DE102010003814B4 - Method for automatic dosing and dosing device - Google Patents

Abstract

Method for the automatic metering of at least one component (H) provided by a metering device (V), the method comprising at least the following steps: - providing the at least one component (H) by means of the metering device (V) in a desired quantity, - measuring a by the dosing device (V) actually provided first actual quantity (Hist) of the at least one component (H) and storage of a measured variable for this measured first actual quantity (Hist) of the at least one component (H) as a measured value, - measurement of at least one second actual quantity ( Hist) after or during further provision of the at least one component (H) and storage of further measured quantities for these second actual quantities (Hist) as further measured values, - calculation of a characteristic curve (L) from the stored measured values and control of the dosage as a function of the characteristic curve ( L), characterized in that the characteristic curve e (L) for which at least one component (H) is determined as a function of the container shape for the at least one component (H).

Description

The invention relates to methods for automatically metering at least one component provided by a metering device, in particular for the automatic metering of a main component and / or an additive of a mixture, and to a metering device with which such a mixture can be fed to a processing machine processing the mixture.

From the document US 4,301,510 A a method for automatic dosing according to the preamble of claim 1 is known.

In particular, in the supply of (plastic) spraying machines with a granulate mixture, which must be supplied as a mixture of a main component and one or more additives of the spraying machine in a predetermined composition or recipe, there is regularly the problem that the injection molding machine to be supplied as a processing machine Mixture portions of the main component and the at least one additive must each be manually entered into a metering device and therefore prone to erroneous entries.

Also can be done by the fixed specification of the amount to be dosed of the components and the resulting amount of at least one additive based on the predetermined and stored in the metering composition or recipe of the supplied mixture no automatic adjustment of the amount of mixture supplied. This means, for example, that with a shutdown of cavities of an injection mold operated with the injection molding machine, less mixture would be necessary, but the metering device would have to be adjusted manually for this purpose.

The US Pat. No. 6,688,493 B2 describes a possibility to adjust manually predefinable set amounts of a material to be supplied via a metering device of a processing machine with the actual amount of material supplied. In this case, it is checked to what extent the actually supplied quantities of material per time correspond to the predetermined set amount of material per time. Depending on the result of this check, a delivery rate of a feed screw connected downstream of the metering device is then varied, with which the supplied mixture is supplied to the processing machine. The result is thus the metering of the US Pat. No. 6,688,493 B2 automatically sure that the downstream processing machine is actually supplied to a set on the metering mixture amount per time.

An automatic adjustment of the dosage to changed operating conditions of the processing machine is the US Pat. No. 6,688,493 B2 just as inferior as a regulation and dosage of individual mixture components to provide a processing machine to be supplied mixture.

Further metering devices are for example in the documents US 4,762,252 A . US 5 110 521 A . DE 10 2009 021 760 A1 and US 811 301 B2 described.

It is an object of the present invention to provide an improved method for automatic metering, improved metering devices and an improved spraying machine.

This object is achieved by the method of claim 1, the metering device according to claim 24 and the injection molding machine according to claim 27.

• – Bereitstellung der mindestens einen Komponente (H) mittels der Dosiervorrichtung (V) in einer Sollmenge, insbesondere in einer von einer mit der Dosiervorrichtung gekoppelten Verarbeitungsmaschine angeforderten Bedarfsmenge,
• – Messung einer durch die Dosiervorrichtung (V) tatsächlich bereitgestellten ersten Istmenge (H_ist) der mindestens einen Komponente (H) und Speicherung einer Messgröße für diese gemessene erste Istmenge (H_ist) der mindestens einen Komponente (H) als ein Messwert,
• – Messung mindestens einer zweiten Istmenge (H_ist) nach oder während weiterer Bereitstellungen der mindestens einen Komponente (H) und Speicherung weiterer Messgrößen für diese zweite Istmengen (H_ist) als weitere Messwerte,
• – Berechnung einer Verlaufskennlinie (L) aus den gespeicherten Messwerten und Steuerung der Dosierung in Abhängigkeit von der Verlaufskennlinie (L).

The method according to claim 1 comprises at least the following steps:

• - Providing the at least one component (H) by means of the metering device (V) in a desired amount, in particular in a demanded by a coupled with the metering processing machine demand quantity,
• Measuring a first actual quantity (H _ist ) of the at least one component (H) actually provided by the dosing device (V) and storing a measured variable for this measured first actual quantity (H _ist ) of the at least one component (H) as a measured value,
• Measuring at least one second actual quantity (H _ist ) after or during further provision of the at least one component (H) and storing further measured quantities for these second actual quantities (H _ist ) as further measured values,
• - Calculation of a characteristic curve (L) from the stored measured values and control of the dosage as a function of the characteristic curve (L).

According to the invention, the calculated characteristic curve for the at least one component is determined as a function of the container shape for the at least one component.

The measurement of the first and the at least one second actual quantity can take place in different ways. As will be explained later, these actual quantities z. B. be determined by the weighings of the container contents and the calculation of corresponding differences. In principle, however, other possibilities of determination are also possible (eg volumetric) for the actual quantity or combinations of determination possibilities.

In particular, a characteristic curve is calculated from measured values (that is, there is an idealized characteristic curve) that can be used for the metering. Such a statistical method dampens measurement errors better than z. B. moving averages.

The calculated characteristic allows the calculation of an expected value for a defined probability of occurrence in trouble-free operation. If a newly measured value lies outside the confidence interval, it can be assumed with some certainty that a fault will occur and countermeasures can be taken.

Cyclic disturbances, such as machine impacts of an injection molding machine, are largely eliminated by the statistical evaluation of the measured values. Since they occur to the same degree with each measurement, they change the slope of the measured idealized characteristic only slightly when a linear regression is used to calculate the characteristic curve.

Advantageously, a fixed and / or an adjustable number of previously stored measured values are used to calculate the characteristic curve.

Advantageously, the calculated characteristic curve for the at least one component is determined in the form of a quadratic dependence for containers with conical sections. The use of a quadratic dependence is an example of a non-linear regression for determining the characteristic curve.

Another mathematical method provides, for example, to determine a characteristic curve, in particular by means of linear regression, and from this a valid average value of a plurality of stored measured values for actual quantities of the main component. A mean value resulting from this calculated characteristic curve, which can be used for measured values for the dosed main components for calculating a desired additive quantity and / or for a plausibility check of subsequent first measured values, is thus significantly more stable and less susceptible to measurement errors.

Advantageously, control parameters of a dosage are obtained by means of at least one test dosage, which form initial values for further dosage. So z. For example, when setting a new recipe setpoint or after a disturbance, set the batch size to a minimum (eg 4 cycles) to achieve a quick settling of the control. When the control is then settled, the batch size can be increased to obtain a more accurate measurement. It makes sense to increase the number of batches by one cycle after each batch until the maximum batch size is reached. Even during the test dosing, a characteristic curve, calculated with a linear or non-linear regression, can be used.

Trajectories calculated by linear or non-linear regression can also be used to determine a mass flow if no dosing device is activated. The thus measured throughput of a main component can then serve as the basis for the addition of a secondary component.

The at least one component may be a flowable substance, in particular a granulate, a liquid, a powder, a free-flowing substance, a pasty mass or a mixture of these substances.

In addition to the weight, the level profile or the pressure profile can also be used as a measured variable. It is also possible to use combinations of the measured values.

• – Bereitstellung der Hauptkomponente mittels der Dosiervorrichtung in einer von der Verarbeitungsmaschine angeforderten Bedarfsmenge, wobei die Größe der Bedarfsmenge von einem Bedarfssignal abhängt, das die Verarbeitungsmaschine an die Dosiervorrichtung übermittelt,
• – Messung einer tatsächlich durch die Dosiervorrichtung bereitgestellten Istmenge der Hauptkomponente und Speicherung mindestens einer Messgröße für diese gemessene Istmenge der Hauptkomponente als mindestens ein erster Messwert,
• – Bestimmung einer Additiv-Sollmenge für das wenigstens eine Additiv auf Basis einer vorgegebenen Zusammensetzung für das zuzuführende Gemisch aus der Hauptkomponente und dem wenigstens einem Additiv, wobei zur Bestimmung der Additiv-Sollmenge mindestens einer der ersten Messwerte für die tatsächlich bereitgestellte Istmenge der Hauptkomponente verwendet wird,
• – Bereitstellung des Additivs mittels der Dosiervorrichtung in der Additiv-Sollmenge und
• – Zuführung des Gemisches aus bereitgestellter Hauptkomponente und bereitgestelltem Additiv an die Verarbeitungsmaschine.

An embodiment of the method for the automatic metering and feeding of a mixture of a main component and at least one additive in a predeterminable composition provides that the mixture is supplied by means of a metering device to a mixture processing processing machine. In this case, the method has at least the following steps:

• Provision of the main component by means of the dosing device in a demand quantity requested by the processing machine, the size of the requirement quantity depending on a demand signal which transmits the processing machine to the dosing device,
• Measuring an actual quantity of the main component provided by the metering device and storing at least one measured variable for this measured actual quantity of the main component as at least one first measured value,
• - Determining an additive setpoint for the at least one additive based on a predetermined composition for the mixture to be supplied from the main component and the at least one additive, wherein at least one of the first measured values for the actually provided actual amount of the main component is used to determine the additive setpoint .
• - Providing the additive by means of the metering device in the additive set amount and
• - Supplying the mixture of provided main component and provided additive to the processing machine.

In carrying out the method, a requirement quantity of the main component is thus initially made available via the metering device, wherein this requirement quantity initially depends exclusively on a demand signal of the downstream processing machine. In this context, a demand signal from the processing machine to the metering device is understood to mean any active or passive information transmission of the processing machine to the metering device, on the basis of which the metering device detects or triggers the provision and / or addition of the main component.

This means, for example, that the opening duration of a slider controlling the feed for the main component of the mixture depends on an (electronic) demand signal which is transmitted from the processing machine, in particular from a conveyor screw of the processing machine, to the metering device. Only after the provision of the demand quantity requested by the processing machine is the actually provided amount of the main component measured. Depending on this determined actual quantity, the at least one additive is then admixed in a desired additive quantity in order to provide a mixture in the manually deposited mixture composition of main component and at least one additive of the processing machine specified in the metering device.

