EP3962708A1

EP3962708A1 - Method for operating a device, computer program product and device for producing a product

Info

Publication number: EP3962708A1
Application number: EP20723079.8A
Authority: EP
Inventors: Christoph NIGL
Original assignee: Alpla Werke Alwin Lehner GmbH and Co KG
Current assignee: Alpla Werke Alwin Lehner GmbH and Co KG
Priority date: 2019-04-29
Filing date: 2020-04-28
Publication date: 2022-03-09
Also published as: WO2020221766A1; CH716122A1; US20220176604A1; CN113795366A

Abstract

The invention relates to a method for operating a device for producing a product, comprising the following steps: - capturing at least one quality data record (10, 10') comprising measured values of one or more quality parameters, each corresponding to one property of the product; - capturing at least one associated machine data record (20, 20') comprising actual values of a plurality of adjustable machine parameters of the device; - chronologically assigning the quality data record (10, 10') to the machine data record (20, 20') and generating a first data record (30) comprising chronologically correlated measured values and actual values; - repeating the preceding steps at least once to generate at least one second data record (30'); - determining a correlation (40) between the quality parameter(s) and the machine parameters by comparing the captured data records (30, 30') and creating a control model for the device, - providing a corresponding target value for at least one of the adjustable machine parameters on the basis of the control model, proceeding from a target value of the quality parameter(s). The invention also relates to a computer program product for carrying out the method and to a device comprising said computer program product.

Description

Method for operating a device, computer program product and device for manufacturing a product

The present invention relates to a method for operating an apparatus for manufacturing a product, in particular an apparatus for molding a hollow body or a

Injection molded part, a computer program product and a device for producing a product according to the preamble of the independent claims.

Different methods for operating an apparatus for manufacturing a product are known from the prior art. Typically, when operating a device for manufacturing a product, a large number of parameters are set on the device before the actual production process. These parameters can change during the manufacturing process, be it through manual adjustment

ren / readjustment or by the fact that properties of the device itself change in the course of the process. For example, the behavior of a device changes depending on the environment in which it is operated or what material is processed in the device. External parameters such as room temperature or outside temperature or air humidity have an influence on the device. Properties of the device itself can also change. For example, a device can heat up during the production process, which has the consequence, for example, that certain parts of the device change in length or that operating resources, such as hydraulic oil or the like, have a viscosity that changes during operation. All of these changes have an impact on the product to be manufactured. It is therefore necessary to continuously check the properties of the products. In certain processes it is sufficient if, for example, every 100th product is tested, other processes make it necessary that every single product is tested. Such test processes are complex and expensive. In addition, not all products can be tested non-destructively.

A method for blow molding containers is known from WO 2011/023155 A1, in which, based on measured parameters characterizing the blow molding process, at least one property of the fully blown container is calculated and compared with a target value. The parameter influencing the blow molding process is changed based on an established deviation. The simulations in question should be carried out using expert knowledge.

This known method is complex and imprecise. Providing expert knowledge requires a lot of effort and is based on subjective judgments.

The object of the invention is to remedy one or more disadvantages of the prior art. In particular, a method and a device are to be provided which make it possible to operate devices for the production of products essentially automatically and to produce products of as constant a quality as possible, and in particular to reduce the effort required to control the products produced.

This object is achieved by the devices and methods defined in the independent patent claims. Further embodiments are the subject of the dependent claims.

A method according to the invention for operating a device for manufacturing a product, in particular a method for operating a device for forming a hollow body or an injection-molded part, comprises the steps:

- Acquisition of at least one quality data record comprising measured values of one or more quality parameters which each correspond to a property of the product; - Acquisition of at least one associated machine data record comprising real values of several, in particular all, adjustable machine parameters of the device;

- Temporal assignment of the quality data record to the machine data record and generation of a first data record comprising time-correlated measured values and real values;

- repeating the preceding steps at least once to generate at least one second data set;

- Determining a correlation between the quality parameter or parameters and the machine parameters by comparing the captured data sets and creating a control model for the device, and

- Provision of a corresponding setpoint for at least one of the adjustable machine parameters using the control model, based on a setpoint of the quality parameter or parameters.

The quality parameters of the product can be, for example, the material distribution in the finished product, the contour of the finished product, the wall thickness of the finished product, the weight of the finished product, the temperature distribution of the immediately demolded product or the color - and surface quality of the finished product.

