Method, computer program, computer readable medium, device and drive system for determining controller configuration for drive system

Abstract

The invention relates to a method for determining a controller configuration (10) for a drive system (1) having a drive (2) using at least one simulation model (7), in which step S1) one or more controlled system measurements (6) are received, the or each controlled system measurement (6) is obtained by measuring a rotational speed control circuit (3) of the drive system (1), step S2) creating a simulation model (7) for the or each rotational speed control circuit (3) having a drive sub-model (8) and one or more controlled system sub-models (9), a system identification is carried out using the controlled system measurement or individual controlled system measurements (6), and step S3) determining a controller configuration (10) of the drive system (1) using a simulation model (7). The invention further relates to a computer program, a computer-readable medium, a device for data processing, and a drive system (1).

Description

Technical Field

[0001] The invention relates to a method for determining a controller configuration of a drive system. The invention also relates to a computer program, a computer-readable medium, a data processing device and a drive system. Background Art

[0002] Complex drive systems, such as printing presses, require specific controller settings for proper operation due to their underlying mechanical controlled system behavior. Due to the huge variability of the drive chain, experts are often required to parameterize and optimize the drives. Due to the complexity and variability of the underlying system, parameterization and commissioning are very time-consuming even for experts.

[0003] Existing controller optimization algorithms are often insufficient to fully cover this variability, requiring expert analysis and optimization.

[0004] Furthermore, manual optimization requires constant accessibility to the machine, as it is an iterative process in which the chosen settings need to be constantly checked by measuring. These waiting times and machine adjustments result in downtime for the end customer, as production is not possible during this time.

[0005] Since changes also occur in the system, the problem usually consists in that the controlled system behavior must be continuously monitored during operation and the current controller configuration may need to be adapted accordingly. Therefore, pure optimization just for the sake of debugging is not sufficient for many situations and applications and requires more effort.

[0006] There remains a particular challenge when it comes to finding a common suitable controller setting for similar drive systems. Different load configurations may arise, for example, due to operating time, aging or retrofitting. The controller settings then have to be rechecked by experts and adapted if necessary, which is very time-consuming. Ideally, one would be able to determine a suitable controller setting for multiple drive systems without having to make constant adjustments in the event of changes, especially on individual axes. Sometimes using expert knowledge it is possible to find suitable controller settings for different load configurations. However, the required validation of the settings is very time-consuming, since tests have to be carried out for all load configurations.

[0007] In many applications, system changes may occur multiple times a day. Depending on the product being manufactured, load inertia or other system parameters (stiffness, damping) may change, which poses a challenge to existing controls. This greatly increases complexity, and in the absence of specific controller configuration guidelines, commissioning personnel can only rely on their own experience and intuition.

[0008] Previous solutions often only addressed part of the commissioning process and still required specialist expertise.

[0009] The applicant knows a method in which the machine is analyzed by an automatic controller optimization algorithm, the controller settings are optimized by measuring the speed control loop and applied directly in the drive. This requires connection to the actual machine, which can lead to downtime and therefore downtime.

[0010] It is also known that controller optimization can be performed using simulations. This requires the creation of a model of the respective machine, which requires that the main mechanical characteristics and parameters are known. This relies on expert knowledge.

[0011] EP 3 518 051 A1 discloses a detection method for optimizing the operation strategy of a processing machine. The method itself does not implement the optimization. This remains the responsibility of the user. However, the user will obtain information about which parameters need to be changed to optimize the operation of the machine, as well as matters that need to be paid attention to during the optimization process.

[0012] EP 3 451 100 A1 describes a method for operating a machine. The method includes normal operation and special operation. In special operation, the control device of the machine obtains actual values ​​through set values ​​and determines a frequency characteristic curve of an implementer based on the set value sequence given in the special operation and the corresponding actual values ​​obtained. Based on the frequency characteristic curve and the parameters of the controller structure, the control device evaluates the implementer and / or the controller structure and decides whether and what kind of message to send to the machine operator or to the computing device via a computer network based on the evaluation result. The method realizes early detection of problematic machine states in a simple and reliable manner.

[0013] According to EP 2 690 513 Bl, a method for condition monitoring of a program-controlled machine is proposed. The method comprises providing a dynamic model of the machine or a part thereof, wherein the dynamic model comprises model parameters such as mass inertia, spring stiffness or damping values, and performing a frequency analysis on the machine or the part thereof. Based on the frequency analysis, at least one new value of at least one model parameter is determined and the new value is compared with the original value. Based on the comparison result, the state of the machine is identified.

