for example, have the components shown in the figure with the exception of the computing unit 110. In the computing unit 110, a control or regulating module 200 and a redundancy module 240 are shown by way of example. These two modules perform different tasks, wherein the representation as separate modules here is only an example and could also be different; rather, the only important thing is that certain tasks can generally be performed, e.g. by the computing unit 110, regardless of the specific implementation. In general, however, this can be referred to as “embedded controller application software”.As also already mentioned, the redundancy module 240 could also be implemented in a different or separate computing unit, in which case suitable data communication would be provided.

[0040] During operation of the system 100, values ​​(measured values) of, for example, several operating variables 220 can be continuously recorded by the system 100, here the electrohydrostatic actuator; the sensors shown in Figure 1 can be used for this purpose. The values ​​of the operating variables 220 are received in the computing unit 110; if necessary, it can also be provided that the computing unit controls the sensors in a suitable manner to record the values.

[0041] The recorded values ​​of the operating variables 220 can then, if necessary after processing and / or integration in a transmission module 210, be transferred to the control or regulation model 200. There, within the framework of a control or regulation, based on the values ​​of the operating variables 220, values ​​of, for example, several control variables 230 for the system 100 are determined. These values ​​are output to the system 100, if necessary after processing and / or integration in the transmission module 210, i.e. the system 100 is controlled according to the values ​​of the control variables 230. This can also be referred to as control signals that are output to system 100. In particular, for example, the electric drive shown in Figure 1 and the control valves can be controlled in a suitable manner.

[0042] This control or regulation sequence of the system 100 can correspond to a conventional control or regulation in which, for example, the valve 140 shown in Figure 1 is opened or closed as needed, for which the actuator 126 must be moved in a suitable manner. In addition, the aforementioned redundancy module 240 is provided, which comprises or represents a digital twin. The redundancy module 240 comprises, for example, the model 242 of the system 100, an estimation module 244 (to update the parameters or state variables from the model 242, e.g., friction, efficiencies, resistances), a malfunction module 246, and a decision module 248. Here, too, it applies that these various modules do not necessarily have to be implemented separately in the software; rather, it depends on their tasks or functions.

[0043] The model 242 can be, for example, a physically motivated model, a data-based model, or a combination of both, as already explained above. The system 100, in this case the electrohydrostatic actuator, is mapped or represented by the model 242. The model receives, in particular, the operating variables 220 and the control variables 230 or their (current) values ​​as input values. Based on this, values ​​of monitoring variables 250 can then be determined using the model. The monitoring variables 250, corresponding in particular to the operating variables 220, are not measured using sensors, but are determined digitally using the model. Just as the actual operating variables 220 are set or result from specifying values ​​of the control variables 230, the monitoring variables 250 or their values ​​can be determined using the model 242 based on the values ​​of the control variables 230.In this respect, the values ​​of the operating variables 220 are not necessary for determining the values ​​of the monitoring variables, but can still be used to obtain more accurate values.

[0044] As mentioned, system 100 is represented in model 242. This particularly includes the representation of state variables (or state parameters) such as efficiencies or friction factors in the system. Such state variables are, in particular, variables that can change over time, e.g., with the aging of the system. For example, efficiency can decrease over time, or friction can increase, e.g., due to wear.

[0045] The estimation module 244 can also receive the operating variables 220 and the control variables 230 or their (current) values ​​as input values. In addition, the state variables 250 or their current values ​​can be known. Based on the operating variables 220 and the control variables 230 or their values, the state variables 252 can, for example, be continuously redetermined or estimated; if necessary, i.e. if the state variables 252 have changed by a certain value, for example, the model 242 can be updated accordingly. The model 242 thus always represents the current state of the system as accurately as possible, in particular including any aging of the system. For example, a value of a control variable that increases over time while the value of the operating variable remains the same can indicate that friction has increased.

[0046] The malfunction module 246 receives as input values, for example, the values ​​of the updated model variables or state variables and / or the operating variables 220 and / or the control variables 230, as well as, for example, the state variables 252. Based on this, it can then be determined whether a malfunction is present or at least to be expected, for example, within a certain time in the future. This is the case, for example, if a monitoring variable determined using the model deviates significantly from the corresponding operating variables or if such a deviation becomes increasingly larger over time. If this is the case, it is determined that a malfunction is present or to be expected.