The product can also be an intermediate product. Accordingly, it may be in the quality parameters corresponding to a characteristic of the product, also a Para act meter of the intermediate product, such as the tempera ture of an art _S toffschmelze or a diameter of a

Tube preforms in extrusion blow molding.

All of these quality parameters are generally recorded by means of suitable sensors and made available as measured values, in particular automatically. However, it is also possible that quality parameters such as color or surface properties, which can be indicators of the surface quality, for example, are assessed by a machine operator and their size is communicated to the system as measured values through manual inputs.

The machine parameters of the device can be, for example, the temperature of a tool, the tempera ture of a corresponding operating medium such as hydraulic oil or compressed air or cooling water and the like or, in the case of a blow molding machine, for example, the temperature of a mold, the volume of a Blowing gas flow or its temperature, to parameters of the stretching process such as Reckgeschwindig speed, to temperatures of a preform, to correspond to the melt pressure in a corresponding cavity, to speed of a screw conveyor, to volume, type or shape and temperature of starting material such as plastic granulate, which ge promotes, act. The static charge of the machine or the supplied plastic granulate, for example, can also be recorded as machine parameters. It goes without saying that not all of the machine parameters mentioned here can be set directly. For example, the shape of the granulate can only be changed by replacing the granulate that has been fed in. It can thus be provided that in addition to adjustable machine parameters, non-adjustable machine parameters are also recorded.

All of these machine parameters can in particular be measured or read out and made available as real values. Preferably, several machine parameters are recorded simultaneously and made available in a machine data record.

An associated machine data record is in the present case a data record of the device with which the program product from which the quality parameter originates.

By capturing a quality data record, several measured values of a quality parameter of a product are included. Preferably, several measured values from several quality parameters are recorded simultaneously and made available in a quality data record.

By acquiring a machine data record, several real values of a corresponding machine parameter are included. These can map a time course of the corresponding real value.

The machine data set is chronologically assigned to the quality data set, in other words, linked to it, and a data set is generated which contains both measured values and real values. In this case, that quality data record is preferably assigned to each machine data record which was relevant for the manufacture of the product or a preliminary product and thus the quality parameters associated therewith.

In principle, it is not necessary to always record all quality parameters at a certain point in time; For example, the weight of the finished product could always be recorded while, for example, the surface quality is only recorded for every tenth product. By comparing several such data sets, a correlation between the quality parameter or the quality parameters and the machine parameter or the machine parameters can be determined. These correlations can be both causal relationships and indirect correlations that are linked to one another via a third variable, for example.

The provision of a setpoint value of at least one machine parameter based on the determined correlation, based on a Setpoint of a quality parameter or several quality parameters makes it possible to precisely control a device for manufacturing a product. In particular, a control of the product after its manufacture can be superfluous, since the prior identification of the relationships between quality parameters and machine parameters, in other words, the determination of the correlation between the respective machine parameters and quality parameters, and the corresponding setting of the machine parameters the result, namely the target value to be achieved for the quality parameter, is predictable.

A large number of setpoint values of machine parameters are preferably provided on the basis of a setpoint value of a quality parameter, such as a wall thickness of a blown body. All setpoint values of the machine parameters that can be influenced in terms of control and / or regulation technology are preferably provided.

Provision can also be made to determine correlations between the real values of several machine parameters, since it is conceivable that several machine parameters also influence one another.

As a result, it can be predicted to what extent changing one machine parameter will affect another machine parameter, and subsequently also to what extent a quality parameter may change.

In addition, at least one environmental data set can be recorded. The environmental data set comprises measured values of one or, in particular, several environmental parameters or environmental parameters. This environmental data set is temporally assigned to the machine data set and thus forms part of the respective recorded data sets. With the acquisition of an environmental data set, it is also possible to determine the influence of the environmental conditions on the corresponding quality parameters and to include them in the correlation and thus in particular to expand the control model by one or more boundary conditions.

The environmental parameters can include parameters such as air temperature or humidity, but also the air pressure, the time of day (especially day or night), the geographical position of the machine and / or the factory in which the machine is operated (and accordingly the associated climatic conditions) or parameters about the status of the factory hall in which the machine is located, for example open or closed gates or windows, which can say something about air movement, such as drafts.

