EP3672903B1 - Method for filling containers with products - Google Patents

Description

technical field

The present invention relates to a method for filling containers with a filling product in a filling product filling plant.

Technical background

In filling product filling systems, it is known to fill containers to be filled with a filling product, the actual introduction of the filling product into the respective container to be filled being carried out by means of a so-called filling valve. The filling valve provides a connection between the filling product reservoir, in which the filling product to be filled is provided before the actual filling, and the container to be filled. The filling process is initiated by means of the filling valve and the product to be filled is guided into the container to be filled and the filling process is terminated again after a specific specification has been reached, for example after a specified filling weight, a specified filling height or a specified filling volume has been reached. Various sensors are known for determining the respective end of filling and thus for determining the respective state or time at which the filling valve is closed again, by means of which, for example, the filling level, the filling weight or the filling volume of the filling product in the container to be filled is determined.

Filling valves are known which merely open and close the respective connection between the filling product reservoir and the container to be filled. A throttle device is often connected upstream of these simply switching filling valves, by means of which a modulation of the filling product flow into the container to be filled can be achieved.

Also known are filling valves, which are also referred to as proportional valves, in which the respective filling valve cone is stepped or stepless relative to its filling valve seat can be raised or lowered, as a result of which the gap or annular gap forming between the filling valve cone and the filling valve seat can be varied in its cross section. Correspondingly, with such a proportional valve, a variation of the effective cross section and thus also a variation of the filling product flow flowing through the proportional valve can be achieved. It is thus possible with the proportional valve to specify or to control a predetermined volume flow profile for the filling of the respective container to be filled. It is thus possible, for example, at the beginning of the filling process, to first introduce a reduced flow of filling product into the container to be filled, in order in this way to reduce the tendency to foam. In the area of the main filling of the container to be filled, on the other hand, the highest possible volume flow is set in order to achieve rapid filling of the container to be filled. Toward the end of the filling process, the volume flow is then reduced again in order to make it possible to reliably reach the specified end of filling and to prevent the filling product from overflowing or squirting out of the container to be filled.

Proportional valves are often coupled in a control loop with a flow meter associated with this proportional valve. In this way, it is possible to use the combination of flow meter and proportional valve to specify a volume flow from a higher-level system control, which is then maintained via the control loop. However, both the flow meter and the corresponding evaluation device and the control of the proportional valve entail a certain inertia and time delay, so that an immediate reaction to changes in the initial conditions and in particular to changes in the supply of the filling product to the proportional valve can only be corrected with a certain time delay be able. Furthermore, flow meters are often dependent on the properties of the respective filling product.

In a configuration of a filling device in a filling product filling device such that each filling valve is connected directly to the filling product reservoir, for example in a configuration in which the filling valves arranged around the circumference of a filler carousel are each individually connected to the filling product reservoir, for example in the form of a central tank or a ring tank , are connected, the filling product reservoir acts as a buffer, such that each filling valve and in particular each proportional valve is operated independently of the other filling valves or proportional valve. In other words, they change Initial conditions for the respective filling valve not when an adjacent filling valve is opened or closed, since the filling product reservoir serves as a large-volume buffer.

In an alternative system design, on the other hand, in which at least two of the filling valves or a large number of filling valves or all of the filling valves are connected to the filling product reservoir via a single common filling product supply line, the initial conditions for each individual filling valve are influenced due to the properties of the line. This is the case, for example, if the filling product reservoir in which the filling product to be filled is provided is designed as a boiler provided and the filling product is connected to all filling valves of the filler carousel via a single filling product feed line, which is routed to the respective filler carousel via a rotary distributor.

In particular, in this embodiment, the pressure provided in the filling product feed line decreases when, at the beginning of the filling operation, starting from a state in which all filling valves are closed, one filling valve after the other is opened. The filling product supply line cannot then act as a quasi-unlimited buffer, but rather the volumetric flow flowing through the filling product supply line is dependent on the line radius to the fourth power.

The filling valves influence each other accordingly--at least until a steady state of equilibrium has been reached. This can mean that even with a filling valve controlled by a flow meter, the actually required flow rate is not achieved at least at the beginning of the respective filling operation due to the inertia of the control loop. This behavior of the filling valves is also observed towards the end of the filling operation, when all filling valves are closed one after the other before production is ended. Here, too, it happens that even with a filling valve controlled by a flow meter, the actually required flow rate is not achieved at the end of the respective filling operation due to the inertia of the control loop.

the EP 2 695 846 A1 describes a method for calculating the filling quantity in a rotary device and also a method for filling a container with a filling product in a filling product filling system having several filling valves by means of differential pressure determination, volume flow calculation and timed valve control based on the determined differential pressure.

the WO 97/00224 A1 describes a method for filling containers with a pressurized liquid. the EP 1 127 835 A1 describes a system and method for filling containers with a liquid product.

Presentation of the invention

Proceeding from the known state of the art, it is an object of the present invention to specify a method for filling containers with a filling product in a filling product filling plant, which shows a further improved filling behavior.

This object is achieved by a method for filling containers with a filling product having the features of claim 1. Advantageous developments result from the dependent claims, the present description and the figures.

Accordingly, a method for filling a container with a filling product in a filling product filling system having a control valve is proposed with the following steps: determining a differential pressure Δ p _v dropping across the control valve and regulating and/or controlling the control valve depending on the determined differential pressure Δ p _v .

Since the control and/or regulation of the control valve is carried out on the basis of the differential pressure Δ p _v , a very reliable regulation can be achieved which responds quickly, is decoupled from the properties of the filling product and no longer has the inertia of a flow sensor. A reliable and fast open-loop and/or closed-loop control behavior can thus be achieved in a filling product filling plant.

A function of the volume flow q ( t , Δ p _v ) for the control valve as a function of the differential pressure Δ p _v dropping across the control valve is preferably determined, the volume flow q ( t , Δ p _v ) through the control valve based on the determined differential pressure Δ p _v is calculated and the control valve is regulated and/or controlled as a function of the calculated volume flow q ( t , Δ p _v ), t is the time.