[0014] From US patent 2008 / 183311 Al a system, a method and a computer program for automatic closed-loop identification of industrial processes in a process control system can be seen.

[0015] EP 2 835 227 A1 describes a robot control device for controlling a robot arm, which device has an elastic mechanism between the rotating shaft of a motor and the rotating shaft of a connection.

[0016] EP 3 349 078 A1 discloses a diagnostic device and a method for monitoring and / or optimizing a control device.

[0017] A method for automatically adjusting a speed controller in a resilient mechanically controlled system is disclosed in DE 197 57 715 A1. Summary of the invention

[0018] Based on the prior art, an object of the present invention is to provide a method for determining a controller configuration, which is relatively simple and easy to implement, relies on professional knowledge as little as possible, does not affect the normal operation of the drive system as much as possible, and can achieve controller optimization even under different load configurations.

[0019] This object is achieved by the method of claim 1. A method for determining a controller configuration for a drive system having a drive using at least one simulation model is disclosed, wherein:

[0020] S1) receiving one or more controlled system measurement results, wherein the controlled system measurement result or each controlled system measurement result is or has been obtained by measuring a speed control loop of the drive system, preferably wherein the controlled system measurement result or each controlled system measurement result constitutes or comprises at least one metrologically obtained frequency response of the speed control loop or each speed control loop,

[0021] S2) creating a simulation model for the speed control loop or each speed control loop, the simulation model having a drive submodel and one or more controlled system submodels, wherein, in order to obtain the controlled system submodel or each controlled system submodel, a system identification is performed using the controlled system measurement results or each controlled system measurement results, and

[0022] S3) Determining a controller configuration for the drive system using the simulation model.

[0023] In other words, the method of the invention allows a defined controller optimization of a drive system taking into account measurement and adjustment criteria based only on at least one specific actual measurement of the drive system. A digitized image of the drive and (multiple) speed control loops is created and used as a dynamic simulation model, from which the optimized controller configuration can then be determined. The simulation model used according to the invention comprises a drive submodel and one or more controlled system submodels. In the case of multiple controlled system measurement results, the controlled system submodel depicts the mechanical and mechanical behavior. According to the invention, the controlled system submodel is determined based on a general system identification. The basis is the actual measurement of (multiple) speed control loops. In combination with the digitized image of the drive, a dynamic model is generated, which can provide an optimal basis for further designing control parameters. Due to the detailed stored simulation model, high-precision simulations can be created, thereby ensuring reliable optimization of the controller configuration. In step S3, the controller parameters can be optimized according to user-specified criteria based on the determined mathematical model, and then simulated and verified using the simulation model, in particular using discrete control loops, which is a preferred method. Examples of feasible criteria include amplitude overshoot and amplitude and phase margins.

[0024] In addition to the higher accuracy than conventional methods, the method according to the invention has other advantages. Since the identified controlled system can be virtually designed and simulated based on a validated simulation model, downtime for actual machine commissioning can be significantly reduced. Controller settings can be tested on the same virtual controlled system submodel without intervening in operation. Access to the actual drive system is not required for carrying out the method according to the invention. Only at least one actual measurement of the speed control loop needs to be carried out to obtain the controlled system measurement results, which can also be done in advance. The method according to the invention can be carried out at almost any time independently of the normal operation of the drive system. It can even be carried out without access to or contact with the drive system.

[0025] In particular, when the method according to the invention is integrated into an edge environment, a continuous, preferably continuous, optimization of the controller configuration can also be performed. The user can perform the optimization himself without expert knowledge. During normal operation of the drive system, the current controller configuration can also be checked and adapted to changes in the system, for which purpose new (individual) speed control loop measurements can be performed at regular or user-specific intervals. Individual measurement phases can be merged or supplemented into the corresponding operating phase.

[0026] The method of the present invention also has very broad application. It is not limited to or focused on a specific machine type.

[0027] For the first time, it is also possible to combine a plurality of models and measurements for controller settings, for example in the case of tool changes and / or changes in other system parameters. The method according to the invention has also proven to be particularly suitable for determining an optimized common controller configuration for different load configurations, which may arise, for example, as a result of a reassembly process. Thus, according to the invention, in step S1 a plurality of controlled system measurement results for different speed control loops are received, in step S2 a simulation model containing a controlled system submodel is created for each speed control loop, and in step S3 a common controller configuration suitable for all speed control loops is determined using the simulation model containing a plurality of controlled system submodels.