For example, during a blowing process, the ambient air temperature can have an influence on the cooling rate of the product, which in turn can possibly have an influence on a corresponding wall thickness of the product if, for example, the material solidifies more slowly at certain points.

It goes without saying that several combinations of setpoint values of machine parameters that differ from one another can be suitable for achieving a specific setpoint value of a quality parameter.

Depending on the boundary conditions, such as the ambient temperature, one or more machine parameters must be set differently. These relationships are recorded by determining the correlations between the measured values and real values and, in the case of an environmental parameter as a boundary condition, the measured values of the environmental parameters. These relationships are then made available in a control model. The data records can be determined at least once on the basis of test results from the device for manufacturing a product and can be made available to create the control model for the device.

This enables data sets to be made available in a specific form. So it is conceivable that individual machine parameters are kept constant over a certain period of time and other machine parameters are varied over this period of time. One or more quality parameters on the product can be measured over the length of this period, and accordingly the measured values of the quality parameters and the corresponding real values of the machine parameters can be transferred to corresponding quality data records or machine data records.

Additionally or alternatively, provision can be made to collect a large number of measured values and real values from production plants with devices for manufacturing products and to make them available in a database, for example. A large number of quality data sets and machine data sets can be created from these measured values and real values, and a large number of data sets can be generated from these. By comparing this multitude of data sets, the determination of correlations is simplified and the corresponding correlations are more precise. This enables the creation of a more precise control model.

The data records can therefore be determined by collecting a large number of measured values and real values from production systems and made available for creating the control model for the device.

The process steps, as described here, essentially correspond to a learning phase. This learning phase creates a very close-knit link between the machine parameters, quality parameters and environmental parameters, which in turn leads to a very precise model of the relationships, which in turn leads to a very precise control model.

The provision of a setpoint for at least one machine parameter described in this method can take place following this learning phase.

The provision of the aforementioned setpoint or several setpoints for machine parameters can take place in the form of a transfer of these values to a control of a corresponding device for manufacturing a product, for example automatically or manually. The controller can be part of a computer and comprise one or more computer program products. Following the handover, the device for manufacturing a product can be operated in accordance with these specifications and carry out a corresponding production process or manufacturing process.

Provision can be made for actual values of several of the machine parameters and at least one of the environmental parameters to be recorded and, based on these actual values, to output the value or values of the quality parameter or quality parameters corresponding to these actual values using the control model. This value or these values are compared with the setpoint or values of the quality parameter or parameters and a deviation is determined. The device is then controlled, in particular regulated, on the basis of this deviation with the aid of the control model, in particular taking into account the actual value of the at least one environmental parameter.

Alternatively or additionally, it can be provided that the actual values of the machine parameters are compared with the respective setpoint. The device can be based on a deviation of the Actual value of the machine parameter controlled by the corresponding setpoint, in particular regulated.

It can be provided that only the actual values of the machine parameters and / or the environmental parameters are recorded during the control of the device. This means that the quality parameters are not recorded. These are automatically correct when the actual values of the machine parameters correspond to the set values of the machine parameters.

It can be provided that a respective value of a machine parameter, such as the speed at which a mold is opened or closed, is directly influenced. Other machine parameters cannot be directly regulated. This applies, for example, to the viscosity of an operating medium. By means of suitable cooling devices, however, the temperature can be influenced, for example, which in turn has an effect on the viscosity. This effect in turn can be derived from one of the previously determined correlations.

In addition, there may be parameters that cannot be influenced in practice, such as environmental parameters, which naturally can hardly be changed. In order to compensate for their deviations from a desired value, correspondingly correlating machine parameters can be controlled or regulated.

A deviation of the actual value from the corresponding target value of the respective machine parameter means that the corresponding target value of the quality parameter, on the basis of which the target value of the machine parameter was determined, also deviates. A control or regulation of the corresponding machine parameter or possibly one or more machine parameters that have a corresponding influence on these machine parameters and / or correlate with them thus makes it possible to achieve the corresponding setpoint value of the quality parameter. The respective actual value of the machine parameters and the at least one environmental parameter can be recorded continuously during operation of the device.

This enables continuous monitoring of the quality parameter.