The requirement quantity of the main component can be provided both in the form of a continuous mass flow and in portions, wherein the supply and supply of the main component takes place via a (first) delivery unit of the metering device designed for this purpose.

In the former case of continuous feeding of the processing machine with the main component, a demand signal from the processing machine may be, for example, a "START" signal leading to initial opening of a feed to the main component at the metering device or triggering operation of the hitherto inactive metering device.

If the provision and addition of the desired additive quantity in such an embodiment variant also continue to depend exclusively on this individual demand signal, z. B. be provided that a time for the beginning of the supply of the additive target amount is determined in each case by a manually specifiable time interval according to the demand signal or "START" signal.

In one embodiment, a release opening of the metering device is permanently opened and thus allows the addition of the main component in a conveyor of the processing machine, such as a screw conveyor of a spraying machine. As long as the conveyor or the processing machine is inactive, the dosing device in this case no main component is withdrawn by the processing machine. Only during operation of the processing machine, the main component is moved by means of the conveyor and thus creates the possibility that from the open release opening of the metering further material can flow into the processing machine. This subsequent flow of the main component or this information that the processing machine has resumed its operation and requires new or further amounts of the main component and thus the mixture is detected and evaluated by the metering device as a "demand signal" of the processing machine.

In a preferred, alternative embodiment variant, the processing machine generates further (electronic) demand signals and transmits them to the dosing device in order to determine the time of delivery of the at least one additive or else several additives into a continuous mass flow of the main component. With such a metering of at least one additive into a main component mass flow which runs in the direction of the processing machine, the further demand signals of the processing machine are also regularly referred to as "picking signals".

Preferably, the measurement of the actual quantities for an actually supplied amount of the main component in the differential or difference weighing method, wherein at the time of transmission of the further demand signals or shot signals for the at least one additive, a total weight for the currently in the metering device (still) located main component measured becomes. For example, we measured and stored the total weight of a weighing container of the metering device, in which the main component is held, as a measured variable. By measuring the total weight of this weighing container in determining a demand signal for the metering or feeding of the at least one additive, the actual amount of the main component actually added can be determined by subtracting successive measured values.

Alternatively, a volumetric measurement of the actual quantities of the main component can be carried out by means of a suitable (first) measuring device of the metering device.

Instead of specifying a time for the addition of the additive via a (further) demand signal or "shot signal" of the processing machine, the method and a subsequent dosing device can be further automated by a time for the provision of the additive target amount and / or a period of time for the supply of the additive target quantity to the processing machine in dependence on the measured, actually provided by the metering device actual amount of the main component is determined.

Thus, for example, the main component can be supplied to the processing machine in a continuous mass flow, and the actual amount of the main component actually supplied within a measuring interval of the processing machine can be determined by forming weight and / or volume differences at discrete, temporally successive measuring times. If the measured actual quantity exceeds a stored threshold value or lies within a stored triggering interval for the main component, the corresponding desired additive quantity is provided and supplied. In this way, it is checked on the basis of a measured actual quantity whether or when the additive target quantity is to be supplied to a continuous mass flow of the main component. A dosing device operating according to such a method is thus almost independent of a downstream processing machine and detects all necessary data for the time of addition of the additive and thus of the mixture to be provided independently.

In this case, it is thus checked, for example, whether the sum of the measured actual quantities has exceeded a stored threshold value since a last supply of the additive, so that the addition of the additive must take place in order to supply the processing machine with a mixture in the predetermined composition.

In principle, the desired additive quantities to be metered can each be calculated directly with the last measured actual quantity of the main component.

However, in order to prevent the additive target quantity from being set anew to a changed measured actual quantity of the main component after each delivery of a specific mixture quantity, it can advantageously be provided that a plurality of further ones are used to determine the additive target quantity of the at least one additive stored first measured values for the actual amounts of the added main component is used.

These further first measured values were stored for previously measured actual quantities of the main component, wherein these previously measured actual quantities were optionally provided on previous demand signals of the processing machine from the metering device in order to supply the processing machine with a mixture having the same composition. Thus, the specification of an additive target quantity and its supply, for example, can only be changed or readjusted if it is ascertained on the basis of successive first measured values for the actually provided actual quantities of the main component that the quantities of the main component, and thus of the quantity provided to the processing machine have been changed (relevant).

For this purpose, provision can be made, in particular, for an average of the first measured values to be used for actually provided actual quantities of the main component for determining the desired additive quantity. Thus, in one exemplary embodiment, the respectively currently supplied or provided actual quantity of the main component can be determined and stored in a list if the dosing device has detected the transmission of a demand signal by the processing machine. By forming the sum of the stored measured values for the actual quantities of the main component over several cycles and dividing this (total) sum by the number of cycles, an average actual quantity of the main component provided can be calculated. This mean actual quantity can be used for the determination or calculation of the additive setpoint.

It is of course not excluded that an average value of the actual quantities of the main component (as well as the mean value of the additive actual amounts explained later) is not determined arithmetically, but via an alternative mathematical method.

In a preferred refinement, erroneous measurements are not taken into account in the calculation of the characteristic curve, which lie outside a predetermined range around a calculated characteristic curve and thus outside of a specific interval by an average calculated over the characteristic curve. A corresponding metering device would therefore be able to obtain a first measured value for a last measured actual quantity which is outside an interval considered to be permissible around the mean value and / or outside the range predetermined range around the calculated characteristic curve, to classify as false measurement.

Alternatively or additionally, in a preferred embodiment of the method, a calculated characteristic curve is determined again as soon as the metering device is refilled with a main component and / or an additive. In this way, a reset of this system, in order to prevent, for example, after a restart of the dosing that is used for a legitimacy check on old and no longer valid measurements.

In order not to allow the arithmetic operations to be performed to increase excessively on the one hand and to detect a permanent change in the metered main component after a relatively short time, a fixed and / or adjustable number of previously stored stored measured values are preferably used to determine a calculated characteristic curve. For example, only the measured quantities of ten or twenty previous measurements are evaluated for the actual quantities.

To ensure that the last measured actual quantity of the provided main component can be used for the above-described (n) calculation method (s) and this is not an erroneous measurement, for example, it is provided in a preferred embodiment of the method that the last stored first The measured value for an actual quantity of the main component is checked for its usability by a stored quality characteristic. Thus, before determining and providing the at least one additive, it is possible to check whether the last measured variable for an actual quantity of the main component is within a predefinable, permissible interval around the mean of the stored first measured values and / or within a predetermined range about a calculated characteristic curve (and analogously to a resulting averaged value or "mean" of the first measured values). Both variants can be used both for a determination of an additive setpoint quantity as a function of a single, last measured actual quantity of the main component and also for a determination of an additive setpoint quantity to be added as a function of a calculated mean value of the actual quantities.

The limits of such an interval or range are preferably specified on the one hand by an average minus the calculated standard deviation for the stored first measured values and on the other hand by the mean value plus the standard deviation for the stored first measured values of the measured actual quantities.

In order to ensure that a permanent change in the processing speed of the processing machine or of the requested mixture quantities is reliably detected, it can furthermore be provided that in the case of a permanent, d. H. A new average and / or a new calculated characteristic curve, excluding previous first measured values, is calculated over several measuring cycles and approximately constant enlargement or reduction of the measured actual quantities relative to an average value.

Alternatively or additionally, it can be provided that the average value is calculated from the measured actual quantities of the main component provided or from the associated first measured values only for a preset number of measuring cycles and thus only from a predetermined number of previously stored first measured values.

In the event that a single first measured value and thus a last measured actual quantity of the main component is not within an interval defined around the mean value and / or within a range defined around the characteristic curve and is therefore not considered permissible, it can continue to be provided, that for the determination of the additive setpoint initially a previous, as measured permissible first measured value is used.

Alternatively, a predefined, fixed desired value or the calculated mean value for the amount of the main component can also be used in order to determine the desired additive quantity on the basis of the prescribed composition or formulation of the mixture to be supplied and then meter it in. Such a fixed desired value can, for example, have been stored manually in a correspondingly designed evaluation unit of the metering device.

In both cases, it is ensured that a single faulty measurement of the actual amount of the main component or a single incorrect dosage, especially at a start of the processing machine, does not lead to a permanent readjustment of the additive setpoint. Rather, a preset additive target quantity can initially be maintained and, for example, reliably determined at a start of the processing machine over several measuring cycles, which actual quantities of the main component are actually requested. After a few measuring cycles, the actually requested mixture quantity or requested actual quantity of the main component is fixed, so that then an automatic adaptation of the additive target quantity takes place.

In this way, any color variations in spraying machines due to changes in the requested actual amount of the main component can be avoided. For example, the composition of the mixture of a main component and at least one additive is decisive for the coloring of a (plastic granulate) mixture to be processed by an injection molding machine. The method (as well as with a metering device and a spraying machine) ensures that a coloration defined by the given mixture composition is maintained even if the mixture quantity requested by the processing machine changes by automatically adjusting the additive target quantity.

Instead of checking on the basis of a mean value, whether a measured actual quantity of the main component (as well as an additive actual quantity explained below) is considered permissible and can be used for the determination (control) of one or more additive target quantities, an alternative mathematical comparison value and / or an alternative mathematical method using previous first measured values for measured actual quantities are used.

In a development of the method according to the invention, an actual amount of additive of the at least one additive actually provided by the metering device is also measured and stored as a second measured value. This not only allows a logging of the actually added additive actual amount, but also a control of the settings for the addition of the additive target amount.

It is thus possible, for example, for a (second) delivery unit of the metering device not to provide a predefined additive setpoint due to fluctuations in the filling weight or density of the additive and for the actually provided additive actual quantity to deviate from the desired additive setpoint.