This means that other components of the control behavior can also be included and, in particular, the transient behavior of the control valve can be taken into account.

According to the invention, at least two filling valves connected in parallel are provided in the filling product filling system, and a function of the volume flow q ( t , Δ p _v ) for at least two of the filling valves connected in parallel is determined as a function of a differential pressure Δ p _v across the filling valves connected in parallel, the differential pressure Δ p _v over the parallel connected filling valves is determined, the volume flow q ( t , Δ p _v ) through at least one of the filling valves connected in parallel is calculated on the basis of the determined differential pressure Δ p _v and the at least one filling valve is controlled as a function of the calculated volume flow q ( t , Δ p _v ). and/or regulated. Regulating and/or controlling the at least one filling valve correspondingly includes compensation for the opening position of the filling valve when the differential pressure Δp _v changes using the calculated current volume flow q ( t , Δp _v ) .

Because a function of the volume flow q ( t, Δ p _v ) is determined as a function of a differential pressure Δ p _v across the filling valves connected in parallel and the at least one filling valve is activated as a function of the calculated volume flow q ( t, Δ p _v ) is controlled, it can be achieved that the control behavior is improved when filling the respective container. In particular, due to the lower inertia of the differential pressure measurement, it is possible to react more quickly to a change in the differential pressure within the device, which is usually due to the fact that further filling valves connected in parallel are switched on or off.

In other words, on the basis of the method, even if a device is designed with several filling valves connected in parallel to one another, which are switched on or off one after the other during the entire filling process, it can be achieved that a reliable and uniform filling result is nevertheless achieved in the containers to be filled becomes.

For example, in a start-up phase, in which the filling process begins and the first filling valve is opened first, while all other filling valves are still closed, a higher differential pressure will result, so that initially, in principle, due to this differential pressure, a higher volume flow through the or the few open filling valves is to be expected. In order to achieve the desired volume flow in the container to be filled, the corresponding filling valve is regulated or controlled according to the calculated volume flow in comparison to the desired volume flow, so that it is opened to a lesser extent. Depending on the calculated volume flow based on the determined differential pressure, the filling product can flow in with the desired volume flow.

As soon as the second filling valve opens in order to fill a container downstream on the filler carousel with the filling product, the differential pressure dropping across the filling valves decreases accordingly, so that the volume flow through the first filling valve and then also through the second filling valve decreases slightly. By determining the differential pressure, it is possible to ensure that the corresponding volume flow at the first filling valve is correctly predicted and the first filling valve is opened a little further as the differential pressure drops in order to continue to maintain the desired volume flow. The regulation based on the differential pressure reacts much more quickly than regulation using a flow meter, for example, could. A time delay in the regulation of the filling valve is therefore shorter and the result achieved, namely the maintenance of the specified volume flow, is therefore more precise by means of the regulation and/or control based on the calculated volume flow.

Accordingly, the determined function of the volume flow, which depends on the differential pressure across the filling valves connected in parallel, can be used to calculate the volume flow through the filling valve and, with a correspondingly varying differential pressure across the filling valves connected in parallel, the respective opening stroke of the filling valves can be adjusted in order to to maintain the desired or predetermined volume flow in the respective container to be filled, regardless of the number and degree of opening of the other filling valves connected in parallel.

As soon as full operation is reached and a settled balance of the simultaneously open filling valves has been established, a variation in the differential pressure resulting from the opening and closing of the respective filling valves can hardly be detected due to the large number of simultaneously open valves. Accordingly, in full operation, the respective filling valves are only readjusted to a very small extent on the basis of the determined differential pressure and the volume flow calculated from it.

It is therefore preferred either to suspend the corresponding regulation and/or control of the filling valve depending on the calculated volume flow during full operation, or to carry out the regulation and/or control only when the resulting stroke of the filling valve exceeds a specific threshold to be set. In other words, it can be prevented in this way that high-frequency control or regulation of the filling valves arranged parallel to one another takes place during full operation. Rather, only longer-term Changes in the pressure behavior, which, for example, reveal a trend, are accepted and corrected. Such a trend can exist, for example, when the filling product reservoir, from which the filling product is supplied to the filling valves connected in parallel with one another, has a changed level or changed pressure conditions. Correspondingly, the proposed method can also be used to achieve compensation in a case in which the total pressure present across the filling valves is changed due to the supply of the filling product.

A volume flow profile predetermined by the filling method for the respective filling product and the respective container to be filled is particularly preferred. The filling valves are, for example, controlled via their respective individual flow meters, driven towards the specified volume flow profile. Accordingly, the respective filling valve is set to a predetermined opening value, which is assumed to correspond to the corresponding volume flow predetermined by the volume flow profile, and is then regulated precisely to this value via the respective flow meter.

This actuation of the filling valve by means of the predetermined volume flow profile is superimposed by the regulation and/or control by means of the proposed method, which enables compensation of the filling product flow into the container to be filled based on the determined differential pressure.

In other words, by measuring or determining the differential pressure, which is much faster than measuring the flow, the corresponding filling valve can be controlled to the appropriate position, which results from the volume flow profile with compensation by the calculated volume flow based on the determined differential pressure . The compensation based on the differential pressure can take place, for example, in a time range of one millisecond due to the fast-reacting pressure sensors. In contrast, regulation by changing the flow rate using the flow meter would require a regulation time of around 50 milliseconds. Correspondingly, due to the compensation modulated onto the volume flow profile provided, a more precise filling behavior can be achieved, so that incorrect filling can be better avoided.

In particular, right at the beginning of the filling process, filling can take place with a correct filling product volume flow. With each additional filling valve that is opened, it decreases the differential pressure across the filling valves connected in parallel to one another, since the total cross-section of the open filling valves increases and the pressure above the filling valves is reduced. The volume flow through the individual filling valves is thus also reduced in such a way that the filling valves have to be opened further in order to maintain the filling product volume flow desired according to the volume flow profile.

The same applies towards the end of the filling process, when the last containers to be filled pass through the system and one filling valve after the other is moved into the closed position and remains in this position. The situation here is that the differential pressure increases with each filling valve that closes and the volume flow that flows through the last filling valve would be greater if countermeasures were not taken and the filling valves were closed more and more based on the calculated volume flow.

