Method for the automated adjustment of a computer-implemented controller module

Abstract

The invention relates to a method for the automated adjustment of a computer-implemented controller module (3) which can be used in the context of a guidance system (4) of a technical installation (2), comprising the following steps: a) carrying out a process (1) in the technical installation (2) and using the computer-implemented controller module (3) to control the process (1), b) monitoring process variables of the computer-implemented controller module (3) by means of a computer-implemented controller service (5) and collecting the process values (7) resulting therefrom by means of the computer-implemented controller service (5), c) analyzing the process values (7) for changes in the process (1) by means of the computer-implemented controller service (5), d) if the changes exceed a specific threshold value, a recording is started and process values (7) are automatically detected by the computer-implemented controller service (5), e) identifying and creating a process model of a part of the process (1) controlled by the computer-implemented controller module (3) by means of the computer-implemented controller service (5) with the inclusion of the process values (7) detected in the previous step, f) generating optimal controller parameters (8) on the basis of the created process model by means of the computer-implemented controller service (5); g) transmitting the generated controller parameters (8) to the computer-implemented controller module (3) by means of the computer-implemented controller service (5).

Description

[0001] 202501325 Foreign version 05.02.2026

[0002] 1

[0003] Description

[0004] Method for the automated adjustment of a computer-implemented controller module

[0005] DESCRIPTION INTRODUCTION

[0006] The invention relates to a method for the automated setting of a computer-implemented controller module that can be used in the context of a control system of a technical plant.

[0007] Furthermore, the invention relates to a computer-implemented controller module for implementing the method.

[0008] STATE OF THE ART

[0009] Standard linear controllers, such as PI, PD, or PID controllers, are widely used, especially in industrial environments, and are an important component in the automation of technical systems. A well-tuned controller contributes to achieving good product quality and high throughput while simultaneously reducing energy consumption. However, in many technical systems, standard linear controllers are not optimally configured due to slow process changes, lack of expertise, or insufficient time. Subsequent adjustment or fine-tuning of standard linear controllers during operation is often not possible, as the need to use characteristic process excitations for adjustment leads to disruptions in the production process and consequently to inferior or even unusable products.

[0010] WO 03081348 A1 describes a method for controlling a component during the operation of a technical plant using a PI controller.

[0011] DE 4433332 A1 describes a device for the adaptive control of a system, in which a model of the system can be dispensed with and which still ensures good control behavior.

[0012] US 2009 / 0112335 A1 describes a process control system in which a controller detects changes in the process, collects process data, and determines new model parameters using a model-based identification method. Model identification is performed using [202501325 Foreign Version 05.02.2026].

[0013] 2

[0014] Several simulations of different process models are performed, and their results are subsequently compared to derive new model data. Data recording in D1 is triggered exclusively by specific events, such as changes in the setpoint, changes in the controller output, or by injecting small output signals ("pulse injection"). D1 further describes how, after model identification, the controller parameters can be automatically adjusted.

[0015] Ohnishi et al., 2011, present a performance-adaptive PID control concept in which a performance index is regularly monitored. If the index exceeds a predefined threshold, a system identification is triggered, followed by a recalculation of the PID parameters. The publication primarily addresses theoretical and experimental methods for model identification and PID retuning in an academic setting.

[0016] The aforementioned documents each describe methods or control concepts in which controller parameters are updated based on process data. The WO document deals with PI control during plant operation, the DE document with a model-free adaptive control approach, while D1 discloses a process control system with model-based identification after trigger events, and D2 describes a power-adaptive PID concept based on a control-performance-oriented threshold. However, none of these documents discloses the specific combination of process steps claimed in the present invention, in particular not the analysis of process values ​​for sufficient dynamic system properties as a decision criterion for initiating model identification, nor the other measures described within the scope of the invention for the safe and reliable automated parameterization of a computer-implemented controller module.

[0017] The invention is based on the objective of providing a method for the automated adjustment of a computer-implemented controller module that can be used in the context of a control system of a technical plant.

