DE102022127260A1 - Process key figure determination - Google Patents

Abstract

In a method for determining process parameters on a cyclically operating machine for processing plasticizable materials to produce molded parts, at least partial curve profiles (10) of measured variables are determined and stored in an iterative process in process cycles of a reference determination cycle, and the curve profiles are preferably processed in at least one stage using mathematical and/or statistical methods. From this, first process parameters (30) at characteristic points of the processed curve profiles (25) and further process parameters (40) found therefrom by indexing (80) as well as deviations from process parameter reference values are determined and stored. By applying deviation-dependent stored values in an automatic compensation intervention to predetermined manipulated variables under changing production conditions, the process parameters (30, 40) are tracked in the production cycle.

Description

Field of the invention

The present invention relates to a method for determining process parameters on a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts according to the preamble of claim 1, a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts for carrying out the method according to the preamble of claim 18 and a computer program product with a program code stored on a computer-readable medium for carrying out the method according to the preamble of claim 19.

The term "plasticizable material" as used here is to be understood broadly and includes in particular, but not exclusively, plastics or silicone or other thermoplastic and/or elastomeric materials, e.g. ceramic, metallic and/or powdery masses as well as paper, cellulose, starch, cork, etc. as well as mixed materials between such plasticizable materials. In principle, these can also be previously plasticized materials or plastic masses that harden automatically or with the use of aids after application. The term also includes recyclates.

The term "process key figure" as used here refers to a quantity obtained from data using mathematical methods that is suitable for quantitatively evaluating a process state (at least partially, if necessary in combination with other process key figures) and thus also making it accessible for quantitative comparisons and, if necessary, compensatory interventions. In the following, the term process key figure is also used if this quantity does not originate from the process itself in the form of a sensor signal, but can be taken from process-relevant data sources such as material data, which can be made accessible for evaluation according to the claimed method.

When the term "curve progression" is used, it does not only have to refer to measurement curve progressions from the process, but can also refer to other curve progressions whose content can be deduced from the claimed method. "Curve progressions" preferably also include partial sections or even just individual measurement points. Where it is helpful for better understanding, the terms are differentiated.

When the term "derived" curves is used, this means any conceivable processing using mathematical methods to arrive at further curves that can be generated from the basic curve. In individual cases, this can also be a mathematical derivation. Where the mathematical derivation is explicitly meant, this is indicated.

State of the art

Input variables for calculating process parameters such as pressure, position, speed, temperature, torque, etc. are known, as are the variables of additional, non-machine-related sensors. All measurement signals that the production plant, e.g. an injection molding machine, needs to function as intended can be used for this purpose without any additional effort. The use of other signals of all conceivable physical variables that can be used explicitly for monitoring purposes in any part of the production plant is also known.

It is also known to determine process indicators such as peak values of signal curves, area integrals under signal curves, etc. as singular cyclic values. These indicators are often used to measure the stability of the process; if necessary, exceedances of set tolerances are used to select finished parts that were supposedly produced under unstable process conditions. If the operator changes the machine setting parameters, suitable tolerances must be regularly re-determined and set. This can significantly increase the complexity of a machine setting change.

Methods for automatically readjusting machine setting parameters of the process with the aim of compensating deviations are also known. The usual setting parameters of the production plant are used for this purpose, such as setpoints for speeds, pressures, temperatures, and other switching points between process phases.

In the WO91/14562 A1 discloses such a method which carries out the following steps: entering a reference characteristic of the material used for the article(s) to be molded; comparing the reference characteristic with a measured actual characteristic from an immediately preceding shot to obtain a setting value; setting the injection molding machine according to the setting value; and executing a shot. Complete pro process signal curves are compared with each other and an offset is determined from this.

The EP 2 583 811 B1 describes a procedure that follows the same approach as the WO91/14562 A1 , whereby transformation methods are used to determine the deviations between the reference and the current process curve. The quantities that can be measured or set on the production plant are used as input and output parameters. Here, too, complete process curves are compared with one another and an offset and/or scaling factor is determined from this.

The main disadvantage of both of the above methods is the need to record complete curve profiles before they can be processed to determine a correction factor.

In the WO96/09926 A1 A method for automatically influencing machine setting variables in cyclical processes is disclosed, in which, after a test determination of characteristic properties of the parts manufactured in the test phase and simultaneous determination of a large number of process curve profiles, a relationship is created between the process key figures characterizing the process curve profiles and the characteristic properties of the parts. Corresponding process key figures are also determined during production. This signal analysis creates the prerequisite for the fact that, when a certain desired characteristic property is specified, the relevant process key figures can be determined in order to determine the machine setting variables that are relevant for the characteristic property. During production, the expected characteristic property can then be calculated so that a difference value can be determined on the basis of a target-actual value comparison, which can be regulated to a minimum by influencing the relevant machine setting variable that is also calculated. This has the disadvantage that a great deal of effort is required to carry out the test and to evaluate and model each individual process.

From the WO97/12740 A2 A method is known for controlling the cavity pressure curve on a cyclically operating machine, in which the cavity pressure is recorded over time and/or over the distance covered by the conveyor during the cycle. The cavity pressure curve is then differentiated according to time or distance. It is then checked whether deviations from the monotonic curve result from the previously determined derivatives. If this is the case, the holding pressure time or the switching time are changed until a monotonic curve of the cavity pressure curve is obtained. The method selectively examines the cavity pressure curve and is therefore limited in its applicability to the specific application, namely to eliminate discontinuities in this signal curve.

The TW201706112 A discloses a method for providing parameters of an injection device. The injection device comprises a conveyor screw having a screw diameter. The method comprises the following steps: obtaining a total injection volume based on a shape model; using shape analysis software to draw a preset shape curve of the screw feed rate versus the mold filling rate, the preset shape curve containing a plurality of fixed points and each of the fixed points containing a mold filling rate and a screw feed rate; using the mold filling rate of each of the fixed points, the screw diameter and the total injection volume to determine a corresponding stroke position; and providing each of the stroke positions corresponding to the fixed points to the injection device. A disadvantage of this method is the necessary recourse to simulation data.

In the EN 10 2018 107 233 A1 describes a method for automatic process monitoring and/or process diagnosis of a piece-based process, in particular an injection molding process. The method comprises the steps of: carrying out an automated reference finding to obtain reference values from values of at least one process variable; carrying out an anomaly detection on the basis of the reference values found in the first step; carrying out an automated root cause analysis and/or an automated error diagnosis on the basis of a qualitative model of process relationships and/or on the basis of dependencies between different process variables. The anomaly detection as the basis of the method limits it to deviations that have deviated statistically significantly (i.e. by a multiple of the natural process variation) from the stable process. It can therefore only react to major process disturbances.

