SYSTEM, METHOD AND COMPUTER PROGRAM FOR CONTROLLING A PRODUCTION PLANT CONSISTING OF MULTIPLE PLANT COMPONENTS, IN PARTICULAR A METALLURGICAL PRODUCTION PLANT FOR THE PRODUCE OF INDUSTRIAL GOODS SUCH AS METAL SEMI-CONDITIONED PRODUCTS AND / OR METAL FINISHED PRODUCTS

Description

[0001] The invention relates to a system for controlling a production plant consisting of several plant components, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products. The invention further relates to a method for controlling a production plant consisting of several plant components, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products, and a computer program for carrying out the method according to the invention, in particular by means of a system according to the invention.

[0002] The production of semi-finished and / or finished metal products involves a multi-stage process chain within a metallurgical production plant. For example, after the production of pig iron or the use of scrap in a steel mill, automotive components typically undergo the following process chain: continuous casting of slabs, hot rolling to 1 to 6 mm, pickling of the strip to remove scale, cold rolling to 0.3 to 3 mm, heat treatment in an annealing furnace to achieve the characteristic microstructure, heating of the strip to a coating temperature of 340 to 480 °C, hot-dip coating, cutting, forming, joining, and paint coating. This process utilizes, among other things, the following components of the metallurgical production plant: continuous casting, hot strip mill, pickling / tandem mill, strip galvanizing line, stamping / pressing line, welding machine, and dip coating.

[0003] The production of long products in a metallurgical production plant involves, for example, the following process chain: continuous casting of billets, hot rolling to 4 to 200 mm diameter round stock in several rolling stands, removal of the scaled surface, and forging. This process chain is represented, for example, by the following components of the metallurgical production plant: continuous billet casting, wire / bill mill, peeling line, and forging press.

[0004] From an abstract perspective, every product undergoes several transformation stages (plant components) before the final product is reached. Each transformation stage has an input quality window, an output quality window, and a process window. The input quality window defines the required quality characteristics of the input product. The output quality window defines criteria after each transformation stage (plant component) to check the quality of the intermediate product. In a production plant consisting of multiple plant components, the output quality window of an upstream transformation stage (plant component) is equivalent to the input quality window of the subsequent transformation stage (plant component). Each transformation stage (plant component) also has a process window, which defines the setpoints that can be implemented by the respective transformation stage (plant component) for the plant automation of that transformation stage (plant component).The set values ​​are adjusted so that the quality defined in the initial quality window of the conversion stage (plant section) is achieved by the conversion stage (plant section).

[0005] Through a transformation stage or plant component, a product changes its state from A to B. For example, in a pickling bath, the scale is removed after hot rolling; in an annealing furnace, the material properties are adjusted through controlled heating and cooling; or in a wire rolling mill, a wire with a diameter of, for example, 5.5 mm is produced from a billet (e.g., 150 x 150 mm). These are just a few examples of transformation stages or plant components in a metallurgical production plant.

[0006] After each conversion stage (plant component), criteria are defined that the product should possess after the conversion. These criteria have lower and upper limits within which the product's properties must lie. All criteria that must be met after a conversion stage (plant component) constitute the quality window. In a production plant consisting of several conversion stages (plant components), the output quality window of an upstream conversion stage (plant component) corresponds to the input quality window of the downstream conversion stage (plant component). In particular, the requirements (input quality window) of a downstream conversion stage (plant component) define the quality requirements (output quality window) for the upstream conversion stage (plant component).Examples of quality windows include the absence of residual scale on the product after pickling, the presence of yield strength and tensile strength within specified limits, or the absence of specified ovality (difference between minimum and maximum diameter) in round material. These are just a few examples of quality windows for conversion stages or plant components of a metallurgical production facility.

[0007] To achieve the desired initial quality after a conversion stage (plant component), the setpoints for the plant automation of that conversion stage (plant component) can be adjusted within the process window of that conversion stage (plant component). Examples of process windows for conversion stages (plant components) include the temperature of the acid in a pickling solution for cleaning scaled surfaces, the holding time in the annealing furnace to adjust the yield strength, or the number of passes between rolling stands to influence ovality. These are just a few examples of process windows for conversion stages or plant components in a metallurgical production plant.

[0008] In each conversion stage (plant section), the sequence of products to be manufactured is determined by a production planning system. For example, in a pickling plant, products with similar scale levels are processed sequentially to minimize necessary temperature changes in the acid within the pickling tanks. Similarly, in an annealing furnace, products with similar annealing temperatures can be produced consecutively, or in a wire rod mill, the product sequence is determined to minimize setup time for roll changes. These are just a few examples of production planning in the conversion stages (plant sections) of metallurgical plants.

[0009] Production planning is carried out for only one conversion stage (plant section), whereby the production sequence is optimized for this conversion stage (plant section). Furthermore, the planning of the production sequence typically assumes an ideal state of the conversion stage (plant section) and / or an ideal input product.