[0018] SOLUTION TO THE TASK

[0019] This problem is solved by a method for the automated adjustment of a computer-implemented controller module with the features of claim 1. Furthermore, the problem is solved by a computer-implemented controller module for implementing the method-202501325 Foreign version 05.02.2026

[0020] 3

[0021] rens according to claim 9 and by a server for a control system with a computer-implemented controller module according to claim 10. Advantageous further developments result from the dependent claims.

[0022] DESCRIPTION OF THE INVENTION

[0023] The inventive method for the automated configuration of a computer-implemented controller module, which can be used in the context of a control system of a technical plant, starts with the execution of a process in the technical plant and the use of the computer-implemented controller module to control the process. Process parameters of the computer-implemented controller module are monitored by a computer-implemented control service, and the resulting process values ​​are collected by the computer-implemented control service. In the subsequent step, the process values ​​are analyzed for changes in the process by the computer-implemented control service.

[0024] If the changes exceed a certain threshold, a recording is initiated, and process values ​​are automatically acquired by the computer-implemented controller service. Subsequently, a process model of the portion of the process controlled by the computer-implemented controller module is identified and created by the computer-implemented controller service, incorporating the process values ​​acquired in the previous step. Based on this process model, optimal controller parameters are generated by the computer-implemented controller service, and these parameters are then transferred to the computer-implemented controller module.

[0025] The process model identification process begins with an evaluation of the process values ​​to determine whether the system's dynamic properties are sufficient. Once this preliminary assessment is complete, the process model is then identified. This can involve, for example, using a PT1Tt or PT2Tt model, or a data-driven machine learning model.

[0026] The optimal controller parameters are derived from the created process model primarily in two ways. First, there are theoretical tuning rules that, depending on a specific process model type, directly determine the controller parameters. Second, the controller parameters can be determined through optimization based on a process model. 202501325 Foreign version 05.02.2026

[0027] 4

[0028] This has the advantage that different process models or process values ​​from different times can be used, allowing a controller to be designed that is able to deliver stable and satisfactory performance despite uncertainties and variations in the process model or process conditions.

[0029] The technical installation can be a plant from the process industry, such as a chemical, pharmaceutical, petrochemical, or food and beverage industry plant. This also includes any plant from the manufacturing industry, such as factories where cars or goods of all kinds are produced. Technical installations suitable for carrying out the process according to the invention can also originate from the energy generation sector. Wind turbines, solar power plants, or power plants for energy generation are likewise included in the term "technical installation."

[0030] In this context, a control system is understood to be a computer-based, technical system that includes functionalities for displaying, operating, and controlling the technical plant and directing process objects within a process. Process objects include, for example, peripheral devices such as sensors for acquiring measured values, as well as various actuators. Furthermore, the control system can include so-called process- or production-related components that serve to control the actuators or sensors. Optionally, the control system can also include additional computing units for more complex control systems and systems for data storage and processing.

[0031] The computer-implemented controller can be a standard linear controller, such as a PI, PD, or PID controller, used to regulate a process that needs to be stably set to and maintained at a setpoint of a process variable, such as temperature. A process can, for example, be a production process, particularly in the process or manufacturing industries. The computer-implemented controller can be implemented on a computing unit within the technical system or on an external computing unit connected to the technical system.

[0032] The process values ​​are measured values ​​that the computer-implemented control system acquires from the process, for example, through sensors attached to the technical equipment. The computer-implemented control system continuously checks whether a control deviation exists, which corresponds to the difference between a desired setpoint and the current measured value. 202501325 Foreign version 05.02.2026

[0033] 5

[0034] A process model is a mathematical representation of a dynamic system or process, describing its essential properties and behavior. Using this process model, system behavior can be analyzed and simulated, and subsequently, control strategies can be developed to maintain stable process values.

[0035] A dynamic system is understood to be a system whose state changes over time and whose behavior is described by the relationship between input and output variables, in particular by differential equations.