From the EN 10 2020 125 655 A1 A method for determining a setpoint curve for at least one process parameter acting as a control variable of a production cycle in an injection molding machine in a first configuration is known. In the method, a setpoint curve for the at least one process parameter acting as a control variable is determined in at least one production cycle. meter so that an actual value curve for the at least one selected variable has a desired property or the desired actual value curve arises automatically. In a second embodiment of the method, the actual value curve present in relation to the first embodiment of the method is specified as a reference value curve for the at least one selected variable and a setpoint curve for the at least one process parameter acting as a manipulated variable is determined by a computer. This is done in such a way that the reference value curve arises as an actual value curve with the desired at least one property or the desired actual value curve arises automatically when the injection molding machine in the second configuration operates in a production cycle according to the selected setpoint curve for the at least one process parameter acting as a manipulated variable. The method aims to adapt a complete curve to a reference curve and therefore has the disadvantages of these methods mentioned above. It cannot be assumed that the reproduction of a desired curve under changed boundary conditions (e.g. changed material properties) keeps the quality of the molded parts produced constant.

The EN 10 2021 109 851 A1 discloses a method for comparing a simulation of a process to be carried out with an injection molding machine with the process actually carried out, wherein as part of the simulation at least one simulation curve of a variable characteristic of the process, in particular a simulated pressure curve, is calculated. In the actual process carried out, at least one measurement curve of the characteristic variable, in particular a measured pressure curve, is measured. Furthermore, first marked points of the curve of the at least one simulation curve and second marked points of the curve of the at least one measurement curve are determined. The first marked points and the second marked points are at least partially assigned to one another, wherein at least one change parameter for the simulation and/or the process is calculated from coordinates of the at least partially assigned first marked points and second marked points, and the simulation and/or the process is changed and carried out again on the basis of the at least one change parameter. The disadvantage is the need to use simulation data for this purpose.

• - erfordern einen hohen Versuchs- und Modellierungsaufwand,
• - sind auf den Rückgriff auf Simulationsdaten angewiesen,
• - zielen nicht auf die Konstanz der produzierten Formteilqualität,
• - sind bereits vom Ansatz her in ihrer Wirkung eng begrenzt,
• - erfordern die Erfassung kompletter Kurvenverläufe vor der Auswertung.

The above documents have one or more of the following problems. The procedures

• - require a high level of testing and modelling effort,
• - rely on simulation data,
• - do not aim at the consistency of the produced molded part quality,
• - are already very limited in their effect,
• - require the recording of complete curves before evaluation.

1. a) Die berechneten Kennzahlen
   • - beruhen jeweils auf nur einem einzigen Prozesskurvenverlauf,
   • - verschieben sich bei bewussten Maschineneinstelländerungen des Bedieners im Prozessverlauf,
   • - ändern ihre Größenordnung, sodass sie ihre Aussagekraft verlieren,
   • - sind nicht mehr auffindbar, weil der Prozesssignalverlauf seine Charakteristik infolge von Maschineneinstelländerungen oder Prozessabweichungen stark ändert,
   • - sind voneinander abhängig, auch wenn sie aus unterschiedlichen physikalischen Größen ermittelt werden..
2. b) Die Eingangssignale sind im Bereich der interessierenden Prozessabweichungen bereits im Bereich von Auflösungsgrenzen der Sensorik, so dass stochastische Abweichungen in den Signalverläufen auftreten.
3. c) Die Signalverläufe sind bedingt durch die Diskretisierung lückenhaft.
4. d) Die Signalverläufe sind von starkem Rauschen beaufschlagt.
5. e) Die Signalverläufe, auch unterschiedlicher physikalischer Größen, sind voneinander abhängig.

Further distinguishing features of the present procedure from the above-mentioned procedures include:

1. a) The calculated key figures
   • - are based on only one single process curve,
   • - shift when the operator deliberately changes the machine settings during the process,
   • - change their magnitude so that they lose their significance,
   • - can no longer be found because the process signal curve changes its characteristics significantly as a result of machine setting changes or process deviations,
   • - are dependent on each other, even if they are determined from different physical quantities.
2. b) The input signals in the range of the process deviations of interest are already within the resolution limits of the sensors, so that stochastic deviations occur in the signal curves.
3. c) The signal curves are incomplete due to the discretization.
4. d) The signal curves are subject to strong noise.
5. e) The signal curves, even of different physical quantities, are dependent on each other.

Summary of the invention

The invention is therefore based on the object of specifying a method for determining process parameters on a cyclically operating machine for processing plastics and other plasticizable materials for producing molded parts, a machine operating in this way and a computer program product with a program code stored on a computer-readable medium for carrying out the method, in which the above-mentioned disadvantages are avoided by determining process parameters in such a way that they change in a suitable manner in the event of process changes. in such a way that their significance is retained.

Description of the invention

This object is achieved by a method according to the features of patent claim 1.

1. a) Ermitteln und Speichern mindestens eines teilweisen ersten Kurvenverlaufs mindestens einer ersten Messgröße mittels mindestens eines Sensors in mindestens einem Prozesszyklus,
2. b) optionales, mindestens einstufiges Verarbeiten des mindestens einen ersten Kurvenverlaufs mittels mathematischer und/oder statistischer Verfahren zur Erzeugung eines davon abgeleiteten Kurvenverlaufs,
3. c) Ermitteln und Speichern mindestens eines charakteristischen Werts des mindestens einen teilweisen ersten Kurvenverlaufs oder optional des verarbeiteten Kurvenverlaufs,

This is a method for determining process parameters on a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts, wherein the machine has a control system for carrying out the method, comprising the following steps of a reference determination cycle:

1. a) determining and storing at least a partial first curve of at least one first measured variable by means of at least one sensor in at least one process cycle,
2. b) optional, at least one-stage processing of the at least one first curve using mathematical and/or statistical methods to generate a curve derived therefrom,
3. c) determining and storing at least one characteristic value of the at least one partial first curve or optionally of the processed curve,

1. d) Nutzung des ermittelten mindestens einen charakteristischen Werts als Index auf mindestens eine zu ermittelnde Prozesskennzahl in einem mindestens einstufig verarbeiteten Kurvenverlauf einer weiteren Messgröße,
2. e) Nachverarbeiten der mindestens einen Prozesskennzahl aus mindestens einem Referenzermittlungszyklus mittels mathematischer und/oder statistischer Methoden zu mindestens einem gewichteten Prozesskennzahl-Referenzwert und Speichern desselben, und umfasst weiterhin die folgenden Schritte eines Produktionszyklus:
3. f) analoges Wiederholen der Schritte a) bis e) aus dem Referenzermittlungszyklus zur Ermittlung mindestens eines gewichteten Prozesskennzahlwerts und Speichern desselben.
4. g) abweichungsabhängiges Berechnen und Speichern mindestens eines Werts und/oder Anwenden mindestens eines gespeicherten Werts in einem Kompensationseingriff auf mindestens eine vorbestimmte Stellgröße mittels mindestens eines Aktors bei sich verändernden Produktionsbedingungen zur Nachführung der mindestens einen weiteren Prozesskennzahl.