This means that both the first filled containers and the last filled containers are still filled correctly in the filling process and the risk of incorrect filling is correspondingly further reduced.

Exact compensation of the respective differential pressure works particularly well in the described system design with a large number of filling valves connected in parallel because the filling valves in the form of proportional valves require a certain amount of time to control the desired degree of opening due to their design. In other words, the respective filling valve is gradually moved from the fully closed position to the desired open position. Accordingly, the filling valves do not open abruptly, as would be the case with pure switching valves, but the opening takes place in such a way that the volume flow through the filling valve slowly increases and, correspondingly, the differential pressure, which results from another filling valve opening, also increases is only slowly reduced.

In combination with the significantly faster pressure sensor, which has a significantly faster detection behavior than, for example, a flow sensor, a corresponding compensation of the respective opening filling valve can be achieved, which ultimately leads to a virtually unaffected or even volume flow in the already open filling valves.

In a further preferred embodiment, the regulation and/or control of the filling valve includes the actuation of an open position of the filling valve based on the current volume flow q ( t , _Δpv ) .

The regulation and/or control of the at least one filling valve is preferably carried out taking into account a predetermined volume flow profile for filling the container to be filled with the filling product.

gegeben, wobei $$q_{\infty }=K_v\cdot \sqrt{\frac{\mathit{\Delta p}}{1\mathit{bar}}\cdot \frac{1000\frac{\mathit{kg}}{m^3}}{\rho }}$$

der Volumenstrom durch das Füllventil im eingeschwungenen Zustand ist.The function of the volume flow q(t, Δ p _v ) as a function of the differential pressure Δ p _v is preferred $$q\left(t,{\mathit{\Delta p}}_v\right)=\{\begin{array}{ll}{q}_{\infty }\cdot \mathit{tanh}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_H}\cdot t+\mathit{arctanh}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0<q_{\infty }\\ q_{\infty }\cdot \mathit{cotton}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_H}\cdot t+\mathit{arccoth}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0>q_{\infty }\\ {\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0,& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0=q_{\infty }\end{array}$$

given where $$q_{\infty }=K_v\cdot \sqrt{\frac{\mathit{<figure-callout\; id="1"\; label="\Delta p"\; filenames="imgf0001.png,imgf0004.png"\; state="\{\{state\}\}">\Delta p</figure-callout>}}{1\mathit{bar}}\cdot \frac{1000\frac{\mathit{<figure-callout\; id="3"\; label="kg\; m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">kg</figure-callout>}}{m^{}}}{\rho }}$$

the volume flow through the filling valve is in the steady state.

In this way, the volume flow can also be calculated for a complex system with a large number of filling valves on the basis of the differential pressure, with the mutual influence of the volume flows of the filling valves on one another being taken into account by these equations. In other words, the calculation in this way enables a more precise calculation of the volume flow and thus an improved filling result.

Brief description of the figures

Figur 1

eine schematische perspektivische Darstellung eines Füllerkarussells mit einem nebengestellten Füllproduktreservoir,

Figur 2

eine schematische Darstellung der exemplarisch gemessenen Volumenströme von vier parallel geschalteten Füllventilen ohne Kompensation;

Figur 3

eine schematische Darstellung eines exemplarisch gemessenen Volumenstroms eines Füllventils beim nachfolgenden Öffnen weiterer, parallel geschalteter Füllventile in einer vergrößerten Detaildarstellung;

Figur 4

eine schematische Darstellung einer Kurve eines Leitwerts Kv über den Hub H eines Proportionalventils;

Figur 5

ein Ersatzschaltbild in einer elektrisch-fluidtechnischen Analogie für den Fülleraufbau gemäß Figur 1;

Figur 6

ein Ersatzschaltbild in einer elektrisch-fluidtechnischen Analogie eines einzelnen Füllventils;

Figur 7

ein Ersatzschaltbild in einer elektrisch-fluidtechnischen Analogie eines einzelnen Füllventils unter Berücksichtigung des Differenzdrucks;

Figur 8

eine schematische Darstellung der einzelnen Maschen in einem Ersatzschaltbild in einer elektrisch-fluidtechnischen Analogie des Fülleraufbaus beispielsweise nach Figur 1;

Figur 9

schematische Darstellung der Überlegung zur Reduktion der Differenzialgleichungen aus den Maschengleichungen unter Berücksichtigung des Differenzdrucks; und

Figur 10

schematische Darstellung einer alternativen Ausführungsform.

Preferred further embodiments of the invention are explained in more detail by the following description of the figures. show:

figure 1

a schematic perspective representation of a filler carousel with an adjacent filling product reservoir,

figure 2

a schematic representation of the exemplary measured volume flows of four parallel filling valves without compensation;

figure 3

a schematic representation of an exemplary measured volume flow of a filling valve during the subsequent opening of further, parallel-connected filling valves in an enlarged detailed representation;

figure 4

a schematic representation of a curve of a conductance Kv over the stroke H of a proportional valve;

figure 5

an equivalent circuit diagram in an electrical-fluidic analogy for the filler structure according to figure 1 ;

figure 6

an equivalent circuit diagram in an electrical-fluidic analogy of a single filling valve;

figure 7

an equivalent circuit diagram in an electrical-fluidic analogy of a single filling valve, taking into account the differential pressure;

figure 8

a schematic representation of the individual meshes in an equivalent circuit diagram in an electrical-fluidic analogy of the filler structure, for example figure 1 ;

figure 9

Schematic representation of the consideration for reducing the differential equations from the mesh equations, taking into account the differential pressure; and

figure 10

schematic representation of an alternative embodiment.

Detailed description of preferred embodiments

Preferred exemplary embodiments are described below with reference to the figures. Elements that are the same, similar or have the same effect are provided with identical reference symbols in the different figures, and a repeated description of these elements is sometimes dispensed with in order to avoid redundancies.