[0010] EP 3 358 431 A1 discloses that a production planning system for a basic materials plant determines its production planning data and provides it to the plant's automation system. A condition monitoring system determines past and future expected states of plant components. A quality assessment system determines the states of input products produced and yet to be produced by the plant and / or past and future states of the plant as a whole. A maintenance planning system and / or the production planning system receive the states of the plant components determined by the condition monitoring system and the states of the input products and / or the plant as a whole determined by the quality assessment system. They take into account the data received from the condition monitoring system and the quality assessment system when determining maintenance planning data and / or production planning data.

[0011] The invention is based on the objective of optimizing the manufacturing process in a production plant consisting of several plant components, in particular a metallurgical production plant for the production of industrial goods such as metallic semi-finished products and / or metallic end products.

[0012] The problem is solved according to the invention by a system according to claim 1 and a method according to claim 9. Advantageous embodiments of the invention are defined in the dependent claims.

[0013] The invention relates to a system for controlling a production plant consisting of several plant components, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products, wherein each plant component has an input quality window, an output quality window, and a process window, wherein the input quality window of a plant component defines the quality characteristics of the input product required by the plant component, and the output quality window of a plant component defines the quality characteristics of the output product permitted by the plant component after processing the input product, wherein, in a production plant consisting of several plant components, the output quality window of an upstream plant component corresponds to the input quality window of the downstream plant component, wherein the process window defines the setpoints that can be implemented by the respective plant component for plant automation of the plant component, wherein each plant component detects the current state by means of sensors and adjusts the process window of the plant component to the detected current state.and wherein the system for controlling the production plant, consisting of several plant sections, determines setpoints for the respective plant automation for each plant section, which lie within the respective process windows, and that the product manufactured in the production plant meets the quality characteristics required by the input quality windows and output quality windows of the several plant sections.

[0014] A plant component within the meaning of the invention can generally also be referred to as a conversion stage.

[0015] According to the invention, the conditions of the components of the production plant are monitored by means of sensors. Based on this monitoring, the process windows of the respective components can be adjusted to the current condition of the component. For example, it is taken into account that the process window narrows over time due to wear and tear. Furthermore, malfunctions or external influences can be detected, and the process window of the affected component can be adjusted accordingly.

[0016] The system according to the invention determines setpoints for the automation systems of the production plant components, with these setpoints lying within the current process windows of the plant components. Thus, the current state of the plant is taken into account when determining the setpoints. The setpoints are selected such that the product manufactured in the production plant meets the quality characteristics required by the input and output quality windows of the various plant components. This ensures that the manufactured product meets the required quality requirements throughout the entire production process. Simultaneously, this guarantees that the final product meets the quality requirements.

[0017] According to an advantageous embodiment of the invention, after processing of the product by a system component, the system updates the setpoints for the automation of subsequent system components based on the initial quality achieved by the processed component and the current process windows of the subsequent components. Thus, after processing of the product by a system component, the subsequent processing by the subsequent components is updated. The processing already performed by the preceding components remains unchanged, and the processing by the subsequent components is determined based on the current process windows of the subsequent components.

[0018] According to one embodiment of the invention, the system determines the sequence of products to be manufactured in the production plant, taking into account, in particular, the current process windows and the achievable quality characteristics of the input products of the plant components. The production of products in the plant is optimized based on the current state of the plant components and the current process windows for several products to be manufactured. For example, products with comparable or similar quality requirements are manufactured sequentially, so that the setpoints for the plant automation of the plant components differ as little as possible. In particular, when determining the sequence of products to be manufactured in the production plant, the system considers the differences and / or similarities of the products to be manufactured.

[0019] In one embodiment of the invention, the system takes into account the possible setpoint change rates achievable by the respective plant automation systems when determining the setpoints for the plant automation systems of the individual plant components. While the plant automation systems of the individual plant components can implement the setpoints within the current process window, this is often only possible within specific time periods. Therefore, no sudden changes in the setpoint value are possible, only changes within a predefined setpoint change rate. These setpoint change rates are taken into account by the system according to the invention when determining the setpoints for the plant automation systems of the respective plant components.

[0020] According to a preferred embodiment of the invention, the system comprises a quality control unit downstream of one or more plant sections for verifying the achieved product quality, in particular for adjusting the setpoints for the plant automation of the subsequent plant sections. In other words, it is checked whether the product manufactured up to that point exhibits the expected quality characteristics at the quality control location. Should the determined quality deviate from the expected quality, the setpoints for the plant automation of the subsequent plant sections can be adjusted so that the product subsequently also meets the quality requirements of the input and output quality windows of the subsequent plant sections.