The procedure also includes the following additional steps of the reference determination cycle:

1. d) use of the determined at least one characteristic value as an index for at least one process key figure to be determined in a curve of another measured variable processed in at least one stage,
2. e) post-processing the at least one process key figure from at least one reference determination cycle using mathematical and/or statistical methods to produce at least one weighted process key figure reference value and storing the same, and further comprises the following steps of a production cycle:
3. f) analogously repeating steps a) to e) from the reference determination cycle to determine at least one weighted process key figure value and storing it.
4. g) deviation-dependent calculation and storage of at least one value and/or application of at least one stored value in a compensation intervention to at least one predetermined manipulated variable by means of at least one actuator under changing production conditions in order to track the at least one further process key figure.

This advantageously means that process key performance indicators are determined in such a way that they shift in an appropriate manner when processes change, so that their significance is retained.

This advantageously avoids the need to record complete measurement curves by stopping the recording after finding defined characteristic figures and carrying out the evaluation directly. This enables process deviations to be compensated for earlier in the cycle, ideally immediately after the deviation and therefore preferably in the same cycle.

Preferably, several signal or curve profiles are examined, whereby they are processed in a cascaded manner by using the "location" of a key figure from one curve profile to index a key figure in another curve profile. Practical tests on the machine and modeling work or recourse to simulation data are not required.

The process can advantageously be carried out without any intervention by the operator and also when the operator intends to make changes to the settings.

Preferably, this involves determining and mathematically and/or statistically processing the curve until the (at least one) process indicator being sought is identified. After identifying a characteristic value at a characteristic point in the curve, which has preferably already been processed, a direct jump is made to another curve, which was recorded and processed at the same time or with a time delay, in order to determine a dependent process indicator there.

This process can preferably also be carried out with another characteristic point so that another dependent process key figure can be determined. From several determined (dependent) process key figures, further dependent process key figures can in turn be determined. This leads to cascading.

This advantageously makes it possible to produce a homogeneous molded part with a constant weight or fill level and a specified quality. This suppresses disturbances in the calculations within a process cycle.

The disadvantage of the aforementioned methods, which compare complete curves with a reference The problem of having to be equal and thus only being able to provide a result of the calculation (e.g. an offset) with a time delay is solved by iteratively searching for characteristic points in the curves and the calculation can usually be completed in the same process phase after they have been recorded. This also makes it possible to detect characteristic points in further curves that would not be detectable as distinctive characteristic points without the indexing from the first curve.

The object is also achieved by a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts according to the features of patent claim 18, wherein the machine has a controller for carrying out the method, at least one sensor and at least one actuator, and is suitable for carrying out the method. The machine can thus be operated more smoothly and energy efficiently.

The object is further achieved by a computer program product with a program code that is stored on a computer-readable medium according to the features of patent claim 19, wherein the program code of the computer program product is suitable for carrying out the method by means of the cyclically operating machine. The advantages of the method are thereby achieved.

Advantageous further developments are the subject of the dependent patent claims.

Preferably, the steps of the reference determination cycle and the steps of the production cycle are carried out analogously to the determination of further process indicators, whereby at least one further process indicator is determined, which in turn/alternatively can be used to track the at least one process indicator. This advantageously makes it possible to determine the "suitable" process indicator depending on the change, which is suitable for a compensation intervention.

Preferably, the method is carried out until a characteristic point is found in the first curve profile processed in at least one stage, using a characteristic value determined at the last examined characteristic point of the derived measured variable as an index to at least one further process key figure of the at least one further derived measured variable.

Preferably, the value of the compensation intervention is determined either from a pre-determined model or during the ongoing process by iteratively observing the reaction of the process to changes in the machine settings, the environmental conditions or the material properties of the plasticizable materials or as a result of wear processes on elements of the machine. This allows the machine to respond advantageously in advance or independently to changed conditions and to be operated reliably and energy-efficiently.

Preferably, when using a pre-determined model, this is determined in the reference determination cycle and/or by introducing process knowledge prepared for the method, whereby the model is either known in advance or is obtained using mathematical methods. The use of process knowledge advantageously allows the machine to access experience that can either be derived from the experience of an experienced machine operator and/or that arises during operation, in particular during operation of the respective machine.

Preferably, in step g) before the stored value of the compensation intervention is used, a logical check of the compensation intervention is carried out, on the basis of which the compensation intervention is either carried out or rejected. This can advantageously counteract settings that could be detrimental to the operation of the machine or the production of molded parts, for example, and the production of defective parts can be avoided. For example, it may be advantageous to divide large compensation interventions into several subsequent cycles and thus prevent a sudden change in the process.

To further improve these advantages, operator intervention can preferably be integrated into the logical test of the compensation intervention, either interactively or by presetting.

Preferably, the logical test advantageously and where appropriate includes settings that can have a systematic influence on the operation of the machine. This can in particular be one or more of the following settings: The value of the compensation intervention, the current process state, a specific machine technology, such as adjustment ranges, time behavior, performance, control engineering aspects and equipment specifics, as well as previous control interventions by an operator. For example, a compensation intervention can advantageously be delayed if a non-stationary operating state exists, e.g. if the compensation parameter is a temperature that must first stabilize before further control intervention, or if the Machine is still in an unstable start-up process.

Preferably, the logical test of the compensation intervention is based on simple or complex if-then conditions and/or on an evaluation using statistical methods, signal processing methods and methods using artificial intelligence. This advantageously ensures a quick test while reducing the computing time. If, for example, several compensation parameters are provided, a faster-acting parameter can be adjusted and returned to overcompensate if a slower compensating parameter slowly increases its effect on the process. In this way, the achievement of the desired process state can be accelerated.