In one exemplary embodiment, a metering screw is used for metering and feeding the at least one additive, the duration of which or its number of angular steps is correlated directly with the amount of additive provided. For example, in an evaluation unit of the metering device, it is stored how many revolutions of the metering screw or which transit time of the metering screw at a certain speed provides a certain amount of additive and thus supplies it to the mixture to be produced. Consequently, the metering screw is operated at a specific delivery rate, that is to say a specific transit time or a specific rotational speed, as a function of the additive target quantity which was calculated on the basis of the measured actual quantity of the main component provided. If the measured actual quantity of additive deviates (considerably) from the desired additive quantity, then an adaptation of the delivery rate of the metering screw can take place. The delivery rate is thus controlled as a function of the second measured value for the measured additive actual amount.

For this control of the feed rate of the metering screw for the additive, a plurality of second measured values for measured additive actual quantities can also be used and evaluated. In particular, an average value can be determined or calculated from a plurality of preceding second measured values, on the basis of which an adjustment or adjustment of the delivery rate takes place in order to approach the desired additive setpoint.

Likewise, for the determination of an average value of a plurality of second measured values for additive actual quantities, a calculated characteristic curve can also be determined here for example by means of linear regression. Thus, on the basis of a plurality of second measured variables analogous to the embodiments already presented above, for the use of a calculated characteristic curve for first measured values, a reliable check can be made as to whether a new second measured value can be utilized or if it is a faulty measurement. The mathematical determination of an approximation function on the basis of previous second (as well as first) measured values allows a simple filtering of fluctuations in the measured variables.

Furthermore, in this case as well, a quality feature or a quality variable or several quality variables can be stored comparable to the permissible interval for a measured actual quantity of the main component, so that adaptation or adjustment of the delivery rate of the metering screw takes place only when a last measured second measured value or a plurality of measured second measured values exceeds or falls below the quality variable (s) for a desired additive quantity. A last measured actual additive quantity must be z. B. within a predetermined interval for an allowable amount of additive, the limits of this interval or range by the additive target amount +/- a permissible (percentage) deviation can be defined.

Exceeding or falling below a "quality quantity" is then equated with exceeding or falling below a limit or a limit value of this interval.

Of course, it is possible in the context of the method, the metering screw described above for the metering and feeding a Additive replaced by an alternative feed unit. This can be, for example, a vibrating chute, a plate or star feeder or a clocked metering slide.

In connection with a further development of the method according to the invention, a processing speed of the processing machine is regulated as a function of the measured actual quantities of the main component actually provided by the metering device. In particular, this is understood to mean the regulation of a delivery rate of the mixture supplied to the processing machine and / or a variation of the demand signal (eg with respect to time, duration) of the processing machine. Thus, in a spraying machine, for example, a screw speed for a screw conveyor into which the mixture is added via the metering device or a drawing time in which the mixture is to be added can be regulated thereby.

Alternatively or additionally, a processing speed of the processing machine can also be regulated as a function of an average of the stored first measured values for actually provided actual quantities of the main component and / or as a function of a calculated characteristic curve for the stored first measured values. For this purpose, an average value for actual quantities of the main component added over the last cycles is preferably used as the reference value in a control algorithm. About this control algorithm is z. B. specify a screw speed of a screw conveyor of the processing machine. The averaged values, in turn, are preferably determined on the basis of a calculated characteristic curve.

Alternatively or additionally, a processing speed of the processing machine can be automatically regulated as a function of the predetermined via the metering composition of the mixture to be supplied, in particular depending on the proportions of the main component and the at least one additive to the supplied mixture.

Both in the regulation of the processing speed as a function of the actual quantity of the main component and also as a function of the composition of the mixture deposited in the dosing device, the measured values recorded in the dosing device or values stored in the dosing device are coupled, for example the mixture composition, with the downstream processing machine.

Thus, in particular in an injection molding machine as a processing machine, an adjustment of a screw speed of the feed mixture conveying conveyor screw of the spraying machine or the collection time of the injection machine for receiving the supplied mixture in response to be transmitted to the injection machine by the metering control signals. If the mixture is in the form of a granule, for example, bulk density fluctuations can occur which are compensated automatically or independently with the described further development of the method according to the invention, that is to say without manual intervention by a user. Thus, a spraying machine can be controlled via information provided by the metering device. In principle, such a regulation is possible irrespective of whether the metering device and the spraying machine are designed as a structural unit or as separate devices which are subsequently connected to one another and coupled to one another via a data connection.

Particularly in connection with injection molding machines as an example of a processing machine which is coupled to a dosing device operating according to the method, millbases can be processed to a greater extent than before, since the injection molding machine can be measured by measuring the actually metered actual quantities and establishing a data connection of the metering device to the spraying machine can react immediately to any bulk density fluctuations.

Thus, usually only fixed metering times and speeds are set on the spraying machine. In the event of fluctuations in the bulk density or the proportion of millbase or proportion of a component of the mixture to be supplied, as a result of which the total bulk density is changed, less material is conveyed by a conveying unit of the injection molding machine. This leads to rejects. For this reason, it is usually ensured that the proportions of the main component and of the at least one additive or even more additives for the mixture to be produced do not change and mill material are processed to a much lesser extent than is the case, for example, with hollow-body bubbles. With the presented embodiment, however, grinding materials can be processed relatively easily or the proportion of grinding stock in the mixture to be supplied can be increased, which results in cost advantages.

In order to measure the actual quantities of the main component actually provided by the metering device and / or the actual amount of additive actually provided by the metering device, it is preferable to measure a weight and / or a volume of the provided quantity of the main component or of the additive.

• – Bereitstellung der Komponente mittels der Dosiervorrichtung in einer Sollmenge, insbesondere in einer von einer mit der Dosiervorrichtung gekoppelten Verarbeitungsmaschine angeforderten Bedarfsmenge,
• – Messung einer durch die Dosiervorrichtung tatsächlich bereitgestellten Istmenge der Komponente und Speicherung einer Messgröße für diese gemessene Istmenge der Komponente als ein Messwert,
• – Messung weiterer Istmengen nach oder während weiterer Bereitstellungen der Komponente und Speicherung weiterer Messgrößen für diese Istmengen als weitere Messwerte,
• – Bestimmung einer berechneten Verlaufskennlinie, insbesondere mittels linearer Regression, aus den gespeicherten Messwerten,

A further aspect of the present invention is a method for the automatic dosing of a component provided by a dosing device, in particular for the automatic dosing of a main component and / or an additive of a mixture, the method having at least the following steps:

• Provision of the component by means of the metering device in a desired quantity, in particular in a demand quantity requested by a processing machine coupled to the metering device,
• Measuring an actual quantity of the component actually provided by the metering device and storing a measured variable for this measured actual quantity of the component as a measured value,
• Measuring further actual quantities after or during further provision of the component and storing further measured quantities for these actual quantities as further measured values,
• Determination of a calculated characteristic curve, in particular by means of linear regression, from the stored measured values,

Again, the component can be provided both in the form of a continuous mass flow and in portions.

Measurements can be made in particular on a demand signal out or in the case of a continuous mass flow in adjustable temporal, z. B. equidistant distances. The determination of the calculated characteristic curve can, for. B. by an internal or external processing unit that can access the stored readings.

In one exemplary embodiment, a measured variable stored and stored that lies outside a predefined range around the calculated characteristic curve can be classified as an erroneous measurement. In this case, it is possible to use one of the preceding measured and stored measured variables or a previously determined value (mean value) of the calculated characteristic curve.

In a further development of the invention, the measurements which are outside a predetermined range around the calculated characteristic curve (incorrect measurements) can not be taken into account in further calculation of this characteristic. A characteristic that is influenced by incorrect measurements in its course can thus be prevented.

Furthermore, it is possible that for determining the calculated characteristic curve according to statistical methods, such. B. linear regression, only a fixed but adjustable number of previous measured total weight values are used. Thus, a sliding regression of the measured values can take place.

As already explained above, this aspect of the invention can also be integrated in the previously explained method sequence with regard to stored (first) measured values for added actual quantities of a main component as well as stored (second) measured values for added actual quantities of an additive.

• – eine erste Fördereinheit zur Bereitstellung der Hauptkomponente in einer von der Verarbeitungsmaschine angeforderten Bedarfsmenge, wobei die Größe der Bedarfsmenge von einem Bedarfssignal abhängt, das die Verarbeitungsmaschine an die Dosiervorrichtung übermittelt,
• – eine erste Messeinrichtung zur Messung mindestens einer Messgröße für eine tatsächlich durch die Dosiervorrichtung bereitgestellten Istmenge der Hauptkomponente,
• – eine Auswerteeinheit zur Speicherung der Messgröße für die gemessene Istmenge der Hauptkomponente als mindestens ein erster Messwert und zur Bestimmung einer Additiv-Sollmenge für das wenigstens eine Additiv auf Basis einer vorgegebenen Zusammensetzung für das zuzuführende Gemisch aus der Hauptkomponente und dem wenigstens einem Additiv, wobei die Auswerteeinheit zur Bestimmung der Additiv-Sollmenge den ersten Messwert für die tatsächlich bereitgestellte Istmenge der Hauptkomponente auswertet,
• – eine zweite Fördereinheit zur Bereitstellung des Additivs in der von der Auswerteeinheit bestimmten Additiv-Sollmenge und
• – eine Freigabeöffnung, über die eine Zuführung des Gemisches aus bereitgestellter Hauptkomponente und bereitgestelltem Additiv von der Dosiervorrichtung an die Verarbeitungsmaschine erfolgt.

Another aspect of the present invention is a metering device according to claim 24 adapted for carrying out a method for automatically metering and supplying a mixture of a main component and at least one additive in a predeterminable composition. In this case, this metering device is used to supply the mixture provided in the predetermined composition of a mixture processing processing machine. The metering device according to the invention has at least the following:

• A first conveyor unit for providing the main component in a demand quantity requested by the processing machine, wherein the size of the requirement quantity depends on a demand signal which transmits the processing machine to the dosing device,
• A first measuring device for measuring at least one measured variable for an actual quantity of the main component provided by the metering device,
• - An evaluation unit for storing the measured variable for the measured actual amount of the main component as at least a first measured value and for determining a desired additive amount for the at least one additive based on a predetermined composition for the supplied mixture of the main component and the at least one additive, wherein the Evaluating unit for determining the additive setpoint evaluates the first measured value for the actually provided actual amount of the main component,
• - A second conveyor unit for providing the additive in the determined by the evaluation unit additive target amount and
• - A release opening, via which a supply of the mixture of provided main component and provided additive from the metering device to the processing machine takes place.