[0047] The decision module 248 receives as input the determination of whether a malfunction is present or expected. Based on this information, it can then be determined whether the values ​​of the operating variables should continue to be used in the control or regulation module 200 or whether, for an operating variable, the values ​​to be used should be replaced by the values ​​of the corresponding monitoring variables—which are determined digitally using the model 242. If a replacement is to take place, the values ​​of the respective monitoring variable can subsequently be transferred from the redundancy module 240 to the control or regulation module 200, and the control or regulation module 200 can be instructed to use the values ​​of the monitoring variable instead of the operating variable with immediate effect. In this way, the operation of the system 100 is still possible with acceptable accuracy, at least for a certain period of time.The goal is, in particular, to continue the system's core services at a minimum acceptable level of quality until maintenance is possible or until the end of the system's service life, if the system so chooses. This information can be communicated, for example, via a communication interface (e.g., CAN or Ethernet), and a decision can then be made whether operation should be continued or whether the devices should be replaced. This means that at least the faults are detected or assumed (as with condition monitoring), but beyond that, the system's operation is still enabled, possibly with limited functionality.

Claims

Claims 1. A method for operating a system (100) comprising: Providing values ​​of one or more operating variables (220) of the system from the system, in particular from one or more sensors (130) of the system, Providing values ​​of one or more control variables (230) for the system, which have been determined within the scope of a control or regulation based on the values ​​of the one or at least one of the plurality of operating variables (220), Determining, based on the values ​​of the one or at least one of the plurality of operating variables (220) and the values ​​of the one or at least one of the plurality of control variables (230), by means of a model (242) of the system, in particular an integrated model, values ​​of one or more monitoring variables (250) of the system, wherein the one or more monitoring variables (250) of the system correspond to the one or each of the plurality of operating variables (220) of the system, if necessary, providing the values ​​of the one or at least one of the plurality of monitoring variables (250) for use in the control or regulation, instead of the values ​​of the corresponding respective operating variable (220), in order to output the values ​​of the one or more control variables (230) to the system (100), in particular a drive of the system.

2. The method according to claim 1, wherein the model (242) is updated based on the values ​​of the one or at least one of the plurality of operating variables (220) and / or based on the values ​​of the one or at least one of the plurality of control variables (230).

3. The method according to claim 2, wherein the model (242) is updated by updating one or more state variables (252), wherein the one or more state variables in particular comprise at least one of the following variables: an efficiency of the system and / or a device to be actuated by means of the system, one or more friction factors of the system and / or the device to be actuated by means of the system device to be operated, one or more indicators of wear or aging processes.

4. Method according to one of the preceding claims, wherein information about one or more of the variables of a device to be actuated by means of the system is determined and output, in particular one of the following variables: an efficiency of the device to be actuated by means of the system, one or more friction factors of the device to be actuated by means of the system, one or more indicators of wear or aging processes of the device to be actuated by means of the system.

5. Method according to one of the preceding claims, further comprising: Determining whether a malfunction of the system is present or is to be expected based on the values ​​of at least one variable, preferably at least one of the following variables: the one or at least one of the plurality of monitoring variables (250), one or more state variables, the one or at least one of the plurality of operating variables (220), and the one or at least one of the plurality of control variables (230); and determining, if a malfunction of the system is present or is to be expected, that the values ​​of the one or at least one of the plurality of monitoring variables are to be used in the control or regulation, or Check, if a malfunction of the system exists or is expected, whether the values ​​of the one or at least one of the several monitoring variables are to be used in the control or regulation according to a test criterion, and Determining that the values ​​of the one or at least one of the plurality of monitoring variables are to be used in the control or regulation if the test criterion is met.

6. The method according to claim 4, wherein it is determined that a malfunction is present or expected if a deviation between the values ​​of the one or more monitoring variables (250) and the corresponding operating variable (220) is or is to be expected to be greater than a predetermined threshold value.

7. Method according to one of the preceding claims, wherein the system (100) comprises or is designed as an electrohydrostatic actuator, wherein the electrohydrostatic actuator comprises an electric drive (120), one or more control valves and an actuator (126), and / or wherein the system is an underwater system.

8. The method according to any one of the preceding claims, wherein the one or more operating variables (220) of the system comprise one or more of the following operating variables: a position of the actuator, one or more pressures in the system.

9. Method according to one of the preceding claims, wherein the one or more control variables (230) for the system comprise one or more of the following control variables: a speed of the electric drive, and a position of one or more control valves.

10. A computing unit (110) comprising a processor configured to carry out the method according to any one of the preceding claims.

11. System (100), in particular subsea system, with one or more sensors (130) for detecting values ​​of one or more operating variables (220) of the system, with a drive (120), and with a computing unit (110) according to claim 10 12. A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the method according to claims 1 to 9.

13. A computer-readable data carrier on which the computer program according to claim 12 is stored.

Images