Preferably, the logical test of the compensation intervention includes information from different machine parts, such as from different drive trains, from machine and tool-related signals, from the environment, from feedback from an operator and from external devices or systems or master computers. This advantageously ensures a systematic, holistic picture of the operation of the machine, which is used for logical testing.

Preferably, an operator blocks compensation interventions on control variables and/or sets them as preferred. If the process provides several independent compensation mechanisms, for example, the operator can advantageously select the currently most suitable compensation parameters based on his or her actual process view and/or deselect unsuitable ones. This means that certain processes during molded part production and any resulting characteristics of the molded parts to be produced can be deliberately suppressed or preferred.

It is particularly preferable to examine several determined process parameters in relation to one another, e.g. using correlation or regression analyses, in order to advantageously determine the most effective intervention in the respective situation and to eliminate a correlation between parameters of different curves. This avoids overcompensation, which can occur in a primary forming process such as injection molding due to interactions between process variables. The process knowledge is taken into account when determining the compensation intervention. If the determination is made during the ongoing process, stored parameters from several cycles are continuously used for this purpose; if the determination is made beforehand, it is introduced statically.

Preferably, measured variables are weighted with previously known and/or process-determined material characteristics of the plasticizable materials to be processed. This advantageously increases the accuracy of the process and allows fluctuations in material properties, especially when processing recyclates, to be dealt with effectively.

Preferably, a mathematical method is used to determine the further curve progression, which is selected and/or parameterized from material properties or material class properties, generates a curve progression derived therefrom, and/or a statistical method is used, which weights material properties or material class properties, generates a curve progression derived therefrom. These curve progressions can advantageously be used material-specifically to determine the key figure and thus represent an additional compensation option.

It is highly preferable to examine the curve only at singular characteristic points of the reference determination/production cycle, using iterative signal methods to search for at least one process key figure and ending the evaluation as soon as the process key figure has been determined. This advantageously leads to reduced computational effort (and reduced complexity). In particular, this approach often allows immediate compensation interventions within the same sub-cycle.

• - mit einem Nachführwert, der mehrere Prozesszyklen überspannt,
• - mit einem Nachstellbetrag, der zyklisch neu ermittelt wird,
• - im aktuellen Teilprozesszyklus,
• - in einem folgenden Teilprozesszyklus,
• - im aktuellen Prozesszyklus,
• - über den aktuellen Prozesszyklus hinaus, z. B. auch über mehrere Prozesszyklen hinweg,
• - zyklisch vollständig,
• - in einem iterativen Ansatz zyklisch anteilig,
• - gedämpft oder verstärkt,
• - direkt bei Erkennung,
• - beliebig zyklisch oder zeitlich verzögert,
• - abhängig von Stabilitätsuntersuchungen,
• - abhängig von logischen Entscheidungsvorgängen.

The tracking of at least one compensation variable is preferably carried out particularly effectively by applying at least one of the following measures:

• - with a tracking value that spans several process cycles,
• - with an adjustment amount that is recalculated cyclically,
• - in the current sub-process cycle,
• - in a subsequent sub-process cycle,
• - in the current process cycle,
• - beyond the current process cycle, e.g. across several process cycles,
• - cyclically complete,
• - in an iterative approach cyclically proportionally,
• - attenuated or amplified,
• - immediately upon detection,
• - arbitrarily cyclical or delayed,
• - depending on stability studies,
• - dependent on logical decision-making processes.

Preferably, stochastic fluctuations in the at least one measured variable are detected and filtered out using statistical and iterative methods in order to exclude them from the further calculation steps. This advantageously further reduces the susceptibility of the process to errors and in particular prevents, for example, different scanning increments, manufacturing tolerances, etc. on different machine types from inadmissibly influencing the compensation interventions.

The features listed individually in the patent claims can be combined with one another in a technologically meaningful manner and can be supplemented by explanatory facts from the description and by details from the figures, whereby further embodiments of the invention are shown.

• 1 die Nutzung eines charakteristischen Werts aus einer ggf. mehrstufig verarbeiteten 1. Messgröße als Index auf eine Prozesskennzahl in einer ggf. mehrstufig verarbeiteten 2. Messgröße,
• 2 die Kaskadierung der Kennzahlermittlung aus mehreren Messgrößen mittels Indizierung auf einen oder mehrere weitere Kurvenverläufe,
• 3 ein Beispiel materialspezifischer Daten, die gem. 1 analog zu Kurvenverläufen aus dem Prozess verwendet werden können,
• 4 - 6b verschiedene Ablaufdiagramme.

The invention will now be explained in more detail using an embodiment. Shown are:

• 1 the use of a characteristic value from a possibly multi-stage processed 1st measurement variable as an index to a process key figure in a possibly multi-stage processed 2nd measurement variable,
• 2 the cascading of the key figure determination from several measured variables by indexing to one or more further curves,
• 3 an example of material-specific data according to 1 can be used analogously to curves from the process,
• 4 - 6b various flow charts.

Description of preferred embodiments

Before the invention is described in detail, it should be noted that it is not limited to the particular components of the device and the particular method steps, since these components and methods may vary. The terms used herein are intended only to describe particular embodiments and are not used in a limiting sense. In addition, when the singular or indefinite articles are used in the description or in the claims, this also refers to the plural of these elements, unless the overall context clearly indicates otherwise.