[0028] The controlled system measurement results may be provided or received, for example, by an engineering system or an engineering platform. The engineering system or the engineering platform may also be used to perform measurements of various speed control loops to obtain respective controlled system measurement results.

[0029] The frequency response curves of all individual speed control loops are preferably superimposed to determine an envelope curve, and based thereon a stable controller configuration is determined as a common controller configuration for all systems.

[0030] Different speed control loops which provide measurement results for a plurality of controlled systems have different designs. In particular, they can differ in at least one component. For example, modifications to the actual drive system due to a retrofit may need to be taken into account. By way of example only, a printing press may need to be operated with different, in particular differently sized, cylinders. An optimized, universal controller configuration can then be found by the method according to the invention, which is suitable for all cylinder sizes. The effort required to find a comprehensive controller configuration is relatively small. In the case of a retrofit, a complex re-parameterization of the controller can also be dispensed with.

[0031] Another advantageous embodiment is characterized in that the or each controlled system measurement result is or has been obtained by measuring the or each speed control loop with a defined excitation, in particular a defined noise excitation, preferably a pseudo-random noise excitation. This has proven to be particularly suitable.

[0032] Alternatively or additionally, it is proposed that for the system identification, the number of poles and zeros is defined for the or each controlled system measurement result and / or the number of poles and zeros is determined in the or each controlled system measurement result. This can be done manually by a user or automatically.

[0033] System identification for the or each controlled system submodel may also comprise providing a controlled system submodel with unknown model parameters and determining the model parameters using one or more controlled system measurements, preferably by mathematical approximation of the measurement curves. Mathematical optimization methods may be used to approximate the model characteristics so as to conform to the or each controlled system measurements, in particular according to a user-defined quality criterion. Merely as an example, a controlled system submodel of a multi-mass, in particular a dual-mass oscillator is mentioned, which model has unknown inertia, stiffness and damping parameters that need to be optimized. The optimization may be performed by a numerical optimization algorithm that minimizes the deviation between the (individual) controlled system measurements, in particular the (individual) measured frequency characteristic curve, and the characteristic curve set by the parameters (e.g. Nelder Mead algorithm).

[0034] System identification for the or each controlled system submodel may also include iterative adjustment of model parameters, in particular until a desired result is obtained, the process continuing until a user-defined quality criterion is met or a settable maximum number of iterations is reached.

[0035] In a further embodiment of the invention, the drive submodel is created, in particular as a mapping of the control and regulation of the frequency converter, using known drive parameters, preferably using known motor parameters of the drive motor. For example, the respective information can be provided, queried or received from an engineering system or an engineering platform. The drive submodel preferably reproduces the behavior required for the control and regulation of the frequency converter so that the torque value matches the speed. The implementation of the drive submodel and its parameterization are preferably carried out in the same way as the actual drive. In the process, the time characteristics of the control cascade and all filters occurring are preferably re-implemented in a similar way to the actual model.

[0036] The controller configuration determined in step S3 can comprise, in a known manner, controller parameters of the drive system, in particular the drive, and current setpoint filters.

[0037] The controller configuration is transferred to the actual drive system in an advantageous manner and controlled accordingly. It can also be introduced into the engineering system or engineering platform.

[0038] Another subject of the invention is a computer program comprising program code means for causing at least one computer to carry out the steps of the method according to the invention, when executed on at least one computer.

[0039] The invention also relates to a computer-readable medium comprising instructions which, when executed on at least one computer, cause the at least one computer to implement the steps of the method of the invention.

[0040] The computer readable medium may be, for example, a CD-ROM, a DVD, a USB or a flash memory. It should be noted that the computer readable medium should not be understood as only a physical medium, but may also exist in the form of a data stream and / or a signal representing a data stream.

[0041] The present invention also relates to a data processing device, comprising

[0042] - processor, and

[0043] - a data storage device on which a computer executable program code is stored, which, when executed by a processor, causes the processor to carry out the steps of the method of the invention.

[0044] In a preferred development, the device is designed as an edge device.

[0045] Finally, a subject of the invention is a drive system which is designed and configured to implement the method according to the invention. The drive system comprises a data processing device according to the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Other features and advantages of the present invention will become clear through the following description of the embodiments of the present invention in conjunction with the accompanying drawings.