[0021] According to the invention, the setpoints of the plant automation of a plant section determine the output quality of the product manufactured in that plant section, in particular whether and within which range of the output quality window the quality of the product manufactured in that plant section lies. Thus, the quality of the output product of a plant section is determined via the setpoints, whereby the setpoints must lie within the current process window of the plant section.

[0022] In one embodiment according to the invention, the sensors of the system components detect the wear, maintenance condition or the like of the system component.

[0023] According to a particularly preferred embodiment of the invention, the system creates a predictive model for the future states of the plant components and takes into account the states of the plant components predicted by the predictive model and the resulting process windows of the plant components when determining the setpoints for the plant automation of the plant components. In particular, the predictive model is used to predict the future state of plant components at the time of processing a specific product. If, for example, the production plant consists of five different plant components, the currently recorded state of the first plant component is taken into account, and the future state of each of the subsequent plant components is predicted using the predictive model.Based on these states, the process windows of the plant components are adjusted, and corresponding setpoints within these process windows are subsequently determined. The predictive model improves the accuracy of the production plant's control, particularly with an increasing number of plant components. The predictive model is especially advantageous for determining a production sequence for the plant, as it can also be used to determine the corresponding states of the plant components and the resulting process windows for future products, and these can then be taken into account when planning the production sequence.

[0024] According to one embodiment of the invention, the predictive model is based on the states of the plant components as detected by the sensors, the product qualities achieved by the plant components, other measured values ​​belonging to the production plant, or the like. In particular, the temporal progression of the aforementioned parameters can be monitored, and a prediction of future states of the plant components can be made based on the historical data. The prediction can also be refined by maintenance information, information on products to be manufactured in the plant, or other parameters relating to the production plant.

[0025] In a convenient variant of the invention, the prediction model is based on methods of statistical data evaluation and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, Lagrange multiplier method or the like.

[0026] According to an advantageous embodiment of the invention, the prediction model is trained continuously or cyclically, in particular based on the states of the plant components detected by the sensors, the product qualities achieved by the plant components, other measured values ​​belonging to the production plant, or the like. The prediction model is thus continuously improved, so that the prediction accuracy steadily increases with the operating time of the production plant.

[0027] According to a preferred embodiment of the invention, the system stops the manufacturing process in the production plant if the required quality of the manufactured product cannot be achieved. Alternatively, the system changes the manufacturing process to a different product whose required quality can still be achieved by the current manufacturing process. Thus, if it is determined that the desired product cannot be manufactured due to the current state of the plant, production is either stopped or changed to a different, still manufacturable product. This can also occur during the manufacturing process, for example, after processing by one or more plant components.For example, if, after processing by a third section of the system, it becomes apparent that the desired product can no longer be manufactured due to the current state of the subsequent sections, the manufacturing process is stopped after the third section. If possible, production is then switched to another product that can still be manufactured.

[0028] In a practical approach, the change to a different product is limited to the production of products planned within a specific timeframe. Therefore, current production can only be changed to a product that is to be manufactured in the near future.

[0029] According to an advantageous embodiment of the invention, the system takes into account several identical plant components, so that the processing of a manufacturing step can alternatively be carried out on different plant components. Thus, if one plant component is not expected to achieve the required product quality, the manufacturing step can be carried out on another identical plant component that is expected to provide the required product quality. This results, for example, from different current process windows of the identical plant components.

[0030] According to one suitable variant, the system includes a central data storage or provides cloud storage for storing and providing system-relevant data, in particular the states of the plant components recorded by sensors, the product qualities achieved by the respective plant components, the set values ​​for the plant automation of the respective plant components, and the like.

[0031] In an advantageous embodiment of the invention, the plant components at least partially comprise optical sensors for acquiring geometric information about the products manufactured in the respective plant component. Particularly in metallurgical production plants, the geometric information of the products manufactured in the respective plant component is a crucial factor. Thus, the system according to the invention can, for example, derive the product quality of the product manufactured in the plant component and / or the plant condition of the component from the geometric information, preferably using statistical methods and / or machine learning. Furthermore, the geometric information can be continuously and easily acquired using the optical sensors and subsequently evaluated, which improves the accuracy of the system.

[0032] According to one advantageous variant, the system according to the invention comprises a user interface for displaying information relevant to the manufacturing process, in particular on a portable device.

[0033] According to the invention, the setpoints for the plant automation of the several plant components are determined by means of a model. The model takes into account, in particular, the input quality windows, the output quality windows, and the current process windows of the several plant components. Furthermore, additional parameters can be considered, such as measured values ​​acquired from the plant components and / or from other data sources, such as a production planning system, maintenance planning system, or other components of the production plant. Based on the input data, the model can determine the optimal setpoints for the respective plant components for the production of a product.

[0034] According to one variant of the invention, the model is based on physical laws. In an alternative variant, the model is based on methods of statistical data analysis and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, the Lagrange multiplier method, or the like.

[0035] According to one advantageous variant, the model is trained continuously or cyclically, i.e., constantly improved.