1. a) Ermitteln und Speichern mindestens eines teilweisen ersten Kurvenverlaufs 12, 27 mindestens einer ersten Messgröße 10 mittels mindestens eines Sensors in mindestens einem Prozesszyklus,
2. b) optionales, mindestens einstufiges Verarbeiten des mindestens einen ersten Kurvenverlaufs 12, 27 mittels mathematischer und/oder statistischer Verfahren zur Erzeugung eines davon abgeleiteten Kurvenverlaufs 20,
3. c) Ermitteln und Speichern mindestens eines charakteristischen Werts des mindestens einen teilweisen ersten Kurvenverlaufs oder optional des verarbeiteten Kurvenverlaufs 20,
4. d) Nutzung des ermittelten mindestens einen charakteristischen Werts 30 als Indizierung 80 auf mindestens eine zu ermittelnde Prozesskennzahl 40, 50, 60 in einem mindestens einstufig verarbeiteten Kurvenverlauf 25 einer weiteren Messgröße,
5. e) Nachverarbeiten der mindestens einen Prozesskennzahl 40, 50, 60 aus mindestens einem Referenzermittlungszyklus mittels mathematischer und/oder statistischer Methoden zu mindestens einem gewichteten Prozesskennzahl-Referenzwert und Speichern desselben,
6. f) analoges Wiederholen der Schritte a) bis e) aus dem Referenzermittlungszyklus zur Ermittlung mindestens eines gewichteten Prozesskennzahlwerts und Speichern desselben,
7. g) abweichungsabhängiges Berechnen und Speichern mindestens eines Werts und/oder Anwenden mindestens eines gespeicherten Werts in einem Kompensationseingriff auf mindestens eine vorbestimmte Stellgröße mittels mindestens eines Aktors bei sich verändernden Produktionsbedingungen zur Nachführung der mindestens einen weiteren Prozesskennzahl 40, 50, 60,
8. h) Wiederholen der Schritte f) und g) zum Fertigen weiterer Formteile.

The invention is, on the one hand, a method for determining process parameters on a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts, wherein the machine has a control system for carrying out the method, comprising the following steps of a reference determination cycle:

1. a) determining and storing at least one partial first curve 12, 27 of at least one first measured variable 10 by means of at least one sensor in at least one process cycle,
2. b) optional, at least one-stage processing of the at least one first curve 12, 27 by means of mathematical and/or statistical methods to generate a curve 20 derived therefrom,
3. c) determining and storing at least one characteristic value of the at least one partial first curve or optionally of the processed curve 20,
4. d) use of the determined at least one characteristic value 30 as an index 80 for at least one process key figure 40, 50, 60 to be determined in a curve 25 of another measured variable processed in at least one stage,
5. e) post-processing the at least one process key figure 40, 50, 60 from at least one reference determination cycle using mathematical and/or statistical methods to produce at least one weighted process key figure reference value and storing the same,
6. f) analogous repetition of steps a) to e) from the reference determination cycle to determine at least one weighted process key figure value and storing it,
7. g) deviation-dependent calculation and storage of at least one value and/or application of at least one stored value in a compensation intervention to at least one predetermined manipulated variable by means of at least one actuator in the event of changing production conditions for tracking the at least one further process key figure 40, 50, 60,
8. h) Repeat steps f) and g) to produce further molded parts.

This advantageously makes it possible to produce a homogeneous and weight-constant molded part with a specified quality. Disturbances in the calculations within a process cycle are suppressed and the process cycle times are shortened by minimizing the computing times.

The first curve of at least one first measured variable can be an existing curve of an already existing measured variable and/or another curve of another measured variable, which is then included in the process as the first measured variable according to the procedure. The same applies to the process key figure, which can be an existing and/or another process key figure that is determined according to the procedure.

In contrast to the state of the art, the determination and storage of curve progressions is generally not carried out completely over the entire process cycle, but can also be carried out only over parts of a process cycle or can also be focused on sections in which significant features or properties of a process cycle regularly become apparent in the process.

It is advantageous to carry out iterative processing of the measurement curves over each individual measurement point until a characteristic point is identified at which further recording can be stopped and the following evaluation steps are initiated. Finding such a characteristic point provides a process key figure, but in particular an index into a further curve progression. In principle, it is also possible to carry out several reference cycles and to evaluate the values or curve sections determined in the process (weight, average, sort, etc.) in order to arrive at the reference values then used.

The measured variables of the sensors can be, for example, an injection pressure, screw path, torque, a pressure or a temperature in a screw antechamber of a conveyor screw or a temperature of a tool of an injection molding machine.

The actuator control variables can be, for example, a switching point, a holding pressure level, a holding pressure duration or an injection profile.

The measured variables can be recorded over one or more process cycles. Furthermore, the manipulated variables can be changed in one or any process cycles.

The processing of at least one curve can be carried out using multivariate mathematical and/or multivariate statistical methods. For example, a pressure curve can be examined using a first process characteristic in order to obtain an index of a position in a processed temperature curve or a processed material property curve. Curves that are not determined in the manufacturing process but contain, for example, basic information on process relationships can be integrated into the process in an analogous manner.

For example, the possibly multi-stage processing of a curve using mathematical and/or statistical methods can be carried out from a simple filtering, a derivation (differential calculus) or an integration (integral calculus) or by applying one or more of these methods several times. Thus, possibly multi-stage processing is to be understood as either single-stage or multi-stage processing.

1 shows an example of the process key figure determination. In 1a) the curve 10 of a first measured value is shown. This is either used as it is, but is usually processed using mathematical and/or statistical methods. The result shows 1b) as curve 20 of the first measured value, possibly processed in several stages. 1c ) shows the determination of a first process key figure 30 in the possibly multi-stage processed curve 20 at a characteristic point of the curve 20. By means of indexing 80 of the first process key figure 30, a second process key figure 40 is determined in the possibly multi-stage processed curve 25 of a second measured variable. The recording and processing of the curves is preferably carried out in partial sections, particularly preferably in the form of individual measuring points, until a termination criterion in the form of a characteristic point is reached. The determined process key figures 30 or 40 and the indexing 80 are usually at the end or shortly before the end of the recorded and processed section of the curves 10, 11, 20, 25.

2 shows analogous to 1d the cascading of the key figure determination from several measured variables by means of indexing 80 to one or more further curve progressions, from which one or more further process key figures 50, 60 are formed. Further process key figures 65 can be determined from several process key figures formed in this way. The measured variables can be weighted with previously known and/or in-process determined material characteristics of the plasticizable materials to be processed.

The value of the compensation intervention can be determined either from a pre-determined model or during the running process by iteratively observing the reaction of the process to changes in machine settings, environmental conditions, the material properties of the plasticizable materials or as a result of wear and tear on elements of the machine.

A pre-determined model can, for example, be a model-based relationship between certain curve characteristics in the curve of a measured variable - or its differentials or integrals - and related characteristics of a second measured variable - or its differentials or integrals - which have already been determined to be demonstrably correct in principle or on the respective machine.

1. 1. Zyklische Ermittlung der Abweichung zwischen Prozesskennzahl-Referenzwerten aus dem Referenzermittlungszyklus und analog definierten und ermittelten Prozesskennzahlen in einem ersten Produktionszyklus,
2. 2. Ausführung eines Kompensationseingriffs mit einer vordefinierten Einstellgröße mit einem vordefinierten Wert im selben Zyklus,
3. 3. Optional: Übergang zu einem zweiten Produktionszyklus,
4. 4. Zyklische Ermittlung der Abweichung zwischen Prozesskennzahl-Referenzwerten und Prozesskennzahlen des aktuellen zweiten Produktionszyklus,
5. 5. Adaptive Anpassung des Werts des Kompensationseingriffs, bis die Prozesskennzahlabweichung ausgeglichen ist.