In figure 1 a schematic perspective view of a filler carousel 10 is shown, which has a large number of filling valves 12 arranged around the circumference of the filler carousel 10, each of which has a filling valve outlet 14 under which the containers to be filled, not shown in this figure, are arranged. The container to be filled, which is arranged underneath, is filled with a filling product via the respective filling valve outlet 14 . The filling valve 12 serves to fill the desired volume, the desired mass or the desired level of filling product into each container to be filled. In the filling operation, the filler carousel 10 rotates about its axis of rotation in order to produce a constant stream of filled containers.

A dedicated fill product reservoir 16 in the form of a dedicated fill product kettle is provided. The filling product is stored in the filling product reservoir 16 before the containers to be filled are actually filled.

The filling level of the filling product in the filling product reservoir 16 can be kept constant via a separate mechanism, for example by means of a filling level sensor in the filling product reservoir 16, via which a supply of filling product into the filling product reservoir 16 is regulated. The advantage of keeping the filling level in the filling product reservoir 16 constant is that the pressure and flow conditions in the system areas downstream of the filling product reservoir 16 can be determined more easily, since the hydrostatic pressure applied via the filling product reservoir 16 is always the same.

Alternatively or additionally, the filling level of the filling product in the filling product reservoir 16 can also be determined via a filling level sensor and the system parts located downstream of the filling product reservoir 16 can be controlled or regulated according to the filling level of the filling product.

The filling product reservoir 16 is connected to the individual filling valves 12 via a filling product line 18 which is routed to the filler carousel 10 via a rotary distributor 19 . Correspondingly, all filling valves 12 are connected to the filling product reservoir 16 that is provided via the filling product supply line 18 and the rotary distributor 19 .

In the embodiment shown, the individual filling valves 12 are connected to one another via a ring line 11 located on the filler carousel 10 and the ring line 11 is connected to the filling product supply line 18 via four distributor lines 17 with the rotary distributor 19 being interposed. Other line-based constellations for connecting the filling product supply line 18 to the filling valves 12 can also be provided here.

This structure of the filler with an adjacent filling product reservoir 16 makes it possible to dispense with the construction of a boiler on the filler carousel 10, which means that costs can be saved. In addition to the filling product reservoir 16, which is simpler to set up, the filler carousel 10 itself can also be dimensioned smaller with regard to the bearings and the statics due to the lower rotating mass, and the required drives and drive energy can be reduced. This not only results in a lower investment volume, but also reduced operating costs.

During the filling operation, containers to be filled are fed to the filler carousel 10 in a manner known per se in the area of the respective filling product outlets 14 of the filling valves 12, filled at these and then the filled containers are discharged from the filler carousel 10 again in a manner known per se.

At the beginning of the respective filling operation, a first container is correspondingly initially supplied and the corresponding filling valve 12 is opened. Then the second container to be filled is fed in and the second filling valve 12 is opened. The process continues in this way until equilibrium has been reached and all filling points in the filling angle are occupied.

Accordingly, at the beginning of the respective filling operation, the filling valves 12 are moved from a situation in which all filling valves 12 are closed to an operation in which a large number of filling valves 12 are open at the same time. In the full filling operation, a large number of filling valves 12 are then operated simultaneously—which is a steady state here Equilibrium is because constantly at the beginning of the filling angle a filling valve 12 is opened and shortly before or shortly after at the end of the filling angle another filling valve 12 is closed. In the full filling operation, the supplied stream of containers to be filled is correspondingly filled with the filling product, and after the filling process is complete, a stream of filled containers can leave the filler carousel 10 again. This operation of a filler carousel 10 is known in principle.

At the filling valves 12, which in the figure 1 are shown, these are so-called control valves or proportional valves, the control valves being designed accordingly in such a way that, in addition to a fully closed position and a fully open position, they also have at least one intermediate position, preferably a large number of intermediate positions or stepless adjustment of the active filling cross section, enable. Correspondingly, a filling valve cone can be lifted out of its corresponding filling valve seat in steps or continuously, so that the annular gap formed between the filling valve cone and the filling valve seat or its cross section can be changed in the said steps or continuously. Accordingly, the filling valve designed in this way as a control valve makes it possible to control the flow of filling product through the filling valve 12 via the position of the filling valve cone relative to the filling valve seat.

Control valves are also used at other positions within a filling product bottling plant in order to vary the flow of media and, in particular, of the filling product. The explanations of the present disclosure given here below are carried out by way of example on a filling device in which control valves are used as filling valves 12 . In principle, however, the considerations can be transferred to the control and regulation of each control valve within a filling product filling system.

The following explanations, which are given on the basis of filling valves 12 designed as control valves, can therefore also be applied, for example, to superstructures of a filling product filling system in which control valves for varying the flow rate are provided in front of the actual filling valves, which are then designed as simple switching valves (open/closed). . The explanations can also be applied, for example, to structures in which a single control valve is provided in the inlet to a filler - such a structure is shown, for example, in figure 10 described.

In the following, however, reference is initially made to a structure in which all of the filling valves 12 under consideration are designed as control valves.

Each filling valve 12 is usually in communication with an individual flow meter or a load cell in such a way that a desired volume flow can be specified, which can then be regulated by the filling valve 12 via its associated flow meter.

For this purpose, the filling valve 12 is usually first moved into a predetermined opening position, which is also referred to as the pilot position, which is assumed to correspond to the desired volume flow, and then the resulting volume flow is measured via the flow meter accordingly by varying the opening stroke of the Filling valve 12 regulated precisely.

The pre-control position has so far been determined for steady-state operation and is based on the conditions in steady-state operation.

In the exemplary embodiment of the filler shown, in which all filling valves 12 are connected to the adjacent filling product reservoir 16 via the filling product feed line 18, the opening of each individual filling valve 12 leads to changing pressure conditions in the filling product feed line 18. This is due, among other things, to the hydraulic inductance of the fluid in the filling product supply line 18 is based. Accordingly, at the beginning of the filling process, when first a first filling valve 12 and then more and more filling valves 12 are opened, starting from an initial differential pressure, the differential pressure decreases more and more slowly, which correspondingly influences the volume flow through the already opened filling valves 12.