[0036] The invention further relates to a method for controlling a production plant consisting of several plant components, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products, wherein each plant component has an input quality window, an output quality window, and a process window, wherein the input quality window of a plant component defines the quality characteristics of the input product required by the plant component, and the output quality window of a plant component defines the quality characteristics of the output product permitted by the plant component after processing the input product, wherein, in a production plant consisting of several plant components, the output quality window of an upstream plant component corresponds to the input quality window of the downstream plant component, wherein the process window defines the setpoints that can be implemented by the respective plant component for plant automation of the plant component, wherein the method comprises the following steps: recording the current states in the several plant components, in particular by means of sensors,Adapting the respective process windows of the multiple plant components based on the recorded current states of the multiple plant components, and determining respective setpoints for the plant automation of the multiple plant components of the production plant, wherein the determined respective setpoints lie within the adapted respective process windows and wherein the product manufactured in the production plant fulfills the quality characteristics required by the input quality windows and output quality windows of the multiple plant components.

[0037] According to the invention, the current states of the components of the production plant are recorded. This is done primarily using suitable sensors. This allows the states to be recorded easily and, in particular, continuously. Based on the recorded current states of the components, the process windows of the respective components are adjusted to the current state of the component. For example, it is taken into account that the process window decreases over time due to wear and tear. Furthermore, malfunctions or external influences can be detected, and the process window of the affected component can be adjusted accordingly.

[0038] The method according to the invention determines setpoints for the automation systems of the production plant components, wherein the setpoints lie within the current process windows of the plant components. Thus, the current state of the plant is taken into account when determining the setpoints. The setpoints are selected such that the product manufactured in the production plant meets the quality characteristics required by the input and output quality windows of the various plant components. This ensures that the manufactured product meets the required quality requirements throughout the entire production process. This simultaneously guarantees that the final product meets the quality requirements.

[0039] In one embodiment of the invention, the method comprises the step of updating the setpoints for the automation of the subsequent plant components after the product to be manufactured has been processed by a plant component. The update is based on the initial quality achieved by the processed plant component and the current process windows of the subsequent plant components. Thus, after the product to be manufactured has been processed by a plant component, the subsequent processing by the subsequent plant components is updated. The processing already carried out by the preceding plant components remains unchanged, and the processing by the subsequent plant components is determined based on the current process windows of the subsequent plant components.

[0040] According to one embodiment of the invention, the method comprises the step of determining the sequence of products to be manufactured in the production plant, particularly taking into account the current process windows and the achievable quality characteristics of the starting materials of the plant components. The production of products in the plant is optimized based on the current state of the plant components and the current process windows for several products to be manufactured. For example, products with comparable or similar quality requirements are manufactured sequentially, so that the setpoints for the plant automation of the plant components differ as little as possible. In particular, the method according to the invention comprises the step of considering differences and / or similarities of the products to be manufactured when determining the sequence of products to be manufactured in the production plant.

[0041] According to one embodiment of the invention, the method includes the step of considering possible setpoint change rates achievable by the respective plant automation systems when determining the setpoints for the automation systems of the respective plant components. While the automation systems of the individual plant components can implement the setpoints within the current process window, this is often only possible within specific time periods. Therefore, no sudden changes in setpoint values ​​are possible, only changes within a predetermined setpoint change rate. These setpoint change rates are taken into account by the method according to the invention when determining the setpoints for the automation systems of the respective plant components.

[0042] One variant of the method according to the invention comprises the step of verifying the achieved product qualities as part of a quality control check after one or more of the plant sections, in particular to adjust the setpoints for the plant automation systems of the subsequent plant sections. This checks whether the product manufactured up to that point exhibits the expected quality characteristics at the quality control location. Should the determined quality deviate from the expected quality, the setpoints for the plant automation systems of the subsequent plant sections can be adjusted so that the product subsequently also meets the quality requirements of the input and output quality windows of the subsequent plant sections.

[0043] According to the invention, the setpoints of the plant automation of a plant section determine the output quality of the product manufactured in that plant section, in particular whether and within which range of the output quality window the quality of the product manufactured in that plant section lies. Thus, the quality of the output product of a plant section is determined via the setpoints, whereby the setpoints must lie within the current process window of the plant section.

[0044] In a suitable variant of the invention, the wear, maintenance condition or the like of the system components are recorded.

[0045] According to a particularly preferred embodiment of the invention, the method comprises the step of creating a predictive model for the future states of the plant components, wherein the states of the plant components predicted by the predictive model and the resulting process windows of the plant components are taken into account when determining the setpoints for the plant automation of the plant components. The predictive model is used, in particular, to predict the future state of plant components at the time of processing a specific product. For each plant component, the predictive model specifies the expected states of the plant component at the respective processing time, on the basis of which the expected process window at that time is determined. The predictive model improves the accuracy of the control of the production plant, especially with an increasing number of plant components.The prediction model is particularly advantageous when determining a production sequence for the production plant, since the corresponding states of the plant components and the resulting process windows can be determined for future products and taken into account when planning the production sequence.