[0047] FIG. 1 is a merely schematic representation of the steps of an embodiment of the method of the present invention. DETAILED DESCRIPTION

[0048] FIG. 1 shows a real drive system 1 and its drive 2 in the form of a purely schematic block diagram. In addition, three real speed control loops 3 of the drive system 1 are shown, which are different from one another and correspond specifically to three different load configurations. The first speed control loop 3 can correspond, for example, to an initial state during commissioning, while the other two are configurations that have been changed, for example, due to retrofit measures. Merely as an example, it is mentioned that a printing press including the drive system 1 can be operated with three different cylinders.

[0049] It should be emphasized that the three speed control loops 3 are merely exemplary and may be any other number, or even only one speed control loop 3 .

[0050] FIG1 also shows an engineering platform 4 connected to the drive system 1 and a computer program 5, which contains program code means, which, when executed on at least one computer, causes at least one computer to implement the steps described below and also shown schematically in the figure. The drive system 1 can include a data processing device, which has a processor and a data storage device, wherein program codes executable by each computer are stored, which, when executed by the processor, cause the processor to implement the described steps. Such a device can also be designed independently of the drive system 1. For example, it can also be an edge device.

[0051] In step S1, the computer program 5 receives a plurality of controlled system measurement results 6, which are obtained after measuring three speed control loops 3 of the drive system 1, the measurements being carried out under a defined noise excitation, in this case a pseudo-random noise excitation. The user can import the controlled system measurement results 6, for example, from the engineering portal 4. The controlled system measurement results 6 are respectively composed of the measured frequency responses of the various speed control loops 3. In this case, the frequency responses 6 have been recorded by the engineering platform 4 before step S1, which contains the relevant functions in a known manner. In the embodiment described here, the frequency responses representing the controlled system measurement results 6 have been recorded before the method is implemented, i.e. before the method is performed. The method steps described here can be implemented completely independently of the operation of the drive system 1 and without access to the system.

[0052] In step S2, a simulation model 7 is created, which contains a drive submodel 8 and a plurality of (here three) controlled system submodels 9. Each controlled system submodel 9 corresponds to a speed control loop 3. To obtain the controlled system submodel 9, a general system identification is performed using the individual controlled system measurement results 6. The user defines the number of poles and zeros (model complexity) that he expects for the individual speed control loops 3. Subsequently, the system identification is performed. First, a controlled system submodel with unknown model parameters is provided for each speed control loop 3, and the model parameters are determined using the individual controlled system measurement results 6. To this end, the individual model parameters are continuously adjusted by a numerical optimization algorithm until the measured frequency characteristic curve of the controlled system is as consistent as possible with the modeled one, or the quality criteria specified by the user are met.

[0053] Once the speed control loop 3 has been identified, in other words the mechanical system and the mechanical behavior, the motor data such as the motor type, the number of pole pairs and the motor constants are transferred from the engineering platform 4 or manually by the user to the drive 2 and a relevant drive submodel 8 is created using the known drive parameters, in particular as a mapping of the frequency converter control and control. The implementation of the drive submodel 8 and its parameterization are preferably carried out in the same way as the actual drive. In the process, the time characteristics of the control cascade and all filters present are preferably also re-implemented in a similar way to the actual model.

[0054] In step S3, the discrete control loops are simulated and verified using the simulation model 7 including the drive submodel 8 and the controlled system submodel 9, and a suitable controller configuration 10 for the drive system 1 is determined. After the optimization, the frequency characteristic curves of the open-loop and closed-loop control loops of the optimized system are displayed for user verification. As an optimized system, in particular, the overall system of the drive 2 and multiple controlled systems 3 after the controller parameter optimization is considered. The optimized parameters are preferably introduced into the drive submodel 8, and the optimized system is simulated again. Simulations with the respectively identified controlled system submodels 9 can be implemented in the simulation model 7 to verify the time behavior of the optimized system in different application-related scenarios and applications. For example, step responses or setpoint curves from actual control can be simulated.

[0055] An average, stable controller configuration 10 suitable for all three speed control loops 3 is generated as a common controller configuration 10. The common controller configuration 10 includes filter and controller settings, can adjust all three speed control loops 3 in the same way, and provides optimized system performance. For example, when a different speed control loop 3 is obtained due to a retrofit, no re-parameterization is required. The controller configuration is still applicable.

[0056] As shown by the arrow in FIG. 1 , a generic controller configuration 10 may be introduced into the engineering platform 4 .