Alternatively, it can be provided that the actual values of the respective machine parameters and of the at least one environmental parameter during operation of the device are recorded (cyclically) at predefinable time intervals.

This can be of particular advantage in the case of parameters that change only slowly or that have only a very slight influence on the quality parameters. In this way, the amount of data and the effort required to record the respective machine parameters can be reduced.

In a preferred form of the method, the correlations, that is to say the control model, are transferred to the device only once. Due to the very close-knit linking of machine parameters, quality parameters and environmental parameters, a statement about the quality parameter can be made with fixed machine parameters or fixed machine parameters without having to re-measure them. In other words, the device can be set to the quality parameters to be achieved and the device selects the respective machine parameters in accordance with the control model. As soon as the measured machine parameters, that is to say the actual values, agree with the selected machine parameters, ie the setpoint values, it can be assumed that the actual value of the quality parameter corresponds to the selected setpoint value of the quality parameter.

The present invention thus also relates to the operation of a device, in particular a method for operating a pre direction, for manufacturing a product, in particular a device for molding a hollow body or an injection molded part, the correlations obtained in a learning phase with the method described here being stored statically, in particular as a static control model, in this device. Starting from a desired target value of a quality parameter, the corresponding target values of the machine parameters are selected and set.

However, it can also be provided that the target parameter is entered in an external database and the target parameters are fed into the device from the external database. This can for example be done manually by means of input by a machine operator. An automatic transfer, for example via an electronic interface, is also possible.

It can be provided that an actual value of at least one of the quality parameters is compared cyclically with the target value of the quality parameter.

These comparisons are particularly advantageous when the device for manufacturing a product is operated in test mode, in other words during the learning phase. These comparisons can be used to determine whether the correlations that have been determined agree with practice and / or in order to obtain further measured values.

The control model can preferably be determined in accordance with the correlation or correlations between the quality parameter or parameters and the machine parameters, taking into account the at least one quality parameter by means of machine learning. In particular, the correlation or the correlations can be determined by means of an implementation of artificial intelligence, in particular with a neural network. This allows the establishment and recognition of connections, i.e. correlations, regardless of whether there is a direct or causal connection between individual measured values and real values. Through the use of machine learning, far-reaching and / or superordinate patterns of a large number of individual measured values and real values can be recognized and / or compared with one another and / or linked with one another.

The machine data record can include several machine parameters, in particular real values of several machine parameters. Each machine parameter can be assigned a weighting of its influence on each quality parameter according to its correlation with the one or more quality parameters of the quality data set. The control model therefore has a weighting of the individual machine parameters.

This means that the greater or better the correlation between a machine parameter and a quality parameter, the greater the influence of a change in a corresponding machine parameter on the respective quality parameter. Accordingly, provision can be made to assign a correspondingly high weighting to such a machine parameter, in other words the corresponding correlation between the respective quality parameter and the respective machine parameter.

However, if this machine parameter has an influence, for example on a further quality parameter of the product, which can also be negative, for example, then it can be provided to assign a somewhat lower weighting to this machine parameter. This means that the machine parameter with the greatest influence on a respective quality parameter is not necessarily the machine parameter with the highest weighting. The highest weighting can, for example, have a machine parameter which, although it has a correspondingly large influence on a certain quality parameter of the product, does at the same time has very little influence on other quality parameters and / or on other machine parameters.

Provision can also be made to determine several weightings, for example in relation to a rapid change and in relation to the least possible influence on further machine or quality parameters.

To achieve the target value of one or more of the quality parameters, one or more of the respective machine parameters can be controlled or regulated with the respective weighting in relation to the quality parameter or parameters in accordance with the control model.

Correspondingly, precise intervention and precise control or regulation are made possible.

In addition, it can be provided that a check is carried out to determine whether one or more of the actual values of the machine or environmental parameters recorded during the control of the device lie within a value range of the machine parameters recorded when the control model was created and / or the recorded environmental parameters.

If, for example, the ambient temperature was recorded as one of the environmental parameters to create the control model, there is typically a highest recorded value and a lowest recorded value. The lowest value indicates the lower limit of the value range, the highest value the upper limit. The value range in which the environmental parameter was recorded can thus extend, for example, from 15 ° to 30 °. For actual values outside this value range, there are no longer any values in the control model that are based on measured data.