[0046] According to one embodiment of the invention, the predictive model is based on the states of the plant components as detected by the sensors, the product qualities achieved by the plant components, other measured values ​​belonging to the production plant, or the like. In particular, the temporal progression of the aforementioned parameters can be monitored, and a prediction of future states of the plant components can be made based on the historical data. The prediction can also be refined by maintenance information, information on products to be manufactured in the plant, or other parameters relating to the production plant.

[0047] In a convenient variant of the invention, the prediction model is based on methods of statistical data evaluation and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, Lagrange multiplier method or the like.

[0048] According to an advantageous embodiment of the invention, the prediction model is trained continuously or cyclically, in particular based on the states of the plant components detected by the sensors, the product qualities achieved by the plant components, other measured values ​​belonging to the production plant, or the like. The prediction model is thus continuously improved, so that the prediction accuracy steadily increases with the operating time of the production plant.

[0049] According to a preferred embodiment, the inventive method comprises the step of stopping the manufacturing process in the production plant if the required quality of the manufactured product cannot be achieved, or of changing the manufacturing process to another product whose required quality can still be achieved by the current manufacturing process. Thus, if it is determined that the desired product cannot be manufactured due to the current state of the plant, production is either stopped or changed to another product that can still be manufactured. This can also occur during the manufacturing process, for example, after processing by one or more plant components.

[0050] In a practical approach, the change to a different product is limited to the production of products planned within a specific timeframe. Therefore, current production can only be changed to a product that is to be manufactured in the near future.

[0051] According to one embodiment of the invention, the method includes the step of considering several identical plant components, so that the processing of a manufacturing step can alternatively be carried out on different plant components. Thus, if one plant component is not expected to achieve the required product quality, the manufacturing step is carried out on another identical plant component that is expected to provide the required product quality.

[0052] According to an advantageous variant of the invention, the method comprises storing data in a central data storage system or a cloud storage system, preferably for storing and providing relevant data, in particular the states of the plant components detected by means of sensors, the product qualities achieved by the respective plant components, the set values ​​for the plant automation of the respective plant components and the like.

[0053] In an advantageous embodiment, the method according to the invention comprises the step of acquiring geometric information about the products manufactured in the respective plant section, in particular by means of optical sensors in the respective plant sections. In metallurgical production plants, the geometric information of the products manufactured in the respective plant section is a crucial factor. In particular, the method according to the invention can derive the product quality of the product manufactured in the plant section and / or the plant section's condition from the geometric information, preferably by means of statistical methods and / or machine learning. Furthermore, the geometric information can be continuously acquired and subsequently evaluated in a simple manner using the optical sensors, which improves the system's accuracy.

[0054] According to one suitable variant, the method includes displaying information relevant to the manufacturing process on a user interface, in particular on a portable device.

[0055] According to the invention, the method comprises the step of creating a model for determining the setpoints for the plant automation systems of the various plant components. The model takes into account, in particular, the input quality windows, the output quality windows, and the current process windows of the various plant components. Furthermore, additional parameters can be considered, such as measured values ​​acquired from the plant components and / or from other data sources, such as a production planning system, maintenance planning system, or other components of the production plant. Based on the input data, the model can determine the optimal setpoints for the respective plant components for the production of a product.

[0056] According to one variant of the invention, the model is based on physical laws. In an alternative variant, the model is based on methods of statistical data analysis and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, the Lagrange multiplier method, or the like.

[0057] According to an advantageous embodiment, the method according to the invention comprises the step of continuously or cyclically training the model to determine the set values.

[0058] The problem is further solved by a computer program comprising instructions which, when the program is executed by a computer, cause it to execute the method according to the invention, in particular that the system according to the invention executes the method according to the invention.

[0059] The invention will now be explained in more detail with reference to exemplary embodiments shown in the figures. The figures show: Fig. 1 shows a schematic view of a first embodiment of a system according to the invention for controlling a production plant consisting of several plant components, and Fig. 2 shows a schematic view of a second embodiment of a system according to the invention for controlling a production plant consisting of several plant components.

[0060] Fig. 1 Figure 1 shows a schematic view of a first embodiment of a system 1 according to the invention for controlling a production plant 3 consisting of several plant components 2, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products. The system according to the invention can be arranged inside or outside the production plant 3, wherein the system 1 is configured for communication with the production plant 3.