With such a metering device, in particular, the above-described methods can be carried out, the evaluation unit of the metering device being expanded by appropriate functionalities for the individual variants already explained, and / or a further (second) measuring device being provided for measuring the additive actual quantity. Thus, the evaluation z. B. further for the calculation of the average values of the measured actual amount of the main component or the measured additive actual amount, checking the permissibility of the last measured actual quantities of the main component or the additive actual amount and the generation of a corresponding control signal for the processing machine as a function of measured Actual quantities of the main component and / or the composition of the supplied via the metering mixture be designed.

The below-mentioned advantageous embodiments of the device for carrying out a method for automatic metering and supply of a mixture are thus to be used in an analogous manner as advantageous embodiments of the method and vice versa.

Advantageously, the metering device is further provided and adapted to be coupled with a spraying machine as a processing machine. In particular, the metering device can preferably be designed as a modular unit which is subsequently to be coupled with an existing spraying machine as a processing machine.

In this case, a coupling of the metering device with the processing machine both on the one hand understood the mechanical connection of the metering device with the processing machine, so that the metering device of the processing machine can supply the mixture via the release opening. On the other hand, this also means that the metering device and the injection molding machine are electronically connected to each other via a data connection, so that, for example, not only a demand signal of the processing machine can be transmitted to the metering device, but also possible control signals for controlling a delivery rate of the processing machine from the metering device the processing machine can be transmitted.

Accordingly, another aspect of the present invention is an injection molding machine having a metering device of the type described above according to claim 27.

Further variants of the method according to the invention and the metering device according to the invention are also given by the subclaims.

In addition, other features and advantages of the present invention will become apparent in the following description of exemplary embodiments.

Show it:

1 an embodiment of a metering device according to the invention;

2 a flow diagram of an embodiment of the inventive method for automatic metering and feeding a mixture to a processing machine by means of a metering device;

3 measured total weights plotted against time and an associated calculated characteristic curve.

The inventive method relates to a metering system for at least one component, wherein this component is liquid or a flowable substance, such as. z. B. may be granules. The method uses measured values for calculating a characteristic curve L, which is then used for controlling and regulating the metering of the at least one component. The details for determining the characteristic curve L are mainly in connection with the 2 and 3 described.

Before that, however, the technological background will be explained by means of a metering device. In this case, a plurality of components, namely a main component H and the additives A are used. In principle, this procedure can also be applied to a single component.

In the 1 is shown in a sectional side view of an embodiment of a metering device V. This is constructed symmetrically along a center line M and allows the supply of a mixture consisting of a main component H and at least two additives via a release opening 90 to a processing machine not shown in the form of a (plastic) injection molding machine.

The main component H and the additives to be metered in the present case are granular and thus form a quantity of granules according to a predetermined composition when fed to the downstream processing machine. The composition of the mixture to be added in the metering device V is stored in an electronic evaluation unit (not shown).

In the shown 1 is the release opening 90 for supplying the mixture at a lower end of the metering device V, below a mixer 9 the metering device V. Immediately above the mixer 9 is a weighing container along the center line M. 2 arranged for the main component H. Symmetrical to the center line M are located on the side of the weighing container 2 for the main component an additive weighing containers 5L (left of the weighing container 2 ) respectively. 5R (right of the weighing container 2 ). In these additive weighing containers 5L and 5R In each case an additive is provided, which must be added to the supplied mixture.

To a certain amount of the main component H via the weighing container 2 in which the main component H is kept, the mixer 9 To feed, is an automatically openable or closable inlet 20 between the weigh bin 2 and the mixer 9 intended. As soon as the inlet 20 Released by a gate valve, part of the main component H may be due to gravity in the mixer 9 or pass through it to the processing machine. In this way, form the weighing container 2 and the feed 20 a (first) conveying unit of the metering device V.

Also on the two additive weighing containers 5L and 5R is one each via a gate valve 10L respectively. 10R adjustable inlet at a lower end of the additive weighing container 5L respectively. 5R intended. However, the respective additive does not come directly from the respective additive weighing container 5L . 5R in the mixer 9 but is via an intermediate conveyor unit consisting of a metering motor 7L respectively. 7R and a dosing screw 8L respectively. 8R the mixer 9 fed. At the end of the two dosing screws 8L and 8R is each a feed 80L respectively. 80R provided for the additive over which the amount of additive to be metered into the mixer 9 arrives.

To make it easier to retrofit and clean the dosing device, there are the individual weighing containers 5L and 5R with a quick release 11L respectively. 11R provided so that the additive weighing containers 5L . 5R easy and quick to remove from the metering device V and are to be attached to it again. So with removed additive weighing containers 5L respectively. 5R no impurities in the mixer 9 can reach, the gate valves remain 10L and 10R at the metering device V and close the corresponding feeds to the respective metering screws 8L respectively. 8R ,

To the weighing container 2 always be able to automatically feed with a sufficient amount of the main component H, is above the weighing container 2 furthermore a material conveyor 1 arranged as part of the dosing device V. For this material conveyor 1 can the weigh bin 2 refilled or the main component H refilled. This is a slider 3 below the material conveyor 1 mounted, the automatic and controllable supply of the main component H from in the 1 shown container of the material conveyor 1 in the funnel-shaped weighing container 2 allowed. The main component H is sucked by the material conveyor and stored therein, so that a (re-) filling of the weighing container 2 is possible as soon as a set filling quantity - here in the form of a filling weight - is exceeded.

Furthermore, the dosing device V shown has a first measuring device in the form of a load cell 4 for the weighing container 2 on. For the additive weighing containers 5L and 5R is in each case a second measuring device in the form of additive weighing cells 6L respectively. 6R intended. About the load cells 4 and 6L respectively. 6R in each case the total weight H _tot or A _{ges of} the respective weighing container 2 respectively. 5L . 5R with the mixture component contained therein, the main component H or one of the two additives. After the addition of a certain amount of the respective mixture component can be from the difference between two measured total weights H _ges and A _{ges of} the respective weighing container 2 respectively. 5L . 5R determine which (actual) _is set H or A mixture of the component _is the mixer 9 was actually fed.

Thus it can be provided that, for example, the opening duration of the inlet 20 for the supply of a certain requirement quantity of the main component H depends on a demand signal of the downstream processing machine. This means, for example, that a spraying machine as a downstream processing machine starts the addition of the mixture via a first demand signal S.

In a further embodiment or variant of the method is the release opening 90 permanently open and thus allows during operation of the processing machine, the constant flow of the main component H in the processing machine, for example in a screw conveyor of a spraying machine. If the processing machine was first switched off or temporarily switched off before a first startup, no further material flows via the release opening 90 from the metering V out. Only with a (renewed) start of the processing machine is the main component H or the mixture further processed and new (required) quantities of the main component H flow over the opened release opening 90 to. This additional flow of the main component H from the weighing container 2 is detected by the metering device V and evaluated as a demand signal S of the processing machine.

Also, individual electronic or mechanical demand signals S of the processing machine could cyclically open and close the inlet 20 condition, so that the main component H is not continuously, but in portions over the mixer 9 and its release opening 90 the processing machine is supplied.

In addition, further demand signals S, so-called "shot signals" z. B. in the form of a running signal of a spraying machine screw conveyor, are detected by the evaluation of the metering V and depending on such further demand signals S times for the metering of the two additives in a mass flow of the main component H are determined.

In any case, a demand quantity of the main component H requested from the processing machine becomes out of the weighing container 2 in the mixer 9 arrive and thus provided for the mixture to be supplied.

It should be noted that one from the weighing container 2 initially removed from the demand signal S of the processing machine or injection machine demand quantity of the main component H is not detected. About the load cell 4 Accordingly, no complicated continuous measurement of the total weight of the weighing container 2 to the opening duration of the inlet 20 depending on an amount specifically requested by the spraying machine. Rather, a demand signal of the spraying machine could each only an adjustment of the at the inlet 20 triggered gate valve so that on a (first) demand signal towards the inlet 20 is released and on another (second) demand signal towards the inlet 20 is closed again. Accordingly, in this embodiment variant, the metering device V does not detect a continuous course of a mass flow with a portionwise supply of the main component H in order to determine whether a quantity of the main component H requested by the feed is required 20 has flowed and the inlet must be closed.

In this way, the supplied amount of the main component H, for example, depends only directly on a processing speed of the supplied mixture in the downstream processing machine. Only after completion of the supply of the demanded by the spraying machine demand quantity to the main component H or to another demand signal S for the metering of one or more additives out by the load cell 4 determines how large the requested requirement quantity actually was or which actual quantity of the main component H actually belongs to the mixer 9 supplied and thus provided for the processing machine.

On the basis of this measured actual amount of the main component H, which is stored as a first measured value in the evaluation unit of the metering device V, the mixer 9 to be added desired amount of the respective in the additive weighing containers 5L and 5R determined additives. For this purpose, the actual quantity of the main component H provided in an equation or formula which represents the predefined mixture composition or mixture formulation _{is used} to calculate in each case an additive setpoint quantity A _soll .

On the basis of this now determined by the evaluation or calculated additive target amount (s) are the second conveyor units, consisting of metering 7L respectively. 7R and dosing screw 8L respectively. 8R , the mixer 9 Add additives.

In the illustrated embodiment of the metering device V, the measured total weights H _{ges of} the weighing container 2 stored for the main component H respectively in an electronic list of the evaluation, so that by subtracting the individual total weights H _ges each of the mixer 9 supplied actual quantity H _is the main component H can be calculated.

The individual measurements of the total weight of the weighing container H _ges 2 with the main component H therein, it is preferably exclusively detected discretely in the detection of a demand signal S of a downstream spraying machine to which the main component H is continuously supplied.