It can be provided that in the event that one or more of the actual values of the machine or environmental parameters recorded during the control of the device are outside the value range It is checked whether the deviation of the respective actual values occurs once or repeatedly, in particular in a certain period of time, and that an error signal can be output if the respective deviation occurs repeatedly.

This enables the machine operator or a higher-level controller to be informed that the process parameters are outside a range that has been secured by measurements. However, the error message can also be interpreted as an indication that a measuring sensor or measuring probe is defective.

Additionally or alternatively, in the event that one or more of the actual values of the machine or environmental parameters detected during the control of the device are outside the value range, a check can be made to determine whether the deviation from the value range is significant. If the deviation is significant, an error signal can be output.

This prevents a correspondingly unnecessary error signal from being output even if there are deviations from the value range that have little or no influence on the quality parameter.

A significant deviation can exist if either the deviating actual value relates to a machine parameter which is regulated with a high weighting on the setpoint of one or more quality parameters, or if the deviation has exceeded an, in particular adjustable, threshold value.

It can be provided that the control model switches off the device or at least requests manual control interventions if the test has shown that the deviation is significant or occurs repeatedly.

This prevents excessive production of products with deviating quality parameters and / or at least requires a corresponding test and instruction from an operator. Another aspect of the invention relates to a computer program product. The computer program product comprises commands which, when executed on a computer, cause the computer to carry out the steps of the method described here.

This enables the method described here to be implemented in a control and / or regulation of a device for manufacturing a product.

Another aspect of the invention relates to an apparatus for manufacturing a product. This device is in particular a device for molding a hollow body. The device comprises a computer program product as described here.

This makes it possible to provide a device for manufacturing a product, on which both a learning process / a learning phase and a production process / a production phase can be carried out. In other words, a device designed according to the invention makes it possible to determine all correlations between the device itself and the product to be manufactured, and subsequently makes it possible to dispense with extensive testing of the respective manufactured products in the later production process.

The device according to the invention can be part of a device for injection molding or for blow molding. For example, the device is an extruder, a dryer for starting material for the production of the product, an extrusion blower or a stretch blower.

The invention is explained in more detail below with the aid of schematic figures, which only represent exemplary embodiments. Show it:

Figure 1: a schematic representation of an apparatus for manufacturing a product FIG. 2: a sequence _S chema of a method according to the invention

FIG. 1 shows a simplified schematic representation of a device for manufacturing a product, which is embodied here as a blow molding machine 100, for example. The blow molding machine 100 has a feed 110, a Blasformbe rich 120 and a discharge 130. By means of the feeder 120, preforms are fed to the blow molding machine in a known manner. In the blow molding area 120, these are blown with compressed air and stretched with the aid of a stretching rod. The completely inflated containers are collected and / or removed in the discharge 130. This manufacturing process as such is known and is therefore not explained in more detail here.

FIG. 1 also shows a controller 200. The controller 200 is connected to sensors which are arranged on the blow molding machine 100 via connections 102 indicated here by dashed lines. The controller 200 can be designed as a separate unit, but will generally be an integral part of the device. The sensors can be temperature sensors, clocks, position sensors and the like. Real values of machine parameters can be recorded with the sensors. The sensors are only indicated schematically here. A temperature sensor 101 is shown as a placeholder for a large number of sensors.

The connections 102 between the controller 200 and the blow molding machine are provided in the illustrated embodiment wired bound via cables. These connections can, however, also be implemented wirelessly or via optical waveguides.

In addition, a human machine interface (HMI unit), that is to say an operating unit 103, is shown in FIG. A machine operator can use this operating unit 103 to monitor the device and, for example, to specify a target value for a quality parameter. FIG. 2 shows a schematic sequence of a method for operating a device for manufacturing a product, for example a method for operating the blow molding machine 100 from FIG. 1.

In a first step, a quality data record 10 is recorded. The quality data record 10 includes, for example, several measured values of the wall thickness of a container.

In a second step, a machine data record 20 is recorded. The machine data record 20 includes real values of the temperature sensor 101 (see FIG. 1).

The quality data record 10 and the machine data record 20 have been recorded simultaneously. Correspondingly, the quality data set 10 can be assigned to the machine data set 20 in terms of time and a data set 30 can be formed. Within the data set 30, the measured values and the real values are correlated over time.