[0036] With regard to the computer-implemented controller, the controller parameters determine the performance and behavior of the controller. Depending on the selected controller (PI, PD, or PID controller), these parameters include the following:

[0037] • The proportional gain (P) determines how strongly the controller reacts to the current control deviation, i.e., the difference between the setpoint and the actual value. A higher P-value results in a stronger reaction to deviations.

[0038] • The integral part (I) takes into account the sum of past rule deviations.

[0039] It helps to eliminate persistent control deviations by increasing the controller's response over time.

[0040] • The differential component (D) responds to the rate of change of the control deviation. It helps to stabilize the system and reduce overshoot.

[0041] The correct adjustment of the controller parameters is of particular importance for the performance of the process or the control system, which is why they must be carefully adjusted.

[0042] Preferably, the method is designed such that the process variables monitored by the computer-implemented controller include current process excitation and / or current performance. The current process excitation and / or current performance makes it possible to identify when new process parameters are needed and when a recalculation of the process parameters is required. For this to work, the process values ​​must be continuously monitored to identify time periods with sufficient dynamic system properties. The process values ​​are collected and included in the calculation when a recalculation is necessary. The recalculation of the process parameters can be performed using the following method: 202501325 Foreign version 05.02.2026

[0043] 6

[0044] can be initiated by process stimulation and / or performance, or, for example, manually at the request of a plant manager or an operator of the control system.

[0045] Process excitation refers to the targeted influencing of a system through external signals in order to analyze and model the system's behavior. This excitation is necessary to identify the system's dynamic properties and to develop suitable control strategies.

[0046] The performance of a control system refers to its ability to achieve desired objectives, such as stability, accuracy, and response time. A well-tuned control system ensures stability, resisting oscillations or instabilities and thus minimizing the deviation between the setpoint and the actual value. The setting of the controller parameters is particularly important here; for example, the proportional, integral, and differential components in a PID controller. The control system is influenced by external factors and disturbances, the inertia, delay, and nonlinearities of the system being controlled, and the accuracy and response time of the sensors and actuators used.

[0047] In an advantageous further development of the invention, the control system is designed such that the computer-implemented controller is a PID controller, a PI controller, or a PD controller. These controller types are widely used in industrial and technical applications due to their adaptability to various system requirements and are characterized in particular by good control accuracy, which minimizes control deviation, maintains system stability, and reduces unwanted oscillations. Furthermore, with these controller types, the controller parameters can be determined by optimization based on a process model or by theoretical tuning rules that, depending on a specific process model type, directly derive the controller parameters.

[0048] The procedure can advantageously be designed so that, in the event of insufficient current process excitation and / or insufficient current performance of the computer-implemented controller module, a recalculation of the controller parameters is initiated based on the already acquired, but not yet considered, process values. The computer-implemented controller service regularly collects and stores process values ​​of the computer-implemented controller.

[0049] 7

[0050] Controller module. If suitable process values ​​are available since the last process parameter creation that have not yet been used in the recalculation, these process values ​​are used to recalculate the controller parameters. This ensures that a recalculation of the controller parameters is always possible if the current process excitation and / or the current performance of the computer-implemented controller module is insufficient.

[0051] Advantageously, the procedure is designed such that if the controller parameters are no longer valid or insufficient, or will not be recorded in the near future, a recalculation of the controller parameters is initiated based on the already recorded, but not yet considered, process values. The computer-implemented controller service regularly collects and stores process values ​​from the computer-implemented controller module. If suitable process values ​​are available since the last process parameter creation that have not yet been used in the recalculation, these process values ​​are used to recalculate the controller parameters. This ensures that a recalculation of the controller parameters is always possible if they are no longer valid or insufficient.

[0052] Preferably, the method is designed such that the process model is identified and created without influencing the process and in the background of the process of the technical plant on an engineering station, on an edge computing platform or on an external server which is connected to a server of the control system of a technical plant.

[0053] This is advantageous because an edge computing platform and an engineering station can reliably continue operating even during network outages. An external server offers high availability through redundant systems. Low latency is also crucial for the method described here. An edge computing platform and an engineering station offer low latency because data processing takes place closer to the data source. External servers can also offer low latency through geographically distributed data centers.