[0057] In step S4 , the determined controller configuration 10 is transferred to the drive system 1 and controlled accordingly. The result is an optimized system behavior that applies to all three speed control loops 3 .

[0058] By implementing the method, downtimes for real machine commissioning can be significantly reduced, since the identified controlled system can be virtually designed and simulated based on a validated simulation model 7. This allows high accuracy to be achieved without requiring intervention in operation.

[0059] Although the present invention has been described in detail by way of preferred embodiments, the present invention is not limited to the disclosed examples and a person skilled in the art may derive other variants therefrom without departing from the scope of protection of the present invention.

Claims

1. A method for determining a controller configuration (10) for a drive system (1) having a drive (2) using at least one simulation model (7), wherein: S1) receiving one or more controlled system measurement results (6), wherein: The controlled system measurement result or each of the controlled system measurement results (6) is obtained or has been obtained by measuring a speed control loop (3) of the drive system (1), preferably wherein the controlled system measurement result or each of the controlled system measurement results (6) constitutes or comprises at least one frequency response of the speed control loop or each of the speed control loops (3) obtained in a metrological manner, S2) creating a simulation model (7) for the speed control loop or each of the speed control loops (3), the simulation model having a drive submodel (8) and one or more controlled system submodels (9), wherein, in order to obtain the controlled system submodel or each of the controlled system submodels (9), a system identification is performed using the controlled system measurement results or each of the controlled system measurement results (6), and S3) determining a controller configuration (10) for the drive system (1) using the simulation model (7), It is characterized in that in step S1, a plurality of controlled system measurement results (6) associated with different speed control loops (3) are provided, and in step S2, a simulation model (7) with a corresponding controlled system submodel (9) is created for each of the speed control loops (3), and in step S3, a common controller configuration (10) applicable to all speed control loops (3) is determined, wherein the various speed control loops (3) have different designs.

2. The method according to claim 1, characterized in that A controller configuration (10) that is stable and averaged over all speed control loops (3) is determined as a common controller configuration (10).

3. The method according to claim 1 or 2, characterized in that: The various speed control loops (3) differ in at least one component.

4. The method according to any one of the preceding claims, characterized in that By measuring the speed control loop or each speed control loop (3), the controlled system measurement result or each controlled system measurement result (6) is or has been obtained after a defined excitation, in particular after a defined noise excitation, preferably after a pseudo-random noise excitation.

5. The method according to any one of the preceding claims, characterized in that For system identification, the number of poles and zeros is defined for the or each of the controlled system measurement results (6) and / or the number of poles and zeros is determined in the or each of the controlled system measurement results (6).

6. The method according to any one of the preceding claims, characterized in that The system identification for the controlled system submodel or each of the controlled system submodels (9) involves providing a controlled system submodel with unknown model parameters and determining the model parameters using the controlled system measurement results or each of the controlled system measurement results (6), preferably wherein a mathematical optimization method is used to approximate the model characteristics so that the model characteristics correspond to the controlled system measurement results or each of the controlled system measurement results (6), in particular according to user-defined quality criteria.

7. The method according to any one of the preceding claims, characterized in that The system identification for the or each controlled system submodel (9) involves iterative adjustment of model parameters, in particular until a desired result is obtained.

8. The method according to any one of the preceding claims, characterized in that The drive sub-model (8) is created using known drive parameters, in particular known motor parameters of a motor of the drive (2).

9. The method according to any one of the preceding claims, characterized in that The method is implemented independently of the operation of the drive system (1), and / or the method is implemented without being connected to the drive system (1).

10. The method according to any one of the preceding claims, characterized in that In step S3, the decentralized control loop is simulated and verified using the simulation model (7).

11. The method according to any one of the preceding claims, characterized in that The controller configuration (10) determined in step S3 comprises controller parameters and current setpoint filters for the drive system (1), in particular for the drive (2), and / or the determined controller configuration (10) is transmitted to the drive system (1) and the drive system is controlled accordingly.

12. A computer program comprising instructions which, when executed on at least one computer, cause the at least one computer to implement the steps of the method according to any one of claims 1 to 11.

13. A computer-readable medium comprising instructions which, when executed on at least one computer, cause the at least one computer to implement the steps of the method according to any one of claims 1 to 11.

14. A data processing device comprising - processor, and - a data storage device having stored thereon a computer-executable program code which, when executed by the processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 11.

15. A drive system (1) comprising a drive and a data processing device according to claim 14.

Images