The value of the compensation intervention can be determined as described in the following steps:

1. 1. Cyclic determination of the deviation between process key figure reference values from the reference determination cycle and analogously defined and determined process key figures in a first production cycle,
2. 2. Execution of a compensation intervention with a predefined setting variable with a predefined value in the same cycle,
3. 3. Optional: Transition to a second production cycle,
4. 4. Cyclic determination of the deviation between process key figure reference values and process key figures of the current second production cycle,
5. 5. Adaptive adjustment of the compensation intervention value until the process key figure deviation is compensated.

When using a pre-determined model, this can be determined in the reference determination cycle and/or by introducing process knowledge prepared for the procedure, which is either previously known or obtained using mathematical methods.

Advantageously, in step g) before using the stored value of the compensation intervention, a logical check of the compensation intervention can be carried out, on the basis of which the compensation intervention is either carried out or rejected.

The logical test can include the value of the compensation intervention, the current process state, a specific machine technology, such as adjustment ranges, time behavior, performance, control engineering aspects and equipment specifics, as well as previous control interventions by an operator.

The logical test of the compensation intervention can be based on simple or complex if-then conditions and/or on an evaluation using statistical methods, signal processing methods and methods using artificial intelligence.

The logical test of the compensation intervention can include information from different machine parts, such as from different drive trains, from machine and tool-related signals, from the environment, from feedback from an operator as well as from external devices or systems or master computers.

The automatic implementation of the compensation intervention can advantageously be preceded by an interactive assessment/approval by an operator. An operator can advantageously block compensation interventions on control variables and/or set them as preferred. This means that certain processes during the molded part production and any resulting characteristics of the molded parts to be produced can be deliberately suppressed or preferred. The operator thus retains the power of decision despite the extensive automation of the process.

Several determined process indicators (30, 40, 50, 60, 65) can be examined in relation to one another, e.g. using correlation or regression analyses. For example, the temperature-dependent part can be calculated from a pressure-dependent process indicator.

Measured variables can advantageously be weighted with previously known and/or in-process determined material characteristics of the plasticizable materials to be processed. This increases the accuracy of the process. 3 shows this schematically using a melt density curve p versus a pressure p of a plasticizable material with the melting temperature T of the material increasing from top to bottom.

The curve can advantageously be examined only at singular characteristic points of the reference determination/production cycle, whereby iterative signal methods are used to search for at least one process key figure (30, 40, 50, 60, 65) and the evaluation is terminated as soon as the process key figure (30, 40, 50, 60, 65) has been determined. This means that compensation can be carried out in the same sub-cycle if necessary.

• - mit einem Nachführwert, der mehrere Prozesszyklen überspannt,
• - mit einem Nachstellbetrag, der zyklisch neu ermittelt wird,
• - im aktuellen Teilprozesszyklus,
• - in einem folgenden Teilprozesszyklus,
• - im aktuellen Prozesszyklus,
• - über den aktuellen Prozesszyklus hinaus, z. B. auch über mehrere Prozesszyklen hinweg,
• - zyklisch vollständig,
• - in einem iterativen Ansatz zyklisch anteilig,
• - gedämpft oder verstärkt,
• - direkt bei Erkennung,
• - beliebig zyklisch oder zeitlich verzögert,
• - abhängig von Stabilitätsuntersuchungen,
• - abhängig von logischen Entscheidungsvorgängen.

The tracking of at least one process key figure (30, 40, 50, 60, 65) is carried out optionally:

• - with a tracking value that spans several process cycles,
• - with an adjustment amount that is recalculated cyclically,
• - in the current sub-process cycle,
• - in a subsequent sub-process cycle,
• - in the current process cycle,
• - beyond the current process cycle, e.g. across several process cycles,
• - cyclically complete,
• - in an iterative approach cyclically proportionally,
• - attenuated or amplified,
• - immediately upon detection,
• - arbitrarily cyclical or delayed,
• - depending on stability studies,
• - dependent on logical decision-making processes.

Using statistical and iterative methods, stochastic fluctuations in at least one measured variable can be detected and filtered out in order to exclude them from the further calculation steps. This further increases the quality of the molded parts to be produced, further reduces the susceptibility to errors in the process and improves the stability of the process.

In the following, the invention is described on the basis of the 4 to 5c The procedures described are explained. 4 explains the basic procedure, while 5 in particular shows further details.

First, according to 4 in step 100, a conventional setting of an injection molding process on the machine for processing plasticizable materials. For this purpose, the injection molding process is set so that the machine can run, i.e. an injection molded part can basically be produced. In the following step 110, a currently available injection molding process is then set as a reference. This reference process was an injection molding process in which a good part can already be produced. However, this is not absolutely necessary. The reference process can also be set in response to an operator input, automated or triggered by external systems, in which the operator or artificial intelligence determines that the current process is basically suitable as a reference.

If such a process is established as a reference, in the next step at least a partial curve 12, 27 of at least one first measured variable is determined and stored by means of at least one sensor in a process cycle, wherein in order to record specific key figures or characteristic points from the curves 10, 25 recorded by sensors, specific process key figures or characteristic points 30, 30 are recorded in step 120.

The sensor data can come from one or more sensors, such as a sensor for determining the injection pressure, a position sensor for determining the screw path or a sensor for recording the torque of the drive. Other sensor data can also be used, such as the pressure in the screw antechamber or the temperature of the injection molding unit and/or mold. Curves from other sources can also be processed in the same way, such as material data or similar that are available in curve form (see. 3 ).

Optionally, at least one first curve is processed in at least one stage using mathematical and/or statistical methods to generate a curve 20 derived therefrom. The recording of process parameters can take place over part of a cycle, an entire cycle or even over several cycles, if necessary using statistical methods and possibly using specially adapted reference sequences. Characteristic points can be, for example, a switchover point between injection and holding pressure, the holding pressure level, the holding pressure duration or a specific injection profile, but generally any point in a curve from sensor data or from other data sources, such as curves of material parameters that can be obtained using mathematical/statistical methods.

If curve progressions were already available previously, the method can also determine further curve progressions based on additional measured variables.