[0054] In an advantageous further development of the invention, the method is designed such that, after the recalculation of the controller parameters is completed, the new controller parameters are checked for functionality by the computer-implemented controller service before being transmitted to the computer-implemented controller module. This procedure increases process reliability and minimizes process failures. 202501325 Foreign version 05.02.2026

[0055] 8

[0056] The procedure can advantageously be designed so that the new controller parameters are automatically transferred to the computer-implemented controller module and / or after approval by an operator of the control system.

[0057] Advantageously, the procedure is designed such that the analysis of the process values ​​includes a check to determine whether the process values ​​exhibit sufficient excitation or variation range for the creation of the process model. This ensures that the changes occurring in the process reflect sufficiently dynamic behavior of the system, enabling the identification of a meaningful process model. Such sufficient excitation or variation range is present, in particular, when the process values ​​show temporal or numerical changes from which relevant dynamic properties of the system to be controlled can be derived. Without such excitation, model identification may be inaccurate or incomplete, which in turn can lead to inadequate or unstable controller parameters.The upstream verification of process values ​​ensures that only data capable of reliably describing system behavior is used for model building, thus supporting a precise derivation of the controller parameters. This increases the robustness of the overall process and contributes to stable and reliable process control.

[0058] The previously formulated technical task is solved by a computer-implemented controller module for implementing the process, as well as a server for a control system with a computer-implemented controller module on that server. The server can be an operator station server and / or an engineering station server.

[0059] FIGURE DESCRIPTION

[0060] The invention will now be described and explained in more detail with reference to the exemplary embodiments shown in the figures.

[0061] FIG 1 shows an arrangement of the building blocks according to the invention in order to be able to carry out the method according to the invention,

[0062] FIG 2 Procedure steps for the automated configuration of a computer-implemented controller module. 202501325 Foreign version 05.02.2026

[0063] 9

[0064] FIG. 1 shows an example of the components according to the invention and their connections to one another. The starting point is the execution of a process 1 using a technical system 2. A computer-implemented controller 3 is used to control the process 1. A process 1 can, for example, comprise a production process, particularly in the process or manufacturing industry. The technical system 2 is connected to a control system 4, via which the process 1 or the technical system 2 is monitored, set, and controlled. A specific process variable, such as a temperature, pressure, or fill level, is to be set and maintained stably at a setpoint value at the technical system 2 or at the computer-implemented controller 3. The process variable is monitored by a computer-implemented controller 5.For this purpose, the technical system 2 is equipped with sensors 6 that detect process values ​​7 such as temperature, humidity, pressure, or brightness. This information is collected by the computer-implemented control system 5 for further processing.

[0065] The computer-implemented control service 5 analyzes the received process values ​​7 for changes or deviations in process 1. If the changes exceed a certain threshold, the computer-implemented control service 5 initiates a recording and automatically acquires process values ​​7. Subsequently, using the acquired process values ​​7, the computer-implemented control service 5 identifies and creates a process model of a part of process 1 controlled by the computer-implemented controller block 3. Based on the created process model, optimal controller parameters 8 are generated and transferred by the computer-implemented control service 5 to the computer-implemented controller block 3, thereby adjusting 9 the manipulated variables (e.g., valve opening) of the technical system 2, which then influence the process variables.

[0066] Figure 2 shows a process sequence for the automated configuration of a computer-implemented controller module 3. The computer-implemented controller module 3 can be used in the context of a control system 4 of a technical plant 2 and comprises the following steps:

[0067] • Execution of process 1 in the technical plant 2 and use of the computer-implemented controller module 3 to control process 1,

[0068] • Monitoring of process variables of the computer-implemented controller module 3 by a computer-implemented controller service 5 and collection of the resulting process values ​​7 by the computer-implemented controller service 5,202501325 Foreign version 05.02.2026

[0069] 10

[0070] • Analysis of process values ​​7 for changes in the process by the computer-implemented control service 5.