This behavior is schematic in figure 2 shown, in which the volume flow through four directly adjacent filling valves a) - d) is shown, which are switched on one after the other at intervals of about 1 second.

When the first filling valves 12 are driven to the pilot control position determined in equilibrium operation, the expected volume flow is therefore not achieved, but rather a higher volume flow, which then gradually decreases. This is schematic again in figure 3 shown in the behavior of the filling valve a) from the figure 2 shown again in a higher resolution. In this particular measured example, the drop in volume flow is more than 100 ml/sec.

The same occurs at the end of the filling operation, when the last containers to be filled are received in the filler carousel 10 and more and more filling valves 12 are closed until finally only one last filling valve 12 is left, which is then closed. A gradual increase in pressure takes place here, so that the flow conditions and in particular the volume flow through the individual, remaining or remaining filling valves 12 change accordingly.

The observed behavior at the end of the filling operation corresponds essentially to that from the figures 2 and 3 , only with the reverse course over time, in which the volume flow of the last filling valve 12 then increases accordingly.

The control loop between the individual filling valve 12 and the flow meter assigned to this filling valve 12 is too sluggish to reliably compensate for these volume flow fluctuations.

In order to better understand this behavior of the filling valves 12, the following considerations should be taken into account.

The basis for the improved control process proposed here is precise knowledge of the filling valve 12 and, in particular, of the control valve used in each case. Knowledge of the relationship between the conductivity K _V and the stroke H of the control valve plays a role here:
First, a function of the conductance K _V (H) of the control valve is determined for each opening position H of the control valve. The conductance K _V is also referred to as the flow factor or flow coefficient of the control valve. It is a measure of the achievable throughput of a liquid or gas through the control valve, is given here in the unit ml/sec and can be interpreted as the effective cross-section. Each K _V value only applies to the associated opening position H of the control valve.

In order to determine the conductance K _V , a specific opening position H _i of the control valve is approached in an initial calibration process, the filling product flow q(H) from the control valve is measured at this opening position H _i and from this in the steady state the conductance K _V is determined, for example by means of a measurement a measuring cell such as a load cell. This is carried out for a large number of discrete opening positions H _i of the control valve.

mit Δp Differenzdruck zwischen Füllventilauslauf und dem Druck oberhalb des Regelventils und p der Dichte des das Regelventil durchfließenden Füllprodukts.The following relationship exists between the K _V value and volume flow q _∞ (volume flow through the filling valve in the steady state): $$q_{\infty }=K_v\ast \sqrt{\frac{\mathit{<figure-callout\; id="1000"\; label="\Delta p"\; filenames="imgf0004.png"\; state="\{\{state\}\}">\Delta p</figure-callout>}}{1000\mathit{mbar}}\ast \frac{1000\mathit{kg}/{\mathrm{<figure-callout\; id="3"\; label="m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">m</figure-callout>}}^3}{\rho }}$$

with Δp differential pressure between filling valve outlet and the pressure above the control valve and p the density of the filling product flowing through the control valve.

Correspondingly, for the exact determination of the conductance _KV , in addition to the above-mentioned measurement of the volume flow at a specific opening position, the differential pressure Δ p and the density p of the filling product flowing through the control valve must also be determined.

The density p of the filling product is usually known or can be determined using known measurement methods. For water and water-like filling products, which are mainly filled in beverage bottling plants, the density can be assumed to be approximately 1000 kg/m ³ , so that it then does not need to be changed for a large number of filling products to be filled.

Correspondingly, the KV value for this opening position can now be determined from the volume flow q measured for a specific opening position _{H i} , the determined differential pressure Δ p and the determined density p by: $$K_v\left(H_i\right)=q_{\infty }\ast \sqrt{\frac{1000\mathit{mbar}}{\mathit{\Delta p}}\ast \frac{\rho }{1000\mathit{kg}/{\mathrm{<figure-callout\; id="3"\; label="m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">m</figure-callout>}}^3}}$$

In order to determine a function of the conductance K _V (H) over the opening positions H _i , after all conductance values K _V (H _i ) have been determined, a compensating curve is determined by the respective conductances K _V (H _i ) determines a function of the conductance over the opening positions of the control valve. The regression curve can be determined, for example, by linear regression, the least squares method, a fit algorithm or other known methods for determining a regression curve from measured values. This determination and calculation is performed for various discrete values of the opening position _Hi .

A sixth-order polynomial, for example, can be used as a regression curve, as is the case, for example, in figure 4 is shown, in which the conductance is plotted against the respective opening position of the control valve. In the figure 4 a first range of values for the opening positions from 0 to 2 mm and a second range of values for the opening positions from 2 mm to 6 mm were used to determine the compensating curve. The discrete values 20 in the first value range and the discrete values 22 in the second value range were used to form the curve of the K _V values 2 over the opening position H of the control valve to form a compensating curve using a 6th order polynomial.

For a specific stroke H of the control valve, the result is, for example, the compensating curve of the conductance K _V : $${\mathrm{K}}_{\mathrm{V}}\left(\mathrm{H}\right)={\mathrm{<figure-callout\; id="6"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_6*{\mathrm{<figure-callout\; id="6"\; label="H"\; filenames="imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}^6+{\mathrm{<figure-callout\; id="5"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_5*{\mathrm{<figure-callout\; id="5"\; label="H"\; filenames="imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}^5+{\mathrm{<figure-callout\; id="4"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_4*{\mathrm{<figure-callout\; id="4"\; label="H"\; filenames="imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}^4+{\mathrm{<figure-callout\; id="3"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_3*{\mathrm{<figure-callout\; id="3"\; label="H"\; filenames="imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}^3+{\mathrm{<figure-callout\; id="2"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_2*{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="imgf0004.png"\; state="\{\{state\}\}">H</figure-callout>}}^2+{\mathrm{<figure-callout\; id="1"\; label="c"\; filenames="imgf0001.png,imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_1*\mathrm{H}+{\mathrm{<figure-callout\; id="7"\; label="c"\; filenames="imgf0004.png"\; state="\{\{state\}\}">c</figure-callout>}}_7$$

Where c ₁ to c ₇ are the corresponding coefficients for fitting the function to the measured values.