In order not to constantly and under certain circumstances incorrectly the additive target amount A _should change and thus, for example, the speed of the metering screws 8L and 8R After every addition of the main component H, an average of the stored first measured values for measured actual quantities H _{ist of} the main component H is calculated in the evaluation unit of the metering device V, so that the additive target quantity A _soll can be calculated on the basis of this mean value H _m ,

Only on the basis of this average weight H _{m of} the added main component H from a plurality of stored total weights H _ges or actual quantities H _is the additive target quantity A _soll calculated, so that the additive target quantity A _should not after each supply of a mixture to the Processing machine needs to be readjusted.

Rather, it is furthermore possible to check whether a last measured actual quantity H _{ist of} the main component H supplied disproportionately deviates from the previously ascertained and actually supplied actual quantities H _{ist of} the main component H. Thus, the standard deviation σ _H can also be calculated in the evaluation unit of the metering device V. This can be an interval around the calculated mean or the mean weight H _m the actual measured quantities H _are +/- the standard deviation σ _{H to be} defined, within which a newly measured actual quantity H _is the main component H, which is the mixer 9 was fed must occur in order for the A further calculation of the quantity of additive desired _to be evaluated by the evaluation unit as permissible. Thus, on the basis of a quality feature stored in the evaluation unit (check whether newly measured actual quantity H _is within a permissible interval or the mean absolute deviation of the last measured actual quantity H _is the main component H from the mean value H _{m of} the previously measured actual quantities H _ist Main component H) checks whether the last measured actual quantity H _is the added main component H, determined by the difference between two measured, successive total weights H _{ges of} the weighing container 2 , to calculate the additive target amount A _{should be} used.

As long as the last measured actual quantity H _{is not} the main component H or its stored first measured value within the permissible interval [H _m - σ _H ; H _m + σ _H ] and thus is not yet released, the evaluation of the metering device V calculates the additive target amount A _soll with a last valid first measured value for the actual amount H _is the added main component H. This last valid first measured value was previously _old stored as a memory value H in the evaluation unit and after each measuring cycle with the last valid first measured value for the measured actual quantity H of the added main component H _is overwritten.

In an alternative or supplementary variant, a calculated characteristic curve is determined by means of linear regression on the basis of a plurality of preceding second measured values to specify a reference or mean value H _m . By determining such a calculated characteristic curve, a plausibility check is less susceptible to error, and the system-related or measurement-related fluctuations are reliably filtered.

For calculation of the characteristic curve can on the one hand, the calculated actual quantities H _is used or on the other hand directly the measured total weights of the weighing container H _ges 2 , In both cases, these are stored measured variables which represent the actually metered quantity of the main component H.

In the first-mentioned case - the calculation of a characteristic curve based on the calculated and actual quantities stored over time _is H - would a linear approximation function in H _is -over-t - chart substantially as an almost horizontal line represent. Thus, the actually metered actual quantity H _{ist of} the main component H ideally should remain permanently constant.

To contrast, in each case to be able to leave out an arithmetic operation, preferably immediately the measured total weights of the weighing container H _ges 2 used as measured values for calculating a characteristic curve, on the basis of which a plausibility check is carried out. If an approximation function for the total weights H _{tot is} calculated as a function of the time t by means of linear regression, then a straight-line which slopes over time will result as the characteristic curve, since the total weight of the weighing container 2 with increasing metering of the main component H in the mixing container 9 diminished (see also 3 ). The slope of this idealizing characteristic curve thus represents the actual quantity H _, measured on average _{, of} the main component H. As already mentioned above, this mean value can be used both to define an interval considered permissible, in which a newly measured first measured value must lie Calculation of the additive target quantity A _{to be added should} be used.

It can be seen that thus in the present metering device V as well as in the thus carried out and based on the 2 and 3 explained in more detail embodiments of the method is a manual specification of the processing machine to be added individual mixture components is not necessary, but on the contrary only the specification of the desired mixture composition is sufficient. The processing machine demands on the metering V the desired required amount of mixture or so that the main component H, which metering device V then provides at the time of the request by the processing engine, the actually supplied actual quantity H _is the main component H and determines exclusively in function of the measured actual quantity H _is the main component H provided the addition of the necessary additive target amount A _soll .

Even without prior, detailed, manual adjustment, the dosing device V _can already detect the amounts of the main component H required by the spraying machine via the calculation of the average values H _{m of} these actual quantities H after only a few measuring cycles. Not only can the material consumption of the main component H be recorded, but also an automatic adaptation of the additive target quantity A _should only take place if it has been ruled out that the actual quantity H _is at the last measurement or if the actual quantity H has been measured _is no faulty measurements and the requested quantity of the main component H has been changed permanently.

As already explained above, the additives of the metering device V are also gravimetric dosed. These are the additive weighing containers 5L and 5R on the additive load cells 6L respectively. 6R assembled. Since the dosing device V shown the 1 preferably mounted on a spraying machine is to be expected with mechanical shocks and vibrations. That's why the additive weighing containers are 5L and 5R on separate additive load cells 6L respectively. 6R assembled. Furthermore, it is provided for this reason that a focus of the respective additive weighing container 5L and 5R below the associated additive load cell 6L respectively. 6R located.

By means of additive weighing cells 6L and 6R , which in each case form a second measuring device of the metering device V, which can the mixer 9 actually supplied additive actual amount A _{is to be} measured and optionally an adjustment of the control of the associated metering screw 8L respectively. 8R respectively.

This is how the dosing screw works 8L . 8R through the associated dosing motor 7L respectively. 7R driven and dosed the respective additive into the mixer 9 , A running time or a number of angular steps of the respective metering motor 7L . 7R is measured by electrical pulses. This is done z. B. with a Hall sensor in the dosing motor 7L respectively. 7R , To determine a desired delivery rate R _{A of} the metering screw 8L . 8R is in the evaluation of the metering V preferably an internal factor deposited. This internal factor indicates how much weight of an additive via the respective metering screw 8L respectively. 8R at a defined running time of the associated metering motor 7L respectively. 7R the mixer 9 is supplied. Typically, this internal factor is set in grams per second (GpS) or grams per pulse (GpP).

By determining the actually supplied additive actual amounts A _is by means of the formation of weight differences from the measured total weights A _{ges of} the additive weighing container 5L . 5R can be controlled whether the stored factors for the specification of the additive target amount A _{should be} correct and thus the desired additive target amounts A _soll can be achieved. It is thus evaluated the extent to which the actual metered additive actual quantities A match _is (on average) _will with the calculated additive A target quantities.

For this purpose, may also be provided that an average value A _{m, is} about the respective stored as a second measurement values in the evaluation unit measures of actual quantities additive A _is formed. In this way it can be determined to what extent a last measured actual quantity of additive A _can be assessed as permissible, or whether it could be a faulty measurement.

Likewise, it can be determined whether a last stored second measured value for an additive actual quantity A _is repeatedly outside an admissible interval about the calculated average value A _m, and therefore an incorrect setting of the delivery rate R _{A of} a second delivery unit for the additive is to be assumed, each by the associated metering motor 7L respectively. 7R and the dosing screw 8L respectively. 8R is formed. The delivery rate R _A can then be increased or decreased accordingly to the targeted additive target amount A _soll to the mixer 9 supply and thus provide for forwarding to the spraying machine.

Here, too, the calculation of a characteristic curve, in particular by means of linear regression, is possible in order to allow a check of the actually supplied Additity actual quantity and a check of the set delivery rate R _A for a certain additive setpoint A _soll .

With the 2 is the above already in connection with the metering V of 1 explained process flow shown in a flowchart. In this case, individual process steps are designated A.1-A.13.2, B1.1-6.2 and C.1, wherein the steps A.5-A.8.2 and A.10-A.13.2 are implemented by the evaluation unit of the metering device V. ,

First of all, the filling of the weighing container takes place 2 in a step A.1 on the material conveyor 1 and the slider 3 , This is followed by a step A.2 the measurement of the total weight H _{ges of} the weighing container 2 with filled main component H. The measured total weight H _ges can then be stored, for example, as an initial value for the further determination of subsequent actual quantities H _ist on supplied main component H by means of a differential weighing method in the evaluation unit.

During operation of the injection molding machine supplied with mixture by the metering device V as an example of a processing machine, the demand signal S of a screw conveyor of the injection molding machine is transmitted to the metering device V in a step B.1. The metering device V then sets, for example, in a step A.3 by opening and closing the inlet 20 on the basis of the demand signal S an initially unknown requirement quantity of the main component H in portions ready and leads them to the mixer 9 to (step B.2 of 2 ). Alternatively, a demand signal S may be a so-called shot signal via which the dosing device V is informed of the point in time at which the addition of the at least one additive or the simultaneous addition of several additives must take place.

As already mentioned above in connection with the 1 explained, can also the reflow of the main component H from the weighing container 2 are detected in the previously inactive processing machine by the dosing device V and evaluated as a demand signal S of the processing machine when the processing machine (re) starts to run.

In response to a demand signal S of the screw conveyor of the spraying machine is towards present in each case in a step A.4 (again) the total weight of the weighing container H _ges 2 measured. This is also understood that the demand signal S can trigger a repeated, time-dependent measurement of the total weight H _ges .

With a subsequent method step A.5, H _{ges of} the weighing container is then determined on the basis of consecutively measured total weights H _ges 2 which actually the mixer 9 supplied Istmenge / weights H _is the main component H determined and stored by means of the evaluation in a list.

In step A.6 the 2 is now by the evaluation unit an average weight H _m of the added actual quantities H _is based on a plurality of stored actual quantities H _is determined and checked in a process step A.7 whether the last measured and stored first measured value for the actual quantity H _is within an allowable Interval lies. The limits of this interval are defined by the mean value H _m and the standard deviation of the stored first measured values for the actual quantities H _ist . Mathematically, the interval to be checked [H _m - σ _H ; H _m + σ _H ].

Alternatively or additionally, in a method step A.7, the comparison of the actual quantities H _ist with a calculated characteristic curve or an average value resulting therefrom can be effected. For this purpose, it is checked, for example, whether a last stored measured variable for the actual quantity H _is within a range considered to be permissible around the characteristic curve. Both first measured values can be used in which measured total weights of the weighing container H _ges 2 (see. 3 ), or first measured values in which actual quantities H _are stored.