[0071] • If the changes exceed a certain threshold, a recording is started and process values ​​7 are automatically recorded by the computer-implemented control service 5.

[0072] • Identification and creation of a process model of a part of process 1 controlled by the computer-implemented controller module 3 by the computer-implemented controller service 5 using the process values ​​7 recorded in the previous step.

[0073] • Generation of optimal controller parameters 8 based on the created process model by the computer-implemented controller service 5.

[0074] • Transfer of the generated controller parameters 8 by the computer-implemented controller service 5 to the computer-implemented controller module 3.

[0075] CONCLUDING REMARKS

[0076] Although the invention has been further illustrated and described by the preferred embodiment, the invention is not limited by the disclosed examples. Variations thereof can be derived by a person skilled in the art without departing from the scope of protection of the invention as defined by the subsequent claims. 202501325 Foreign version 05.02.2026

[0077] 11

[0078] Reference symbol list

[0079] 1 process

[0080] 2 technical system

[0081] 3 computer-implemented controller module 4 control system

[0082] 5 computer-implemented controller service 6 sensors

[0083] 7 process values

[0084] 8 controller parameters

[0085] 9. Adjustment

Claims

202501325 Foreign version 05.02.2026 12 Patent claims 1. Method for the automated setting of a computer-implemented controller module (3) that can be used in the context of a control system (4) of a technical plant (2), comprising the following steps: a) Carrying out a process (1) in the technical plant (2) and using the computer-implemented controller module (3) to control the process (1), b) Monitoring of process variables of the computer-implemented controller module (3) by a computer-implemented controller service (5) and collecting the resulting process values ​​(7) by the computer-implemented controller service (5), c) Analysis of the process values ​​(7) for changes in the process (1) by the computer-implemented control service (5), d) If the changes exceed a certain threshold, a recording is started and process values ​​(7) are automatically recorded by the computer-implemented control service (5), e) Identification and creation of a process model of a part of the process (1) controlled by the computer-implemented controller module (3) by the computer-implemented controller service (5) taking into account the process values ​​(7) recorded in the previous step, f) Generation of optimal controller parameters (8) based on the created process model by the computer-implemented controller service (5), g) Transfer of the generated controller parameters (8) by the computer-implemented controller service (5) to the computer-implemented controller module (3).

2. The method of claim 1, wherein the process variables monitored by the computer-implemented controller module (3) include current process excitation and / or current performance. 202501325 Foreign version 05.02.2026 13 3. The method of claim 1, wherein the computer-implemented controller module (3) is a PID controller or a PI controller or a PD controller.

4. Method according to claim 1, wherein, in the event of insufficient current process excitation and / or current performance of the computer-implemented controller module (3), a recalculation of the controller parameters (8) is started on the basis of the already recorded, but not yet taken into account, process values ​​(7).

5. Method according to claim 4, wherein, if the controller parameters (8) are no longer valid or insufficient or will not be recorded in the near future, a recalculation of the controller parameters (8) is started on the basis of the already recorded, but not yet taken into account, process values ​​(7).

6. Method according to claim 1, wherein the process model is identified and created without influencing the process (1) and in the background of the process (1) of the technical plant (2) on an engineering station, on an edge computing platform or on an external server which is connected to a server of the control system (4) of a technical plant (2).

7. Method according to claim 1, wherein, after the recalculation of the controller parameters (8) is completed, the new controller parameters (8) are checked for functionality by the computer-implemented controller service (5) before being transmitted to the computer-implemented controller module (3).

8. Method according to one of the preceding claims, wherein the new controller parameters (8) are automatically and / or after approval by an operator of the control system (4) are transferred to the computer-implemented controller module (3).

9. Method according to one of the preceding claims, wherein, as part of the analysis of the process values, it is checked whether the process values ​​(7) have a sufficient excitation or range of variation for the creation of the process model.

10. Computer-implemented controller module (3) for implementing the method according to one of the preceding claims. 202501325 Foreign version 05.02.2026 14 11. Server for a control system (4) with a computer-implemented controller module (3) according to claim 10.

Images