This time correlation is explained in more detail using the following simplified example. At time tl, for example, the wall thickness Dl of a first container B1 in discharge 130 (see FIG. 1) is measured. The wall thickness Dl corresponds to a quality parameter. At the same time, the temperature Kl of the cavity in which a second container B2 is inflated at the same point in time t1 is measured in the blow molding area 120 (see FIG. 1). The temperature Kl corresponds to a machine parameter. Again at the same time, the temperature VI of a preform of a third container B3 is measured in the feed 110 (see FIG. 1). This temperature VI can be treated as a machine parameter or as a quality parameter. In the present example, the temperature VI is treated as a machine parameter.

In a next step, all containers B1, B2 and B3 are moved on a station. That is, the preform of the third Container B3 is moved from the feed 110 into the blow molding area 120 and the inflated second container B2 is moved from the blow molding area 120 into the discharge area 130, while the first container B1 is removed from the discharge area 130. A new preform of a fourth container B4 is provided in the feed 110.

The wall thickness D2 of the container B2 is now measured at time t2. The preform of the container B3 is located in the cavity for inflating the container B3, the temperature K2 being measured. At the same time, the temperature V2 of the new preform of a container B4 is measured again. These processes are now repeated for the other containers B2, B3 to Bn.

This means that the temperature Kl of the container B2 is measured at the time point t1, but the effect of this temperature Kl on the wall thickness D2 of the container B2 is not measured until the time t2. In other words, for a measured value of a quality parameter, in this example D2, there are a large number of real values which are temporally at different distances before the point in time of the acquisition of the measured value of the quality parameter. These temporal correlations are created in the data set 30.

Following the creation of the data record 30, a second data record 30 'is created. This includes the acquisition of a second machine data record 20 'and the acquisition of a second quality data record 10' and accordingly the generation of a data record 30 'as described above.

The quality data records 10 and 10 'can comprise further quality parameters, which are preferably all recorded simultaneously. Thus, in addition to the wall thickness, another quality parameter can be, for example, the opacity of a wall of the container or the concentricity of a closure with respect to a container base. The machine data records 20 and 20 'can also have measured values of further machine parameters.

In the next step, the data sets 30 and 30 'are brought together and a correlation 40 is determined between the measured values and the real values.

On the basis of these correlations, a setpoint 50 of the recorded machine parameters can be determined for each setpoint value of a quality parameter, or in the case of several machine parameters in the respective machine data sets 20, 20 ', setpoint values 50 for each detected machine parameter. As already stated, there can be several combinations of setpoint values 50 of the machine parameters which lead to the same setpoint value of the quality parameter. In the event that, for example, a machine parameter cannot be changed (or an unchangeable environmental parameter is included in the process), the target values must be determined using two fixed values (target value of the quality parameter and the unchangeable value of the device / environment), which possible combinations. These relationships are stored in a control model for the device, or a control model for the device is created on the basis of these relationships.

Since it is expedient to generate a large number of data records 30, 30 ′, represented as data records 30 ″, 40 machine learning is used to determine the correlations. This makes it possible to compare a large number of values with one another and to determine similarities or patterns and the like and to link them to one another, even if the relationships are no longer obvious.

In a subsequent step of the method, a setpoint value of the quality parameter is selected and corresponding setpoint values 50 of machine parameters are transferred to the controller 200 (see FIG. 1) in order to make the blow molding machine 100 (see FIG. 1) operate accordingly. The control model can be transferred to the control, or just the corresponding values of the machine parameters that were determined using the control model. As soon as the measured real values of the blow molding machine 100 (see FIG. 1) deviate from the setpoint values 50, the control can readjust them. If, for example, values should change that the controller cannot influence (for example outside temperature), the setpoint values 50 can be adjusted according to a specification for the corresponding value that cannot be influenced. For this purpose, the setpoint values 50 are anchored in a data matrix, in particular in a control model, which has been transferred to the controller 200 (see FIG. 1) as part of the setpoint value 50 of the quality parameter.