Depending on the result of the check in method step A.7, the evaluation unit executes either method step A.8.1 or method step A.8.2. If it is determined that the last stored first measured value for the actual quantity H _{ist of} the main component H lies within the permissible interval, the designation of an additive target quantity A _soll to _{be added} continues _to be carried out as a function of the currently calculated mean value H _{m of} the actual quantities H _{ist of} the main component H . Since such a mean value H _m subject after a running time of the molding machine and error-free operation thereof hardly larger fluctuations, which keeps the desired amount of additive a _will initially obtained substantially.

Furthermore, in method step A.8.1, a memory value H _{old is occupied} with the first measured value found to be permissible for the last measured actual quantity H _ist . This memory value H _old is then available to the evaluation unit in method step A.8.2, if it is determined that the last measured actual quantity H _is the main component H belonging to the mixer 9 was fed outside the allowable interval. The calculation of the additive target quantity A _{to be} supplied on the basis of the mixture composition or formulation stored in the evaluation unit now takes place on the basis of the last memory value H _{alt which was} found to be permissible.

Alternatively, it may be provided that in the method step A.8.1 determining the additive target amount A _should always _is directly related to the actual quantity H deemed acceptable is correlated and / or in process step A.8.2 only when found to be inadmissible actual quantity H _is the from the weighing container 2 supplied main component H, the additive target amount A _{should be determined} with the calculated average value H _m .

Also, the memory value H _{alt may be} a manually stored fixed value, which _{is used} to determine an additive target quantity A _soll only in the case of an actual quantity H _being evaluated as inadmissible.

In a further embodiment, it can also be provided that only a (first) demand signal S from the processing machine triggers the start or operation of the dosing device V and then this up to a (second) demand signal S or triggered by a user "STOP" Signal works completely automatically. In this case also the above-explained operation can be carried out in which the metering V based discrete measured actual quantities H _is determined whether or when an additive or more additives to the mixer 9 is or are to be supplied.

In the in the 2 illustrated embodiment of the method will now be in a step A.9 continue to measure the total weight A _{ges of} the additive weighing container 5L respectively. 5R performed to determine what actual additive amount A _is from the respective additive weighing container 5L . 5R the mixer 9 is actually supplied.

The actual addition of the respective additive via the respective metering screw 8L . 8R is preferably carried out after the measurement of the total weight A _ges . So can already in advance of the addition of the respective additive into the mixer 9 be determined by the evaluation, to what extent the last-added additive actual quantity A _has coincided with the desired additive target amount A _should .

The actual quantity actually supplied to additive-A _is determined in a step A.10 after measuring the total weight of A _ges or calculated by this total weight of A _ges of a previous measured total weight of A _ges is subtracted.

In a further method step A.11, the evaluation unit calculates an average additive actual quantity or average additive weight A _{m, which is} from the stored additive actual amounts A _ist . Thus, it is also possible here in a subsequent method step, not shown, to check whether the last stored second measured value for a measured additive actual quantity A _can be assessed as permissible. For this purpose, the respective second measured value or the associated measured additive actual quantity A _{must be} within an interval, which is also preferred here by the mean value, here A _m , and the associated standard deviation of the measured additive actual amounts A _is defined.

Unless the predetermined additive setpoint set A _{is to} been found in such admissible, an evaluation of the actually dosed additive actual quantity A _is or a plurality of measured additive actual quantities A _is made as to whether it _{is to} accurately enough with the desired additive target amount A has agreed or have. For this an interval [A _soll - x _{A, 1} ; A _is x + _{A, 2]} the additive target amount A _is set within which the last measured actual quantity _is additive A or - as in a process step of the A.12 2 - the mean value A _{m, is} the last measured, z. B. ten, additive-A _is actual quantities must be.

If additive actual quantity A _is or average value A _m, the actual quantity of additive A _is within the interval [A _soll -x _{A, 1} ; A _should + x _{A, 2} ], is in a in the 2 illustrated process step A.13.1 the setting of the metering screw 8L . 8R or of the dosing motor driving it 7L . 7R as appropriate.

If, on the other hand, it is determined that the last measured actual amounts of additive A _are permanently outside the interval considered admissible, then one in the 2 executed with A.13.2 process step, after which an adjustment of the delivery rate R _{A of} the metering screw 8L . 8R becomes necessary to achieve the desired additive set amount A _soll . Here, the delivery rate R _A as a function of the calculated average value A _{m, is} determined, from which it now emerges for the evaluation, which additive actual amounts A _is by the previously set delivery rate of dosing 8L . 8R actually achieved.

Instead of the method step explained above, a comparison of the last measured actual additive quantity A _{is also possible} with a calculated calculated characteristic curve for stored second measured values (cf. 3 ) respectively.

Are as measured variables that represent the actual added additive actual amounts A _is , for. If, for example, the total weights A _{ges are} stored directly as second measured values, it is also possible on the basis of these values to determine a calculated characteristic curve directly by means of linear regression, with which a check of the delivery rate R _{A is} possible.

In the in the 2 The method shown in the method steps A.11, A.12, and A.13.1 or A.13.2 the calculated average A _m, the additive actual amounts A _is used to check whether a set delivery rate R _{A of} the respective dosing screw 8L . 8R the desired additive target amount A _should deliver. In each case, a fixed percentage deviation x _{A, 1} , x _{A, 2} of the desired additive target quantity A _{soll is} preferably tolerated. Due to the dependence on the calculated mean value A _m, an adjustment of the delivery rate A _{A is} only performed if the measured additive actual amounts A _{are not} over a longer period of time, that is, over several measurement cycles, with the desired additive target quantity A _soll match or lie in the vicinity. Alternatively, however, it can also be provided that it _is checked directly on the basis of a measured additive actual quantity A, whether this value stored as the second measured value is within the interval [A _soll -x _{A, 1} ; A _{should be} + x _{A, 2} ]. Thus, an adjustment of the delivery rate R _A already take place when the measured additive actual amount A _is once deviates from the desired additive target amount A _soll (considerably).

It should also be noted that the limits of the interval considered admissible [A _soll - x _{A, 1} ; A _should + x _{A, 2} ] for an additive actual quantity A _is both symmetrical and asymmetric by the value for the additive setpoint A _{should be} selected. _Is a "symmetrical" choice of the allowable interval for an additive actual quantity A would accordingly [A _is - x _A; A _is x + _A] loud, so that here the additive from the desired amount of A _{is to} be subtracted or hinzuzuzählende amount x _A as permitted considered deviation from the desired amount of additive A _should in each case the same.

Instead of the speed of the dosing motor 7L respectively. 7R To change, this can also be operated at full speed. The distribution of a metered addition of the additive over the collection time of the processing machine is then by pulsing the metering motor 7L respectively. 7R reached. This makes it possible to evaluate the speed control on the respective metering motor 7L respectively. 7R save on.

As is clear from the above statements, not only a material consumption of the main component H can be detected and logged with the described method and the metering device V, but also a change in the demanded by the processing machine demand quantity is automatically detected and the amount of additive to be added automatically adjusted.

Especially with regard to a spraying machine as an example of a processing machine supplied with a mixture by the metering device V, such color fluctuations due to changes in the required quantity are avoided. Such color variations result from the fact that the (granular) mixture of main component and additive or additives leads to a processed (plastic) product whose color depends directly on the deposited mixture composition. If the mixture composition is not maintained due to a change in the required amount of main component H or a total amount of the mixture to be supplied, for example, because the amount of additive supplied is not correct, color variations occur over a production cycle of the injection machine on the final products produced. The presented automatic measurement and adaptation of the quantities of the mixture components (in the present case a main component H and two additives) to be provided by the metering device V can reliably prevent such color fluctuations even in fully automated operation.

Furthermore, as already stated above, the addition of the additive into the mixer 9 as a function of the demand signal S or another demand signal S transmitted by the processing machine. In this case, the start and stop of the respective associated metering screw 8L respectively. 8R depend on such a demand signal S of the processing machine. Alternatively, of course, it is also possible that, in response to a single demand signal S from the processing machine, the addition of the additive into the mixer 9 takes place and the duration of the dosing screw 8L . 8R On the basis of the measured actual quantity H _, the supplied main component H is determined. Depending on how large _the supplied actual quantity H is or was, then, based on the predetermined mixture composition or -rezeptur the desired additive target amount A _is added over a certain time.

A development of the metering device V can also provide that the individual additive weighing container 5L . 5R can also be fed automatically with additive with one or more material conveyors.

Furthermore, it can be provided that a processing speed of the downstream processing machine _is adjusted or regulated as a function of the measured actual quantities H _{ist of} the supplied main component H. This is understood to mean in particular an adjustment of the screw speed of a screw conveyor of an injection molding machine and / or the pull-in time of such a spraying machine. In this way, any bulk density fluctuations of the main component H as well as the additives can be automatically corrected. The same applies to a possibly provided adjustment as a function of set on the metering V mixture composition or formulation.

Thus, by coupling a metering device with a spraying machine or generally a processing machine via a data line or data connection possible that by measuring the proportions of the individual mixture components, a supplied with mixture spraying machine (processing machine) compensates bulk density fluctuations independently by feeding more material.

Moreover, by detecting the consumption of the main component H, it is possible to specify an order quantity. By correlating the actual quantities consumed _is H can be controlled to the supply of the mixture or of the main component H over a measured feed time, that a job end no more material in the material conveyor 1 and in the weighing container 2 is present. This shortens any changeover times and reduces material loss by spreading granules, provided that the mixture to be supplied is in the form of a granulate mixture.

It should be noted that with the metering device V and the method of 2 was turned off exclusively to a measurement of the weight to an actual amount H _ist or an additive actual amount A _is to determine. However, it may of course also a volumetric sensing the actually supplied actual quantity H _is the main component H and / or the additive actual quantity A _is the actually supplied amount of additive can be used.