By determining the compensating function, all intermediate values of the opening positions can then also be taken into account during filling. This means that the corresponding volume flow can be calculated for steady states for each opening position: $$q_{\infty }\left(H\right)=K_v\left(H\right)*\sqrt{\frac{\mathit{<figure-callout\; id="1000"\; label="\Delta p"\; filenames="imgf0004.png"\; state="\{\{state\}\}">\Delta p</figure-callout>}}{1000\mathit{mbar}}\ast \frac{1000\mathit{kg}/{\mathrm{<figure-callout\; id="3"\; label="m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">m</figure-callout>}}^3}{\rho }}$$

It should be noted, however, that this function of the conductance K _V (H) of the control valve for each opening position is the respective volume flow in the steady state, i.e. after the opening position has been kept constant and after a longer period of waiting. At the Opening, closing or moving the control valve from one open position to another open position, on the other hand, have other dynamic influences.

In order to consider the dynamic influences caused by the opening or closing of the adjacent or other filling valves of the filler carousel designed as control valves, an analogy from the field of electrical engineering is first drawn, whereby the electrical-mechanical analogy given in the following table is used: electrical consideration mechanical consideration ohmic resistance Kv value tension differential pressure electricity flow rate inductance accelerated mass

In figure 5 The flow-mechanical structure of some filling valves 12 a) to d) designed as control valves, which are in communication with the filling product reservoir provided via the filling product feed line 18, is shown schematically in an electrical-fluidic analogy with the help of an equivalent circuit diagram.

are in figure 5 : K _VInlet : Conductance inflow K _V1-n : Conductance single filling valve L _inlet : Hydraulic inductance feed L _1-n : Hydraulic inductance filling valve Δp: differential pressure q: flow rate inlet q1-n: Flow rate filling valve n: Number of filling valves

The opening position or the degree of opening of the filling valve 12 influences the system variables KV1-n and L _1-n and thus indirectly the potential and flow variables.

The filling product supply line 18 accordingly comprises a hydraulic inductance L _supply and a conductance K _V-supply , with which the behavior of the filling product supply line 18 can be described according to the electrical-fluidic analogy.

The entire volume flow q, which is supplied from the filling product reservoir provided, is supplied to the individual filling valves 12 accordingly via the filling product feed line 18 .

The individual filling valves 12 are connected in parallel to one another and are all connected to the filling product supply line 18 . Correspondingly, each filling valve 12 also has a hydraulic inductance L ₁ and a conductance K _V1 , by means of which the flow behavior of each filling valve 12 can be represented according to the electrical-fluidic analogy.

In order to achieve an improved control and/or regulation behavior of the filling product filling system 1, in particular at the beginning and at the end of the respective filling operation, the following additional considerations must be made:
In figure 6 the structure of an individual filling valve 12 designed as a control valve is shown schematically.

The differential pressure across the conductance results in: $${\mathit{\Delta p}}_{K_v}\left(t\right)={\left(\frac{q\left(\mathrm{<figure-callout\; id="2"\; label="t\; K\; v"\; filenames="imgf0004.png"\; state="\{\{state\}\}">t</figure-callout>},\right)}{K_v}\right)}^2\cdot 1\mathit{bar}\cdot \frac{\mathrm{<figure-callout\; id="1000"\; label="\rho "\; filenames="imgf0004.png"\; state="\{\{state\}\}">\rho </figure-callout>}}{1000\frac{\mathit{<figure-callout\; id="3"\; label="kg\; m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">kg</figure-callout>}}{m^{}}}$$

The differential pressure across the hydraulic inductance results in: $${\mathit{\Delta p}}_{L_H}\left(t\right)=L_H\cdot \frac{\mathit{dq}\left(t\right)}{\mathit{German}}$$

mit

• l = wirksame Leitungslänge
• ρ = Dichte der Flüssigkeit
• A = wirksamer Fließquerschnitt

Where the hydraulic inductance is given to $$L_H=\frac{l\cdot \rho }{A}$$

With

• l = effective cable length
• ρ = density of the liquid
• A = effective flow area

The formula can be used for more complicated line geometries in infinitesimally small sections. The resulting individual inductances are then to be added up or integrated to form a total inductance.

Now the differential equation of the individual valve is set up and solved according to the volume flow. This calculated volume flow is finally transferred to a normal control algorithm to compensate for the drops in volume flow - for example by adjusting a pilot control position.

In the figure 7 shows schematically the consideration for a single stitch on this basis. From this consideration, the differential pressure Δ p _v of the control valve under consideration over this individual mesh results in: $${\mathit{\Delta p}}_v={\mathit{\Delta p}}_{K_v}+{\mathit{\Delta p}}_{L_H}={\left(\frac{q\left(\mathrm{<figure-callout\; id="2"\; label="t\; K\; v"\; filenames="imgf0004.png"\; state="\{\{state\}\}">t</figure-callout>},\right)}{K_v}\right)}^2\cdot 1\mathit{bar}\cdot \frac{\mathrm{<figure-callout\; id="1000"\; label="\rho "\; filenames="imgf0004.png"\; state="\{\{state\}\}">\rho </figure-callout>}}{1000\frac{\mathit{<figure-callout\; id="3"\; label="kg\; m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">kg</figure-callout>}}{m^{}}}+L_H\cdot \frac{\mathit{dq}\left(t\right)}{\mathit{German}}$$

This mesh equation is now to be set up for each of the filling valves 12 of the respective filling product bottling plant 1, from which a complex differential equation system results accordingly.

The structure of the differential equation system is shown schematically in figure 8 , in which the respective meshes I, II, ..., each representing a row of the differential equation system, are shown.

This differential equation system describes the mutual influence of the filling valves 12 when the filling valves 12 are connected in parallel given the differential pressure Δ p _v dropping across these filling valves 12 .

However, such a differential equation system can no longer be solved analytically, but must be solved numerically. However, with the available computing power of a control computer, this is not practicable in filling operation and would be too slow. Furthermore, would have - as can be seen from the figure 8 results, also the material quantities K _{v Intake} and L _inflow can be determined and measured for the respective machine.