In the following step 130, the process key figures or characteristic points of the reference are compared with the corresponding values in the current process cycle.

In step 140, the comparability of the reference and the current cycle is validated. If this comparability is given, in step 150, new setting data is calculated or the process is actively intervened on the basis of the previously recorded differences in the process parameters between the reference and the current cycle. In this case, the formation of the molded part is completed in step 170 and the process is continued in step 130 with the changed setting data.

If step 140 shows that the reference and the current cycle are not comparable, no setting data is calculated according to step 160. Instead, an output or an intervention in the process takes place by marking the manufactured molded part as a defective part, stopping the cycle at the end and/or an error evaluation takes place.

5a to 5c describe the concrete procedure, which corresponds to the rough flow charts in accordance with. 4 and 6 is underlined.

Out of 4 , Steps 110 and 120 show that characteristic points or process indicators are first determined in (one or more) reference cycles. This section is 5a further detailed.

5a shows how partial curves are recorded from two curves 10, 25 (any number of curves can be included in the processing), whereby partial recording can, in the simplest case, mean a single measuring point 12, 27 per processed curve. The recorded section of each curve is preferably processed (if necessary, several times). Processing is understood to mean any application of mathematical methods, such as simple counting, sorting, filtering, derivation, integration, etc.

The partial processes are recorded step by step in an iterative manner. As soon as a predetermined characteristic point is reached on curve 1, a process key figure can be determined from it if required. In any case, however, the characteristic point is used as a termination criterion for the recording and processing in all curves examined simultaneously and is used to index curve 1 in the other curve(s), where it is used to determine characteristic points 39 and process key figures 40 of at least one other curve 25.

The process key figures 40 are determined over at least one reference cycle and can be post-processed using mathematical/statistical methods (step 41), e.g. in the form of averaging, weighting or the like. After completion of the reference cycles, one or more reference process key figures 42 from one or more curve profiles 25 are available.

Out of 4 , Step 130 shows that the characteristic points or process indicators of the reference are compared with those of the current cycle.

For this purpose, as in 5b shown the process of 5a to determine the reference value for each of the z production cycles, the process is carried out identically and the one or more production process key figures 42 of each individual production cycle are compared with the one or more reference process key figures 42 (step 130). The production process key figures can also be further processed mathematically over several cycles in step 41 in order to arrive at cross-cycle production process key figures 42.

According to 5c The comparison (step 130) between the production process key figure and the reference process key figure produces a value for the distance 135 between the process key figures. Any mathematical distance measure can be used for this purpose.

The distance value is then subjected to a logical test (step 140), which determines whether a compensation intervention on a specified setting value is carried out (step 150) or omitted (step 160). An example of such a logical test can be to determine the comparability of the reference curves and the production curves. Other possibilities were mentioned in the description of the exemplary embodiments. All appropriate mathematical methods can be used. A decisive operator intervention can also be possible at this point.

6a , 6b show an overview of an application of the procedure based on a concrete influence on the switching point Up. It should be noted that for the implementation in detail, procedural steps according to. 1 or the process 5a - 5c take place.

As per 4 As explained, in steps 100 and 110, a conventional injection molding process is first set up and then a current process cycle is defined as a reference.

The process starts with the injection or the formation of the molded part in step 620. In step 630, a path point of the conveyor screw, namely the characteristic point "Start filling" sf _ref , is recorded as a characteristic point, which corresponds to the time at which plasticizable material is injected into the injection mold. In step 640, the screw path sf _ref of the reference process cycle is compared with the value sf from the current cycle. $${\text{ds}}_{\text{i}}={\text{sf}}_{\text{ref}}-{\text{sf}}_{\text{i}}$$

wobei

• Up_neu der angepasste Umschaltpunkt,
• Up_soll der vorher gesetzte Sollwert des Umschaltpunkts, sowie
• k ein konfigurierbarer zyklusabhängiger Faktor und/oder ein materialabhängiger Faktor ist.

In step 650, it is now checked whether the maximum permissible difference value ds _max is greater than or equal to the determined value ds _i of the cycle. If this is the case, the switching point Up is changed in step 655 as follows: $${\text{Up}}_{\text{new}}={\text{Up}}_{\text{should}}+{\text{ds}}_{\text{i}}*\text{k}$$

where

• Up _new the adjusted switching point,
• Up _{should be} the previously set setpoint of the switching point, as well as
• k is a configurable cycle-dependent factor and/or a material-dependent factor.

However, if the comparison in step 650 shows that the value ds _i > ds _max , the maximum adjustment path according to the set upper limit is exceeded, so that in step 660 the molded part is reported as a defective part with subsequent error evaluation.

wobei

• s_{NDdiff, ref} = Volumendifferenz im Referenzzyklus
• s_{NDdiff, akt} = Volumendifferenz im aktuellen Zyklus

In step 670, a switch to holding pressure takes place after injection has taken place, whereby a further process key figure is now a characteristic volume difference s _NDdiff , e.g. between the start and end of the holding pressure in step 680 in 6b In step 690, this volume difference s _NDdiff is compared with the reference value from the reference determination cycle to $$\Delta {\text{s}}_{\text{NDdiff}}={\text{s}}_{\text{NDdiff},\text{ref}}-{\text{s}}_{\text{NDdiff},\text{act}}$$

where

• s _{NDdiff, ref} = volume difference in the reference cycle
• s _{NDdiff, act} = volume difference in the current cycle

In step 691, the holding pressure is changed depending on ND _ref and Δs _NDdiff .

With step 692, the post-printing is completed and thus the molding is also completed in step 693, so that the next cycle is continued with the information obtained in step 620. This allows corresponding moldings to be manufactured in the following cycles.

The invention also relates to a cyclically operating machine for processing plastics and other plasticizable materials for producing molded parts, wherein the machine has a controller for carrying out the method, at least one sensor and at least one actuator, wherein the machine is suitable for carrying out the method.

Furthermore, the invention relates to a computer program product with a program code stored on a computer-readable medium, wherein the program code of the computer program product is suitable for carrying out the method by means of the cyclically operating machine.

List of reference symbols

10

Curve of a measured quantity,

11

Curve of another measured value,

12

partial curve / measuring point

20

processed curve of a first measured value

25

processed curve of another measured value

27

partial curve / measuring point

30

Process key figure for the first measured variable, possibly processed in several stages

29, 39

characteristic point

40 2.