The 3 illustrates a further aspect of the present invention, in which by means of statistical methods, a calculated course characteristic L is determined from measured values and used for plausibility check for new measured values.

Among other things is in the 3 the determination of a mean actual quantity H _m that the mixer 9 was actually supplied on average, shown and seen as a slope of the calculated characteristic curve L. In the graph of this are measuring points for total weights H _ges of the weighing container 2 applied at different times over time t. These total weights H _{tot come} from measurements taken one after the other and are subject to certain fluctuations.

Following a first filling of the weighing container 2 , the measurement of the total weight of the weighing container H _ges 2 with then filled main component H (H ₀ ). The recording of further measured values of the total weight H _ges (H _n ) can be done after each opening and closing of the inlet 20 due to a demand signal S or continuously after a single opening of the inlet 20 (continuous mass flow) in any, in particular equidistant time steps carried out. The measured values for the total weight of the weighing container H _ges 2 are each stored as a first reading. From these measured values then a calculated characteristic curve (continuous line L in the graph) by means of statistical methods, such. B. by linear regression determined. Subsequently, a determination of the gradient of the characteristic curve L can take place. This gives the average actual quantity H _m actually metered on average, that of the mixer 9 was fed to. The value H _m can be used in an equation or formula which represents the predefined mixture composition or mixture formulation _{, in order} to calculate in each case a desired additive quantity A _soll .

Alternatively or additionally, a new total weight value H_ges To check whether it lies in a predetermined range around the calculated characteristic curve L. If this is the case, then from this total weight value H_ges and a previous total weight value H_ges  an actual quantity H_is are calculated, which are used in an equation or formula representing the predefined mixture composition or mixture formulation, in each case a desired additive amount A_should to calculate. For incorrect measurements that are not within the specified range, one of the preceding weight values H_is or the value for the mean or average actual quantity H_m be resorted to.

Furthermore, in a further developed method, incorrect measurements that lie outside a predetermined range around the calculated characteristic curve L can not be taken into account in further calculation of this characteristic curve.

In a particular modification of the method, the determination of the characteristic curve L is carried out in a sliding manner, ie, only a fixed but adjustable number of preceding total weight values H _{ges is} used to calculate the characteristic curve L.

In a further embodiment, the method for calculating the characteristic curve L restarts as soon as the weighing container 2 is refilled.

In a further embodiment based on a calculated gradient characteristic, an average measurement value for the ausdosierte additive component from the weight values A _tot may alternatively or additionally be determined. In the 3 are not only measured total weights H _{ges of} the weighing container for illustrative purposes 2 but also measured total weights A _{ges of} an additive weighing container 5L . 5R applied over time t.

Following a filling of the weighing container 2 and a first addition of an additive from one of the additive weighing containers 5L . 5R , the measurement of the total weight of A _ges and the additive weighing container 5L . 5R with then filled additive (A ₀ ). The recording of further measured values of the total weight A _ges (A _n ) can then analogously after each addition of the additive in the mixer 9 take place in any, in particular equidistant time steps.

That in the 3 illustrated and already previously explained several times dosing and measuring method, in which on the basis of measured variables, the actual metered actual amount of a component, eg. As a main component H or an additive to represent, by means of statistical methods, a calculated characteristic curve is determined, offers significant advantages over z. B. purely arithmetic calculations of a (mean) actual quantity.

In the case of an arithmetic determination of an actual quantity metered in on average, for example, several total weights H _{ges of} the emptying weighing container are successively determined 2 measured. By subtracting two successive measured values, it is possible to deduce a metered actual quantity in a specific time. If, however, the sum of the actual quantities H thus calculated _is only divided by the number of measurements made, the resulting average value does not reveal any possible scattering between the individual actual quantities. Also "outliers" (ie strongly deviating from the other measured values individual results) are equally weighted in the calculation of the average and increase or decrease this seriously, so he z. B. is undesirable error-prone for a reliable determination of an additive target amount.

In contrast, by determining a calculated characteristic curve for measured actual quantities, in particular by means of linear regression, short-term or slight fluctuations in the measured values are reliably filtered out. An average resulting from this characteristic curve, z. B. as a slope of this characteristic curve (see characteristic curve L of 3 ), thus provides a reliable data source for subsequent calculations, such. B. the calculation of a desired additive amount A _soll , the plausibility check for subsequent measurements of an actual amount H _is a main component H and / or an additive or the specification of a delivery rate R _A for a metering device V downstream processing machine.

The method using a characteristic curve L can be used both in quantity increase and in quantity decrease. With an increase in quantity, the calculated characteristic curve L can be used to determine the remaining dosing time until a desired final weight is reached. It can also be monitored when filling containers, whether the filling speed is within a permissible range. This is useful when filling silos to avoid abrasion and angel hair.

• 1. Zunächst wird ein Startwert für die Förderrate in Gr./Sek. (GpS-Wert) bzw. Gr./Sek./Dz (GpSD-Wert) festgelegt, wobei die Drehzahl (bzw. eine anderweitige Vorschubgröße) Dz = 100% = 1 ist. Dies kann durch Eingabe empirischer Daten in einer Bildschirmmaske erfolgen oder durch eine Testdosierung.
• 2. Dann wird eine Anzahl an Prozesszyklen (Prozess-Zyklenanzahl) festgelegt, die zusammen eine Charge ergeben. Sinnvoll können dabei 4 bis 20 Prozesszyklen sein.
• 3. Warten auf das Start-Signal für einen Prozesszyklus.
• 4. Festlegen der Rezept-Soll-Menge aus der Benutzereingabe von Schlussgewicht und Farbanteil, wenn Farbadditive verwendet werden.
• 5. Erstes Messen des Gewichtes und Speicherung in einem FiFo-Speicher.
• 6. Wenn die Anzahl der Prozesszyklen die Chargen-Zyklenanzahl erreicht hat, dann
• 7. Berechnen einer Kennlinie aus dem im FiFo-Speicher vorhandenen Werten und Ableiten der ausdosierten Ist-Menge. In diesem Fall wird eine lineare Regression durchgeführt, um eine Regressionsgerade zu erhalten. Die Ist-Menge ergibt sich unmittelbar aus der Steigung der Regressionsgeraden.
• 8. Durchführung eines Soll-Ist-Vergleichs und Errechnen einer fortlaufenden Prozess-Sollmenge.
• 9. Errechnen einer Dosierzeit für das Dosierorgan aus der Prozess-Sollmenge.
• 10. Löschen des FiFo-Speichers und Rücksetzen des Prozesszyklus-Zählers.
• 11. Ausdosieren des Materials für die Dauer der festgelegten Dosierzeit.
• 12. Weiter mit Warten auf Start-Signal.

In the following, an embodiment of a dosage will be indicated by means of a cyclical process (eg injection molding process). In this case, a dosage with a constant conveying speed.

• 1. First, a starting value for the delivery rate in Gr./Sek. (GpS value) or Gr./Sek./Dz (GpSD value), whereby the rotational speed (or another feed amount) Dz = 100% = 1. This can be done by entering empirical data in a screen mask or by a test dosing.
• 2. Then a number of process cycles (process cycle number) are determined, which together result in a batch. 4 to 20 process cycles can make sense here.
• 3. Waiting for the start signal for a process cycle.
• 4. Define the recipe target quantity from the user input of final weight and color content when using color additives.
• 5. First measuring the weight and storing in a FiFo memory.
• 6. If the number of process cycles has reached the batch cycle number, then
• 7. Calculate a characteristic curve from the values available in FiFo memory and derive the metered actual quantity. In this case, a linear regression is performed to obtain a regression line. The actual quantity results directly from the slope of the regression line.
• 8. Carrying out a target / actual comparison and calculating a continuous process target quantity.
• 9. Calculate a dosing time for the dosing device from the process setpoint.
• 10. Clear the FiFo memory and reset the process cycle counter.
• 11. Dosing the material for the duration of the specified dosing time.
• 12. Continue with waiting for start signal.

The inventive method is thus integrated into this exemplary process flow. By combining several process cycles into one batch, actual quantities can be measured with a higher accuracy than the measuring equipment allows for individual measurements.

In step 11 a fixed dosing time is used. In a modification of this procedure, it is possible to subdivide the desired dosing into several time steps, which are then distributed in the period between two START signals. This allows a uniform discharge of material to be achieved, even if a drive with a constant conveying speed (eg metering screw with a non-speed-adjustable motor) is used.

In the example given above, the running time of the drive is used as a manipulated variable for the dosage. Alternatively, other parameters, such. B. the angular steps of a rotary drive with attached metering screw can be used. It is also possible to use the number of strokes of a linear plunger or a Vibroantriebes or the chamber-wise discharging Sterndosierem or metering as a manipulated variable.

The following describes the sequence of a process for a cyclic process (as above) using a metering device or an adjustable speed drive.

In a first variant, the dosing time or the delivery units are used as manipulated variable. Basically, the control is carried out as described above in the 12 steps. However, in this variant, the speed of the drive is added as a manipulated variable. The speed of the drive is adjusted so that the dosing spreads as evenly as possible on the cycle time minus an adjustment.

The speed is kept constant during control and only adjusted when the cycle time or the recipe setpoint changes. The variations of the manipulated variable "dosing time" are within the adjustment range.

This first variant makes sense in metering devices in which the delivery rate varies with the drive speed. This is the case with most dosing screws due to friction. The adjustment of the speed increases the possible control range and causes a better distribution of the metering over the cycle time.

In a second variant, the drive speed is used as a manipulated variable. The dosing time is kept constant and with each new batch a new drive speed is calculated as a control variable of the control algorithm.

In cyclic processes, the first variant may be advantageous because the dosing time can be finer resolved than the desired speed of a drive. Also, the speed-dependent friction can cause different changes in the delivery rate and affect the quality control.

In a continuous process, such. As the extrusion is usually desired, a continuous material discharge. The dosing time can not be used as a manipulated variable, as there are time gaps in the dosed material flow.

In the continuous process, the dosing time is set to the full cycle time and the drive speed (as in the second variant above) is used as the manipulated variable. The continuous mass flow is divided into individual batches within the control, the cycle time being equal to the dosing time.