In order to circumvent these problems, the underlying equivalent circuit diagram and thus the differential equation system are reduced. It turned out how figure 9 it can be seen that the equivalent circuit diagram can be reduced by measuring the differential pressure Δ p _v across the parallel connection of the filling valves 12 and a separate determination of the conductance and the hydraulic inductance can be dispensed with.

In other words, by measuring the differential pressure Δ p _v across the individual filling valve 12 or across the parallel connection of the active filling valves 12, a simplification of the determination of the throughflow can be achieved.

The differential pressure Δ p _v can be determined in a simple manner in the filling product bottling plant 1 by means of appropriate pressure sensors. The pressure sensors have a very short response time, for example in the range of 1ms, and are sufficiently accurate. This results in a very quick measurement of the differential pressure Δ p _v and thus the possibility of a quick determination of the resulting volume flow through the respective filling valve.

With this, the following solution for the volume flow q _n ( t ) of the respective nth individual valve when a measured differential pressure Δ p _v is present can be found: $$q_n\left(t\right)=\{\begin{array}{ll}{q}_{n_{\infty }}\cdot \mathit{tanh}\left(\frac{{\mathit{\Delta p}}_v}{q_{n_{\infty }}\cdot L_H}\cdot t+\mathit{arctanh}\left(\frac{q_{n_0}}{q_{n_{\infty }}}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q\; n"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_{n_{}}<q_{n_{\infty }}\\ q_{n_{\infty }}\cdot \mathit{cotton}\left(\frac{{\mathit{\Delta p}}_v}{q_{n_{\infty }}\cdot L_H}\cdot t+\mathit{arccoth}\left(\frac{q_{n_0}}{q_{n_{\infty }}}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q\; n"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_{n_{}}>q_{n_{\infty }}\\ {\mathrm{<figure-callout\; id="0"\; label="q\; n"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_{n_{}},& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q\; n"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_{n_{}}=q_{n_{\infty }}\end{array}$$

Where q _{n 0} is the volume flow of the nth filling valve at the beginning of the analysis and the volume flow q _{n ∞} of the nth filling valve in the fully settled state results in: $$q_{n_{\infty }}=K_v\cdot \sqrt{\frac{\mathit{<figure-callout\; id="1"\; label="\Delta p"\; filenames="imgf0001.png,imgf0004.png"\; state="\{\{state\}\}">\Delta p</figure-callout>}}{1\mathit{bar}}\cdot \frac{1000\frac{\mathit{<figure-callout\; id="3"\; label="kg\; m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">kg</figure-callout>}}{m^{}}}{\rho }}$$

However, in a first approximation, the same pressure prevails at the filling valve outlet 14 of all filling valves 12 . This pressure can, for example, the ambient pressure in a free jet process or the Pressure of a defined applied bias in the container to be filled. The corresponding pressure at the filling valve outlet 14 is therefore known in principle and is equal to a first approximation for each filling valve 12 at the respective start of filling.

Furthermore, due to the joint connection of all filling valves 12 to the filling product supply 18 - for example via the ring line 11 - the same pressure also prevails above the filling valves 12 in a first approximation. Accordingly, to simplify the method, an individual consideration of the individual filling valves 12 can be dispensed with. In other words, the measured differential pressure Δ p _v corresponds to the differential pressure across all active control valves that are present in the corresponding parallel circuit.

This results in the volume flow q(t) of the respective individual filling valve 12, taking into account the assumptions mentioned for each filling valve 12 on the basis of the measurement of the pressure in the filling product feed 18 or in the ring line 11, the knowledge of the pressure at the filling valve outlet 14 and the determination of the resulting differential pressure Δ p _v to: $$q\left(t,{\mathit{\Delta p}}_v\right)=\{\begin{array}{ll}{q}_{\infty }\cdot \mathit{tanh}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_H}\cdot t+\mathit{arctanh}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0<q_{\infty }\\ q_{\infty }\cdot \mathit{cotton}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_H}\cdot t+\mathit{arccoth}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0>q_{\infty }\\ {\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0,& \mathit{if}{\mathrm{<figure-callout\; id="0"\; label="q"\; filenames="imgf0004.png"\; state="\{\{state\}\}">q</figure-callout>}}_0=q_{\infty }\end{array}$$

With $$q_{\infty }=K_v\cdot \sqrt{\frac{\mathit{<figure-callout\; id="1"\; label="\Delta p"\; filenames="imgf0001.png,imgf0004.png"\; state="\{\{state\}\}">\Delta p</figure-callout>}}{1\mathit{bar}}\cdot \frac{1000\frac{\mathit{<figure-callout\; id="3"\; label="kg\; m"\; filenames="imgf0004.png"\; state="\{\{state\}\}">kg</figure-callout>}}{m^{}}}{\rho }}$$

Accordingly, when determining the differential pressure Δ p _v by connecting the filling valves 12 in parallel, which applies accordingly to each filling valve 12, the mutual influence of the filling valves 12 is fully included in the individual calculation of the volume flow.

The volume flow q ( t , Δ p _v ) calculated in this way based on the differential pressure Δ p _v is then transferred to a controller or regulator in order to achieve appropriate control of the valve position of the respective control valve to maintain the specified target volume flow.

This is used in particular for the pilot control of the respective control valve, with the control valve then being controlled on the basis of the respectively currently measured differential pressure Δ p _v when it is opened in such a way that the desired volume flow is precontrolled.

In this way it can be achieved that, especially at the beginning of the filling operation or at the end of the filling operation when only a few filling valves 12 are active, a compensated activation of the pilot control position and the operating position of the filling valves 12 is achieved.

The control, which is carried out on the basis of the volume flow q ( t , Δp _v ) calculated based on the currently measured differential pressure Δ p _v , can be modulated onto the remaining control and/or regulation steps of a higher-level system control.

The remaining control and/or regulation behavior of the individual filling valve 12—for example, to achieve a predetermined flow curve for filling the container to be filled according to a volume flow profile appropriate to the filling product and the container—is not changed as a result. Rather, the compensation via the volume flow q ( t , Δp _v ) calculated based on the currently measured differential pressure Δ p _v achieves more precise compliance with the required volume flow profile, regardless of the number of filling valves 12 opened at the same time.