[0061] Each plant section 2 of the production plant 3 has an input quality window 4, an output quality window 5, and a process window 6. The input quality window 4 of a plant section 2 defines the quality characteristics of the input product required by plant section 2, and the output quality window 5 of a plant section 2 defines the quality characteristics of the output product permitted by plant section 2 after processing the input product. In a production plant 3 consisting of several plant sections 2, the output quality window 5 of an upstream plant section 2 corresponds to the input quality window 4 of the downstream plant section 2, which in Fig. 1 each is represented by a common rectangle covering the two adjacent plant sections 2.

[0062] Process window 6 defines the setpoints 7 that can be implemented by the respective plant section 2 for plant automation of plant section 2. Process window 6 is in Fig. 1 Each is symbolized by a dashed rectangle within plant section 2. Within the respective process window 6 lie the setpoints 7 for the plant automation of plant section 2. The setpoints 7 are in Fig. 1 symbolized by hexagons. The setpoint values ​​7 of the plant automation of a plant section 2 determine the output quality of the product manufactured in plant section 2, in particular whether and within which range of the output quality window 5 the quality of the product manufactured in plant section 2 lies. This is in Fig. 1 Each of these is symbolized by the line from input quality window 4 through the set values ​​7 within process window 6 to output quality window 5. Across all plant sections 2 of production plant 2, this results in a continuous line, symbolizing the manufacturing process within the production plant.

[0063] Each plant component 2 records its current state using suitable sensors 8. Based on the recorded current states of the plant components 2, the process windows 2 of the plant components 2 are adjusted. The process windows 6 can, for example, shift within a specific parameter space and / or change their size. This is in Fig. 1 represented by the different dashed rectangles within the several plant sections 2.

[0064] The system 1 according to the invention for controlling the production plant 3, which consists of several plant sections 2, determines setpoints 7 for the respective plant automation for each plant section 2. The setpoints 7 determined by the system 1 according to the invention lie within the respective process windows 6 and are selected such that the product manufactured in the production plant 3 meets the quality characteristics required by the input quality windows 4 and output quality windows 5 of the several plant sections 2. Since the product manufactured in the production plant 3 lies within the output quality window 5 of the last plant section 2, the quality requirements placed on the manufactured product are met.

[0065] Preferably, after processing of the product to be manufactured by a plant section 2, the system 1 according to the invention updates the setpoints 7 for the plant automation of the subsequent plant sections 2 based on the initial quality achieved by the processed plant section 2 and the current process windows 6 of the subsequent plant sections 2. The setpoints 7 for the upcoming plant sections 2 are thus continuously adjusted, taking into account the processing already carried out by preceding plant sections 2.

[0066] In an advantageous embodiment, system 1 determines a sequence of products to be manufactured in production plant 3, in particular taking into account the current process windows 6 and the achievable quality characteristics of the input products of plant components 2. When determining the sequence of products to be manufactured in production plant 3, system 1 preferably considers the differences and / or similarities of the products to be manufactured.

[0067] When determining the setpoints 7 for the plant automation systems of the respective plant components 2, the system 1 takes into account the possible setpoint change rates that can be implemented by the respective plant automation systems.

[0068] According to a preferred variant, system 1 includes a quality control step after one or more plant sections 2 to verify the achieved product quality, in particular to adjust the setpoints 7 for the plant automation of the subsequent plant sections 2. Thus, at least partially after plant sections 2, it is checked whether the manufactured product lies within the initial quality window 5. Depending on the determined quality, i.e., the position of the manufactured product within the initial quality window 5, the setpoints 7 for the plant automation of the subsequent plant sections 2 are adjusted.

[0069] The sensors 8 of the plant components 2 detect, for example, the wear, maintenance status, or similar characteristics of the plant component 2. In particular, the plant components 2 include at least partially optical sensors 8 for acquiring geometric information about the products manufactured in the respective plant component 2. From this geometric information, the system 1 can derive the product quality of the product manufactured in the plant component 2 and / or the plant condition of the plant component 2, preferably using statistical methods and / or machine learning.

[0070] According to a particularly preferred embodiment of the invention, the system 1 creates a predictive model 9 for the future states of the plant components 2 and takes into account the states of the plant components 2 predicted by the predictive model 9 and the resulting process windows 6 of the plant components 2 when determining the setpoints 7 for the plant automation of the plant components 2. The predictive model 9 is based, for example, on the states of the plant components 2 detected by means of the sensors 8, the product qualities achieved by the plant components 2, other measured values ​​belonging to the production plant 3, or the like.

[0071] Predictive model 9 is based on methods of statistical data evaluation and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, Lagrange multiplier method or the like.

[0072] The prediction model 9 is trained continuously or cyclically, for example, on the basis of the states of the plant components 2 recorded by means of the sensors 8, the product qualities achieved by the plant components 2, other measured values ​​belonging to the production plant 3 or the like.

[0073] In one advantageous variant, System 1 stops the manufacturing process in production plant 3 if the required quality of the manufactured product cannot be achieved. If possible, System 1 can change the manufacturing process to a different product whose required quality can still be achieved by the current manufacturing process. Advantageously, the change to a different product is limited to the production of products planned within a specific timeframe.