The selected cycle time depends advantageously on the delivery behavior of the dosing, the desired mass flow, the desired control range and the control behavior.

For the display, the actual quantity is converted into a throughput value.

LIST OF REFERENCE NUMBERS

1

Materialförderer

10L, 10R

Gate valves

11L, 11R

quick release

2

Weighing container (main component)

20

Inlet (main component)

3

pusher

4

Load cell (main component)

5L, 5R

Additive weighing container

6L, 6R

Additive weighing cell

7L, 7R

metering

80L, 80R

Feed (additive)

8L, 8R

dosing screw

9

mixer

90

release access

A _is

Additive actual quantity

A ₀ , A _n , H ₀ , H _n

measuring point

A _ges

Total weight of the additive weighing container

A _{m, is}

Mean additive actual quantity

A _should

Additive target amount

H

main component

H _ges

Total weight of the weighing container

H _is

Actual quantity (main component)

H _m

Mean actual quantity (main component)

L

Running characteristic

M

center line

R _A

delivery rate

S

demand signal

V

metering

x _A , x _{A, 1} , x _{A, 2}

permissible deviation

σ _H

standard deviation

Claims (27)

Method for the automatic metering of at least one component (H) provided by a metering device (V), the method comprising at least the following steps: - providing the at least one component (H) by means of the metering device (V) in a desired quantity, - measuring a by the metering device (V) actually provided first actual quantity (H _ist ) of the at least one component (H) and storing a measured variable for this measured first actual quantity (H _ist ) of the at least one component (H) as a measured value, - measuring at least a second Actual quantity (H _ist ) after or during further provision of the at least one component (H) and storage of further measured quantities for these second actual quantities (H _ist ) as further measured values, - calculation of a characteristic curve (L) from the stored measured values and control of the dosage depending on from the characteristic curve (L), characterized in that the course characteristic curve (L) for which at least one component (H) is determined as a function of the container shape for the at least one component (H). A method according to claim 1, characterized in that for calculating the characteristic curve (L) a fixed and / or adjustable number of previous stored measured values are used. A method according to claim 1 or 2, characterized in that the calculated characteristic curve (L) for the at least one component (H) is determined in the form of a quadratic dependence in containers with conical sections. Method according to at least one of the preceding claims, characterized in that the calculated characteristic curve (L) of the actual quantities (H _ist ) _is stored from a plurality of stored first measured values for actual quantities of at least one main component (H _ist ) and / or the characteristic curve from a plurality of stored ones second measured values for actual quantities (A _ist ) of at least one additive is determined. A method according to claim 4, characterized in that the calculated characteristic curve (L) of the actual quantities (H _ist ) from a plurality of stored first measured values for actual quantities of at least one main component (H _ist ) by means of linear regression or non-linear regression and / or the characteristic curve from a plurality of stored second measured values for actual quantities (A _ist ) of at least one additive is determined by means of linear regression. Method according to at least one of the preceding claims, characterized in that the at least one component (H) is a flowable substance. Method according to at least one of the preceding claims, characterized in that the model of the calculated characteristic curve (L) is coupled to a model for disturbance variables. Method according to at least one of the preceding claims, characterized in that control parameters of a dosage are obtained by means of at least one test dosage, which form output values for further dosage. Method according to at least one of the preceding claims, wherein the method comprises at least the following steps: - Provision of a main component (H) by means of the dosing device (V) in a demand quantity requested by the processing machine, the size of the requirement quantity depending on a demand signal (S) , which transmits the processing machine to the metering device (V), - measuring an actual quantity (H _ist ) of the main component (H) provided by the metering device (V) and storing at least one measured quantity for this measured actual quantity (H _ist ) of the main component (H ) as at least one first measured value, - determination of a desired additive quantity (A _soll ) for at least one additive based on a predetermined composition for the mixture to be supplied from the main component (H) and the at least one additive, wherein for determining the additive target quantity ( A _soll ) at least one of the first measured values for the Actual provided actual quantity (H _ist ) of the main component (H) is used, - Provision of the additive by means of the metering device (V) in the additive set amount (A _soll ) and - supply of the mixture of provided main component (H) and provided additive to the processing machine. Method according to at least one of the preceding claims, characterized in that the requirement quantity of the main component (H) is provided in the form of individual portions of the main component (H) or in the form of a continuous mass flow of the main component (H). A method according to claim 10, characterized in that the continuous mass flow of the main component (H) by a single demand signal (S) of the processing machine is triggered. Method according to at least one of the preceding claims, characterized in that a point in time for the provision of the desired additive quantity (A _soll ) is likewise transmitted by the demand signal (S) or by another demand signal (S) transmitted by the processing machine to the metering device (V). depends. Method according to at least one of the preceding claims, characterized in that a time for the provision of the additive setpoint quantity (A _soll ) and / or a time duration for the supply of the additive setpoint quantity (A _soll ) to the processing machine as a function of the measured, actually provided by the metering device (V) actual amount (H _ist ) of the main component (H) is determined. A method according to claim 13, characterized in that, if the main component (H) is supplied in the form of a continuous mass flow of the processing machine, based on a measured actual amount (H _ist ) is checked whether the additive target amount (A _soll ) should be provided. Method according to at least one of the preceding claims, characterized by measuring an actually provided by the metering device (V) additive actual amount (A _is ) of the at least one additive and storage of this additive actual quantity (A _is ) as a second measured value. Method according to at least one of the preceding claims, characterized in that erroneous measurements that are outside a predetermined range around the characteristic curve (L) are not taken into account in calculating the characteristic curve (L) and / or a characteristic curve (L) is determined again as soon as a refilling of the metering device (V) with a main component (H) and / or an additive takes place. Method according to at least one of the preceding claims, characterized in that a plurality of further stored first measured values is used to determine the additive setpoint (A _soll ) of the at least one additive, the further first measured values for previously measured actual quantities (H _ist ) being the Main component (H) and these previously measured actual quantities (H _ist ) were provided by the metering device (V) to supply the processing machine with a mixture having the same composition. Method according to at least one of the preceding claims, characterized in that it is checked before the determination of the additive target quantity (A _soll ) and the provision of the additive in the additive target quantity (A _soll ), whether the measured variable for a last measured actual quantity (H _is ) of the main component (H) in a predefinable permissible interval (H _m -σ _H , H _m + σ _H ) about an average value (H _m ) of the stored first measured values for actually provided actual quantities (H _ist ) of the main component (H) and / or in a predetermined range around a calculated characteristic curve (L) of the stored first measured values for the actual quantities (H _ist ) of the main component (H). Method according to at least one of the preceding claims, characterized in that a first measured value for a last measured actual quantity H _is outside the permissible interval (H _m -σ _H , H _m + σ _H ) about the mean value (H _m ) and / or may be located outside of the predetermined range by the calculated curve characteristic (L) is, classified as a false measurement and for determining the additive target amount (a _soll) for the supplied mixture, a predefined, fixed target value for the amount of the main component (H), (the average H _m ) or a first stored measured value (H _alt ) is used for an actual quantity (H _ist ) of the main component (H) which is within the predeterminable interval ([H _m -σ _H ; H _m + σ _H ]) about the mean value (H _m ) and / or within the predetermined range around the calculated characteristic curve (L) was. Method according to at least one of the preceding claims, characterized in that a plurality of further stored second measured values for respectively measured additive actual amounts (A _ist ) is used for controlling the additive target quantity (A _soll ), wherein the further second measured values are for before measured additive actual amounts (A _ist ) of the at least one additive are stored and these actual amounts of additive (A _ist ) were provided for previous preparations of a mixture to be supplied of the same composition. A method according to claim 20, characterized in that an average value (A _{m, is} ) of these second measured values and / or a calculated characteristic curve of these second measured values is used. Method according to at least one of the preceding claims, characterized in that as a function of the measured actual quantity (H _ist ) of the main component (H) provided by the metering device (V), as a function of an average value (H _m ) of the stored first measured values for actual provided actual quantities (H _ist ) of the main component (H) and / or depending on the characteristic curve (L) for the stored first measured values, a processing speed of the processing machine is controlled. Method according to at least one of the preceding claims, characterized in that the composition of the mixture to be supplied via the metering device (V) is predetermined and depending on the composition of the mixture, a processing speed of the processing machine is controlled. Dosing device designed for carrying out a method for automatic dosing and feeding a mixture of a main component (H) and at least one additive in a predetermined composition to a mixture processing processing machine, wherein the dosing device (V) for performing a method according to one of claims 1 to 23 is formed and wherein the dosing device (V) has at least: - a first conveyor unit ( 2 . 20 ) for providing the main component (H) in a demand quantity requested by the processing machine, the size of the requirement quantity depending on a demand signal (S) which the processing machine transmits to the metering device (V), - a first measuring device ( 4 ) for measuring at least one measured quantity for an actual quantity (H _ist ) of the main component (H) provided by the metering device (V), - an evaluation unit for storing the measured quantity for the actual quantity (H _ist ) of the main component (H) as at least a first measured value and for determining a desired additive amount (A _soll ) for the at least one additive based on a predetermined composition for the mixture to be supplied from the main component (H) and the at least one additive, wherein the evaluation unit for determining the desired additive quantity (A _soll ) evaluates the first measured value for the actually provided actual quantity (H _ist ) of the main component (H), - a second conveying unit ( 7L . 8L ; 7R . 8R ) for providing the additive in the additive target quantity (A _soll ) determined by the evaluation unit, and - a release opening ( 90 ), via which a supply of the mixture of provided main component (H) and provided additive from the metering device (V) to the processing machine takes place. Dosing device according to claim 24, characterized in that the metering device (V) is provided and adapted to be coupled with a spraying machine as a processing machine. Dosing device according to claim 25, characterized in that the metering device (V) is designed as a modular unit, which is to be subsequently coupled with a spraying machine as a processing machine. Spraying machine for processing plastic mixtures of a main component (H) and at least one additive with a metering device (V) according to one of claims 24 to 26 for metering and supplying the plastic mixture to a processing unit of the injection molding machine.

Images