The method for compensation can be used at the beginning and at the end of the respective filling operation until a steady equilibrium of the number of filling valves 12 opened in parallel has resulted in full operation.

However, the method can also be continuously compensated during the entire filling operation to compensate for the opening position of all filling valves 12, taking into account the differential pressure Δ p _v .

• Füllventil n ist offen, und der Volumenstrom durch das Füllventil n ist konstant eingeschwungen
• Das Füllventil n+1 wird geöffnet. Dadurch ändert sich der Differenzdruck Δp_v über der Parallelschaltung der Füllventile
• Dies wird durch die entsprechenden Drucksensoren erkannt und auf dieser Grundlage der Volumenstrom berechnet, der entsprechend sinkt
• Der berechnete Volumenstrom wird an die Regelung als Regelgröße übergeben
• Die Regelung erhöht am Füllventil n den Öffnungshub so, dass der gewünschte Soll-Volumenstrom (Führungsgröße) gehalten wird

The control procedure can also be carried out like this, for example:

• Filling valve n is open and the volume flow through filling valve n has stabilized constantly
• The filling valve n+1 is opened. This changes the differential pressure Δ p _v over the Parallel switching of the filling valves
• This is detected by the corresponding pressure sensors and the volume flow is calculated on this basis, which decreases accordingly
• The calculated volume flow is transferred to the controller as a controlled variable
• The controller increases the opening stroke at the filling valve n in such a way that the desired target volume flow (command variable) is maintained

This procedure also works well because the differential pressure Δ p _v can be sampled and measured in a short cycle of 5 ms, for example, and because the filling valve n+1 triggers a change in the differential pressure Δ p _v relatively slowly with the (sluggish) opening stroke.

In figure 10 an alternative design of the circuit is provided. A control valve 180 is provided in the filling product supply line 18, by means of which the common inflow to the separate individual filling valves 12 can be controlled. In the exemplary embodiment shown, the filling valves 12 are not designed as control valves, but as simple switching valves (open/closed).

Accordingly, the control behavior of the filling valves 12, which is achieved in the above-described embodiments via the filling valves designed as control valves, can be taken over in this embodiment by the one control valve 180 arranged in the filling product feed line 18.

It is thus possible to regulate the filling valves 12 using standardized control specifications without having to impose a pilot control behavior controlled by the number of open filling valves 12 .

The control valve 180 in the inlet 18 thus exhibits a behavior in which, at the start of production, it is initially adjusted to a lower conductance value K _V and then the first filling valve 12 is opened. The conductance K _V of the control valve 180 is then gradually increased synchronously with the increase in the number of open filling valves 12, so that each individual filling valve 12 sees the same differential pressure in principle.

In other words, the pressure drop Δp _inlet is varied by means of the control valve 180 in the inlet, so that Δp _valve can be kept constant.

Reference character list

1

filling product filling plant

10

filler carousel

11

loop line

12

filling valve

14

filling valve outlet

16

fill product reservoir

17

distribution line

18

filling product supply line

180

control valve

19

rotary distributor

Claims (6)

1. Method for filling a container with a filling product in a filling-product filling system having at least two filling valves (12) connected in parallel with one another as control valves, the method having the steps of:

   - determining a function of the volume flow q(t, Δp_v ) for at least two of the filling valves (12) connected in parallel depending on a differential pressure Δp_v decreasing across the filling valves (12) connected in parallel;

   - determining the differential pressure Δp_v decreasing across all filling valves (12) connected in parallel;

   - calculating the volume flow q(t, Δp_v ) through at least one of the filling valves (12) connected in parallel on the basis of the differential pressure Δp_v determined; and

   - regulation and/or control of the at least one filling valve (12) depending on the calculated volume flow q(t, Δp_v ), wherein the the regulation and/or control of the at least one filling valve (12) comprises compensation of the opening position of the filling valve (12) in the event of varying differential pressure Δp_v based on the calculated current volume flow q(t, Δp_v ).
2. Method according to claim 1, characterised in that the the regulation and/or control of the filling valve (12) comprises the adjustment of an opening position of the filling valve (12) based on the current volume flow q(t, Δp_v ).
3. Method according to claim 1 or 2, characterised in that the regulation and/or control of the at least one filling valve (12) is carried out taking into account a predetermined volume-flow profile for filling the container to be filled with the filling product.
4. Method according to any of claims 1 to 3, characterised in that the the regulation and/or control of the at least one filling valve (12) is carried out depending on the calculated volume flow q(t, Δp_v ) only at the start and/or end of the respective filling operation, preferably until a steady-state equilibrium is reached at simultaneously opened filling valves (12).
5. Method according to any of claims 1 to 4, characterised in that the the regulation and/or control of the at least one filling valve (12) is carried out depending on the calculated volume flow q(t, Δp_v ) only when the resulting regulation and/or control exceeds a predetermined threshold.
6. Method according to any of the preceding claims, characterised in that the function of the volume flow q(t, Δp_v ) is given depending on the differential pressure Δp_v by $$q\left(t,{\mathit{\Delta p}}_v\right)=\{\begin{array}{ll}{q}_{\infty }\cdot \mathit{tanh}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_h}\cdot t+\mathit{arctanh}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathrm{if}{q}_0<q_{\infty }\\ q_{\infty }\cdot \mathit{coth}\left(\frac{{\mathit{\Delta p}}_v}{q_{\infty }\cdot L_h}\cdot t+\mathit{arccoth}\left(\frac{q_0}{q_{\infty }}\right)\right),& \mathrm{if}{q}_0>q_{\infty }\\ q_0,& \mathrm{if}{q}_0=q_{\infty }\end{array}$$

   

   wherein $$q_{\infty }=K_v\cdot \sqrt{\frac{\mathit{\Delta p}}{1\mathit{bar}}\cdot \frac{1000\frac{\mathit{kg}}{m^3}}{\rho }}$$

   

   is the volume flow through the filling valve (12) in a steady state.

Images