Claims

1. A method for operating a device for manufacturing a product, in particular a device for molding a hollow body or an injection-molded part, comprising the steps:

- Acquisition of at least one quality data record (10, 10 ') comprising measured values of one or more quality parameters which each correspond to a property of the product;

- Acquisition of at least one associated machine data record (20, 20 ') comprising real values of several, in particular of all, adjustable machine parameters of the device;

- Temporal assignment of the quality data record (10, 10 ') to the machine data record (20, 20') and generation of a first data record (30) comprising time-correlated measured values and real values;

- repeating the preceding steps at least once to generate at least one second data set (30 ');

- Determining a correlation (40) between the quality parameter or parameters and the machine parameters by comparing the recorded data sets (30, 30 ') and creating a control model for the device

- Provision of a corresponding setpoint (50) for at least one of the adjustable machine parameters using the control model, starting from a setpoint of the or the quality parameter.

2. The method according to claim 1, characterized in that additionally at least one environmental data set comprising measured values of one or in particular several environmental parameters is recorded and this is temporally assigned to the machine data record (10, 10 ') so as to be a component of the respective recorded Form data records (30, 30 ').

3. The method according to claim 1 or 2, characterized in that the data sets (30, 30 ') are determined at least once on the basis of test results of the device for manufacturing a product and are provided for creating the control model for the device.

4. The method according to any one of claims 1 to 3, characterized in that the data sets (30, 30 ') are determined by collecting a large number of measured values and real values from production systems and are provided for creating the control model for the device.

5. The method according to any one of claims 1 to 4, characterized in that actual values of several of the machine parameters and at least one of the environmental parameters are recorded and, based on these actual values using the control model, the value or values of the quality parameter (s) corresponding to these actual values is output and this The value or these values are compared with the setpoint value (s) of the quality parameter (s) and a deviation is determined and the device is controlled, in particular regulated, based on this deviation using the control model, in particular taking into account the actual value of the at least one environmental parameter .

6. The method according to claim 5, characterized in that the actual values of the machine parameters are compared with the respective corresponding target value (50) and the device is controlled, in particular regulated, based on a deviation of the actual value of the machine parameters from the target value (50).

7. The method according to claim 5 or 6, characterized in that the actual values of the machine parameter and the at least one environmental parameter are continuously recorded during the operation of the device.

8. The method according to claim 5 or 6, characterized in that the actual values of the machine parameter and of the at least one environmental parameter are recorded cyclically during operation of the device.

9. The method according to any one of claims 1 to 8, characterized in that the control model according to the correlation between the quality parameter or parameters and the machine parameters taking into account the at least one quality parameter by means of machine learning, in particular by means of an implementation of artificial intelligence, in particular with a neural network.

10. The method according to any one of claims 1 to 9, characterized in that the machine data set (20, 20 ') comprises several machine parameters and each machine parameter according to its correlation (40) with the one or more quality parameters of the quality data set a weighting of its Influence on each quality parameter is assigned.

11. The method according to claim 10, characterized in that in order to achieve the target value of one or more of the quality parameters, one or more of the respective machine parameters are regulated with the respective weighting in relation to the quality parameter or parameters according to the control model.

12. The method according to any one of claims 5 to 11, characterized in that it is additionally checked whether one or more of the actual values of the machine or environmental parameters recorded during the control are within a value range of the machine parameters and recorded during the creation of the control model / or the detected environmental parameters lie.

13. The method according to claim 12, characterized in that in the event that the actual values are outside the value range, a check is carried out to determine whether the deviation of the actual values occurs repeatedly and that an error signal is output if necessary.

14. The method according to claim 12, characterized in that in the event that the actual values are outside the value range, a check is carried out to determine whether the deviation from the value range is significant for the control model and that an error signal is output if necessary.

15. The method according to claim 14, characterized in that there is a significant deviation if either the deviating actual value affects a machine parameter that is regulated with a high weighting on the setpoint of one or more quality parameters, or if the deviation is a particular adjustable

Exceeded the threshold.

16. The method according to any one of claims 12 to 15, characterized in that the control model switches off the device or at least requests manual control interventions if the test has shown that the deviation is significant or occurs repeatedly.

17. Computer program product comprising instructions which, when executed on a computer, cause the computer to carry out the steps of the method according to one of claims 1 to 16.

18. Device for producing a product, in particular a device for forming a hollow body, comprising a computer program product according to claim 17.

19. The device according to claim 18, characterized in that the device is part of a device for Injection molding or for blow molding, in particular an extruder, a dryer for starting material for the production of the product, an extrusion blow molding device or a stretch blow molding device.