[0074] System 1 can also take into account several similar plant components 2, so that the processing of a manufacturing step can alternatively take place on different plant components 2.

[0075] System 1 from Fig. 1 furthermore includes a central data storage 10 or a cloud storage 10, for storing and providing system-relevant data, in particular the states of the plant components 2 recorded by means of sensors 8, the product qualities achieved by the respective plant components 2, the set values ​​7 for the plant automations of the respective plant components 2 and the like.

[0076] To display information relevant to the manufacturing process, system 1 includes a user interface 11. The user interface is provided, for example, on a portable device, so that the information is also available within the production plant, e.g., during an inspection.

[0077] The system 1 according to the invention further comprises a model 12 for determining the setpoints 7 for the plant automation of the several plant sections 2. The model 12 is based on physical laws or on methods of statistical data evaluation and / or machine learning, in particular linear or quadratic programs, genetic optimization, reinforcement learning with Q-tables, neural networks, simulated annealing, Metropolis, swarm algorithms, hill climbing, the Lagrange multiplier method, or the like. Like the predictive model 9, the model 12 can also be trained continuously or cyclically.

[0078] Fig. 2 Figure 1 shows a schematic view of a second embodiment of a system 1 according to the invention for controlling a production plant 3 consisting of several plant components 2, in particular a metallurgical production plant for the production of industrial goods such as semi-finished metallic products and / or finished metallic products, in comparison to a control system according to the prior art.

[0079] The invention differs from the prior art in that it detects the current states in the several plant sections 2, in particular by means of sensors 8, adapts the respective process windows 6 of the several plant sections 2 based on the detected current states of the several plant sections 2, and determines respective setpoints 7 for the plant automations of the several plant sections 2 of the production plant 3, wherein the determined respective setpoints 7 lie within the adapted respective process windows 6 and wherein the product manufactured in the production plant 3 fulfills the quality characteristics required by the input quality windows 4 and output quality windows 5 of the several plant sections 2.

[0080] According to the current state of the art, it is possible that, based on the processing by a preceding plant section 2 and the current process window 6 of the current plant section 2, the setpoints 7 for the plant automation of the current plant section cannot be set in such a way that the manufactured product lies within the output quality window 5 of the current plant section 2. This is in Fig. 2 symbolized by the solid line, which lies outside the output quality window 5 of plant section 2 after the second plant section from the left.

[0081] According to the present invention, the current process windows 6 of all plant components 2 are taken into account when determining the setpoints 7 for the plant automation of the plant components 2. Thus, the process windows of the plant components 2 at the end of the manufacturing process are already considered at the beginning of the manufacturing process. The setpoints 7 are therefore determined so that the product manufactured in the production plant 3 lies within all input quality windows 4 and output quality windows 5 of the plant components 2. This in Fig. 2 represented by the dashed line, which lies within all input quality windows 4 and output quality windows 5. Reference symbol list

[0082] 1 System 2 Plant component 3 Production plant 4 Input quality window 5 Output quality window 6 Process window 7 Setpoints 8 Sensors 9 Prediction model 10 Data storage / Cloud storage 11 User interface 12 Model for predicting setpoints

Claims

1. System (1) for controlling a production plant (3) consisting of a plurality of plant parts (2), particularly a metallurgical production plant for producing industrial goods such as metallic semi-finished products and / or metallic end products, wherein each plant part (2) has an input quality window (4), an output quality window (5) and a process window (6), wherein setting values (7) determine the output quality of the product produced in the plant, wherein the input quality window (4) of a plant part (2) defines the quality characteristics, which are required by the plant part (2), of the input product and the output quality window (5) of a plant part (2) defines the quality characteristics, which that are permitted by the plant part (2) after processing the input product, of the output product, wherein in the case of a production plant (3) consisting of a plurality of plant parts (2) the output quality window (5) of an upstream plant part (2) corresponds with the input quality window (4) of the downstream plant part (2), wherein the process window (6) defines the setting values (7) which can be implemented by the respective plant part (2) for a plant automation unit of the plant part (2), wherein each plant part (2) detects the current state by means of sensors (8) and adapts the process window (6) of the plant part (2) to the detected current state, and wherein the system (1) for controlling the production plant (3) consisting of the plurality of plant parts (2) determines for each plant part (2) setting values (7), which lie within the respective process windows (6), for the respective plant automation unit so that the product produced in the production plant (3) meets the quality characteristics required by the input quality windows (4) and output quality windows (5) of the plurality of plant parts (2), wherein the setting values (7) for the plant automation units of the plurality of plant parts (2) are determined by means of a model (12).

2. System (1) according to claim 1, wherein after the product to be produced has been processed by a plant part (2) the system (1) updates the setting values (7) for the plant automation units of the subsequent plant parts (2) on the basis of the achieved output quality of the processed plant part (2) and the current process windows (6) of the subsequent plant parts (2).