Process key figure by indexing using a 1st process key figure on a possibly multi-stage processed 2nd curve of a measured variable

41

Post-processing

42

Reference process key figure

50, 60, 65

further process key figure

80

Indexing of process metrics

100 - 130

steps

135

Step

140 - 170

steps

620 - 650

steps

655

Step

660 - 690

steps

691 - 693

steps

QUOTES INCLUDED IN THE DESCRIPTION

This list of documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions.

Cited patent literature

• WO 9114562 A1 [0009, 0010]
• EP 2583811 B1 [0010]
• WO 9609926 A1 [0012]
• WO 9712740 A2 [0013]
• TW201706112A [0014]
• DE 102018107233 A1 [0015]
• DE 102020125655 A1 [0016]
• DE 102021109851 A1 [0017]

Claims (19)

Method for determining process parameters on a cyclically operating machine for processing plastics and other plasticizable materials to produce molded parts, the machine having a controller for carrying out the method, comprising the following steps of a reference determination cycle: a) determining and storing at least one partial first curve profile (12, 27) of at least one first measured variable (10) by means of at least one sensor in at least one process cycle, b) optional, at least one-stage processing of the at least one first curve profile (12, 27) by means of mathematical and/or statistical methods to generate a curve profile (20) derived therefrom, c) determining and storing at least one characteristic value of the at least one partial first curve profile or optionally of the processed curve profile (20), characterized by , d) using the determined at least one characteristic value (30) as an index (80) for at least one process parameter (40, 50, 60) to be determined in a curve profile (25) of a further Measured variable, e) post-processing the at least one process key figure (40, 50, 60) from at least one reference determination cycle by means of mathematical and/or statistical methods to at least one weighted process key figure reference value and storing the same, f) analogously repeating steps a) to e) from the reference determination cycle to determine at least one weighted process key figure value and storing the same, g) deviation-dependent calculation and storing of at least one value and/or application of at least one stored value in a compensation intervention to at least one predetermined manipulated variable by means of at least one actuator under changing production conditions in order to track the at least one further process key figure (40, 50, 60), h) repeating steps f) and g) to produce further molded parts. Procedure according to Claim 1 , characterized in that steps a) to e) from the reference determination cycle and steps f) and g) are carried out analogously to determine further process key figures (50, 60), whereby at least one further process key figure (65) is determined, which in turn/alternatively can be used to track the at least one process key figure (40). Procedure according to Claim 1 or 2 , characterized in that step d) comprises an iterative repetition of steps a) to c) on at least one further partial curve of the first measured variable (10) and/or a further measured variable (11) until a characteristic point is found in the first curve (20) processed in at least one stage, using a characteristic value (30) determined at the last examined characteristic point of the derived measured variable (25) as an index (80) to at least one further process key figure (40) of the at least one further derived measured variable (25). Method according to one of the preceding claims, characterized in that the value of the compensation intervention is determined either from a pre-determined model or in the ongoing process by iterative observation of the reaction of the process to changes in at least one of the variables, comprising machine settings, environmental conditions, material properties of the plasticizable materials, or as a result of wear processes on elements of the machine. Procedure according to Claim 4 , characterized in that when using a pre-determined model, this is determined in the reference determination cycle and/or by introducing process knowledge prepared for the method, which is either previously known or obtained by means of mathematical methods. Method according to one of the preceding claims, characterized in that in step g) before the use of the stored value of the compensation intervention, a logical check of the compensation intervention is carried out, on the basis of which the compensation intervention is either carried out or rejected. Procedure according to Claim 6 , characterized in that the logical test includes the value of the compensation intervention, the current process state, a concrete machine technology, such as adjustment ranges, time behavior, performance, control engineering aspects and equipment specifics, as well as previous control interventions by an operator. Procedure according to Claim 6 or 7 , characterized in that the logical test of the compensation intervention is based on simple or complex if-then conditions and/or on an evaluation using statistical methods, signal processing methods and methods utilizing artificial intelligence. Method according to one of the Claims 6 until 8th , characterized in that the logical test of the compensation intervention includes information from different machine parts, such as from different drive trains, from machine and tool-related signals, from the environment, from the feedback of an operator and from external devices or systems or master computers. Method according to one of the preceding claims, characterized in that the automatic execution of the compensation intervention is preceded by an interactive assessment/approval by an operator. Method according to one of the preceding claims, characterized in that an operator blocks and/or preferentially sets compensation interventions on manipulated variables. Method according to one of the preceding claims, characterized in that several determined process parameters (30, 40, 50, 60) are examined in their relationship to one another, e.g. by means of correlation or regression analyses. Method according to one of the preceding claims, characterized in that measured variables are weighted with previously known and/or process-determined material characteristics of the plasticizable materials to be processed. Method according to one of the preceding claims, characterized in that , to determine the further curve progression, a mathematical method is used which, selected and/or parameterized from material properties or material class properties, generates a curve progression derived therefrom, and/or a statistical method is used which, weighted from material properties or material class properties, generates a curve progression derived therefrom. Method according to one of the preceding claims, characterized in that the curve progression is only examined at singular characteristic points of the reference determination/production cycle and wherein the at least one process characteristic number (30, 40, 50, 60) is searched for by means of iterative signal methods and the evaluation is terminated as soon as the process characteristic number (30, 40, 50, 60) has been determined. Method according to one of the preceding claims, characterized in that the tracking of the at least one process characteristic (40, 50, 60) takes place optionally: - with a tracking value that spans several process cycles, - with an adjustment amount that is determined anew cyclically, - in the current sub-process cycle, - in a subsequent sub-process cycle, - in the current process cycle, - beyond the current process cycle, e.g. also across several process cycles, - cyclically completely, - in an iterative approach cyclically proportionally, - dampened or amplified, - directly upon detection, - arbitrarily cyclically or with a time delay, - dependent on stability tests, - dependent on logical decision processes. Method according to one of the preceding claims, characterized in that stochastic fluctuations of the at least one measured variable are detected and filtered out by means of statistical and iterative methods in order to exclude them from the further calculation steps. Cyclically operating machine for processing plastics and other plasticizable materials for producing molded parts, the machine having a control for carrying out the method, at least one sensor and at least one actuator, characterized in that the machine is suitable for carrying out the method according to one of the Claims 1 until 17 to execute. Computer program product with a program code stored on a computer readable medium, characterized in that the program code of the computer program product is suitable for carrying out the method according to one of the Claims 1 until 17 by means of the cyclically operating machine Claim 18 to execute.

Images