3. System (1) according to claim 1 or claim 2, wherein the system (1) determines a sequence of products to be produced in the production plant (3), particularly by taking into account the current process windows (6) and the achievable quality characteristics of the output products of the plant parts (2).

4. System (1) according to any one of the claims 1 to 3, wherein the system (1) takes into account the possible setting value change rates, which can be realised by the respective plant automation units, when determining the setting values (7) for the plant automation units of the respective plant parts (2).

5. System (1) according to any one of the claims 1 to 4, wherein the system (1) creates a prediction model (9) for the future states of the plant parts (2) and takes into account the states of the plant parts (2) predicted by the prediction model (9) and the process windows (6), which result therefrom, of the plant parts (2) when determining the setting values (7) for the plant automation units of the plant parts (2).

6. System (1) according to claim 5, wherein the prediction model (9) is based on the states of the plant parts (2) detected by means of the sensors (8), the achieved product qualities of the plant parts (2), other measured values belonging to the production plant (3) or the like.

7. System (1) according to any one of the claims 1 to 6, wherein the system (1) takes into account a plurality of plant parts (2) of the same type so that the processing of a production step can alternatively take place on different plant parts (2).

8. System (1) according to any one of the claims 1 to 7, wherein the plant parts (2) comprise at least in part optical sensors (8) for detecting geometric data of the products produced in the respective plant part (2).

9. Method for controlling a production plant (3) consisting of a plurality of plant parts (2), particularly a metallurgical production plant for producing industrial goods such as metallic semi-finished products and / or metallic end products, wherein each plant part (2) has an input quality window (4), an output quality window (5) and a process window (5), wherein the input quality window (4) of a plant part (2) defines the quality characteristics, which are required by the plant part (2), of the input product and the output quality window (5) of a plant part (2) defines the quality characteristics, which are permitted by the plant part (2) after processing the input product, of the output product, wherein in the case of a production plant (3) consisting of a plurality of plant parts (2) the output quality window (5) of an upstream plant part (2) corresponds with the input quality window (4) of the downstream plant part (2), wherein the process window (6) defines the setting values (6), which can be implemented by the respective plant part (2), for a plant automation unit of the plant part (2), wherein the method comprises the following steps: detecting the current states in the plurality of plant parts (2) and adapting the respective process windows (6) of the plurality of plant parts (2) by way of the detected current states of the plurality of plant parts (2) on the basis of the monitoring, creating a model (12) for determining setting values (7) for the plant automation units of the plurality of plant parts (2), wherein the setting values (7) determine the output quality of the product produced in the plant part, and determining the respective setting values (7) for the plant automation units of the plurality of plant parts (2) of the production plant (3), wherein the determined respective setting values (7) lie within the adapted respective process windows (6) so that the product produced in the production plant (3) meets the quality characteristics required by the input quality windows (4) and output quality windows (5) of the plurality of plant parts (2).

10. Method according to claim 9, further comprising the step of updating the setting values (7) for the plant automation units of the downstream plant parts (2) after the product to be produced has been processed by a plant part (2), wherein the updating is carried out on the basis of the achieved output quality of the processed plant part (2) and the current process windows (6) of the downstream plant parts (2).

11. Method according to claim 9 or claim 10, further comprising the step of determining a sequence of products to be produced in the production plant (3), particularly taking into account the current process windows (6) and the achievable quality characteristics of the output products of the plant parts (2).

12. Method according to any one of the claims 9 to 11, comprising the step of taking into account possible setting value change rates, which can be realised by the respective plant automation units, when determining the setting values (7) for the plant automation units of the respective plant parts (2).

13. Method according to any one of the claims 9 to 12, comprising the step of creating a prediction model (9) for the future states of the plant parts (2), wherein the states of the plant parts (2) predicted by the prediction model (9) and the process windows (6), which result therefrom, of the plant parts (2) are taken in account when determining the setting values (7) for the plant automation units of the plant parts (2).

14. Method according to claim 13, wherein the prediction model (9) is based on the states of the plant parts (2) detected by means of the sensors (8), the achieved product qualities of the plant parts (2), other measured values belonging to the production plant or the like.

15. Method according to any one of the claims 9 to 14, comprising the step of taking into account a plurality of plant parts (2) of the same type so that the processing of a production step can alternatively take place on different plant parts (2).

16. Method according to any one of the claims 9 to 15, comprising the step of detecting geometric information of the products produced in the respective plant part (2), particularly by means of optical sensors (8) in the respective plant parts (2).

17. Computer program comprising commands which when the program is executed by a computer cause this to execute the method according to any one of claims 9 to 16, particularly in that the system (1) according to any one of claims 1 to 8 executes the method according to any one of the claims 9 to 16.