EP3538476A1 - System and method for automated adjustment of a container filling machine - Google Patents

Abstract

The invention relates to a system (20) for the automatic adjustment of a container filling machine (1) comprising: a number of filling units (10) for filling containers (2) during a production phase , coupled to a rotary conveyor (4) adapted to rotate about an axis of rotation (A); and a machine control unit configured to handle filling operations by the filling units (10) according to a filling recipe. The system (20) is provided with a processing module (12, 28), coupled to an HMI module (27) and a storage database (29) for determining a new filling recipe, based on received user information via the HMI module (27) and known features of the filling machine (1) stored in the storage database (29); the processing module being further configured to control the execution of filling operations at a validation filling station (30) during a recipe validation phase preceding the production phase; receiving feedback information regarding measured parameters associated with the filling operations that are performed; and iteratively adjusting the new fill recipe until the measured parameters satisfy a desired relationship with desired fill parameters.

Description

SYSTEM AND METHOD FOR AUTOMATED ADJUSTMENT OF A CONTAINER FILLING MACHINE

The present invention relates to a system and method for automated adjustment of a container filling machine, designed to fill containers with a pourable product.

In general, the present solution may be used in machines for filling any type of containers, such as containers or bottles made of glass, plastics (PET) , aluminum, steel and composites, and with any type of pourable products filled in the containers, including carbonated liquids, such as sparkling water, soft drinks, beer; non-carbonated liquids, such as still water, juices, teas, sport drinks, wine; liquid cleaners, emulsions, suspensions, or high viscosity liquids.

As it is known, a container processing plant includes a number of processing machines cooperating for processing of containers, such as: a blowing machine, designed to blow heated preforms received from an oven into fully shaped containers; a filling machine, designed to fill the formed containers received from the blower, with a pourable product, e.g. a still or carbonated liquid; a capping machine, designed to close the filled containers with a cap; a labeling machine, designed to label the filled containers received from the filler; and a palletizer, designed to receive the labelled and filled containers from the labeler and to form organized pallets, ready for shipping and marketing.

In general, each processing machine comprises: a feed line for receiving a succession of containers to be processed; a processing line, typically comprising a rotating conveyor (so called "carousel") , carrying a number of processing units, each designed to perform processing operations on a respective container, the processing units mounted to rotate continuously about a longitudinal axis, carried by the carousel; and an output line, to transfer the processed containers towards a next-in^¬ line processing machine in the container processing plant. Generally, containers are transferred between the various machines of the processing plant via transfer wheels or conveyors .

Figure 1 schematically shows a filling machine (or "filler") 1 of a container processing plant, configured for filling containers 2, for example glass or PET bottles, with a filling fluid, for example a carbonated pourable food product, such as a CSD - Carbonated Soft Drink (it is however again underlined that other types of containers and carbonated, or non-carbonated, liquids may as well be envisaged) .

Filling machine 1 comprises a conveying device, including a rotating conveyor (or carousel) 4, which is mounted to rotate continuously (anticlockwise in Figure 1) about a substantially vertical longitudinal axis A.

The rotating conveyor 4 receives a succession of empty containers 2 (e.g. originating from a blower in the same processing plant) from an input wheel 5, which is coupled thereto at a first transfer station 6 and is mounted to rotate continuously about a respective vertical longitudinal axis B, parallel to axis A.

The rotating conveyor 4 releases a succession of filled containers 2 to an output wheel 8 (e.g. so as to be received by a labeling machine and/or a capping machine of the processing plant) , which is coupled thereto at a second transfer station 9 and is mounted to rotate continuously about a respective vertical longitudinal axis C, parallel to axes A and B.

Filling machine 1 comprises a number of filling units 10, which are equally spaced about axis A, are mounted along a peripheral edge of rotating conveyor 4, and are moved by the same rotating conveyor 4 along a path P extending about axis A and through transfer stations 6 and 9.

Each filling unit 10 is designed to receive at least one container 2 to be filled, and to perform, during its rotation along path P, filling operations according to a desired plan, the so called filling "recipe", in order to fill the container with a fluid (e.g. a carbonated liquid) . Each filling unit 10 generally includes one or more fluidic conduits and flow regulators (here not shown) , including solenoid valves that are designed to selectively couple the container to one or more feed devices, or product tanks (also not shown) , of the filling machine 1.

The filling recipe may for example envisage one or more of the following operations, e.g. performed in a preset sequence: a first air-extraction operation to remove air from the container 2 by fluidically coupling the container 2 to an air- suction device;

flushing the container 2, by feeding a stream of carbon dioxide to the container 2; a second air-extraction operation to remove any remaining air from the container 2, again by fluidically coupling the container 2 to the air-suction device (so called "snift" operation) ;

pressurizing the container 2, by feeding a stream of carbon dioxide into the container 2;

filling the container 2, by feeding a stream of filling liquid; and

depressurizing the container 2, by fluidically coupling the container 2 to the air-suction device.

The filling recipe defines the sequence of operations to be performed, the respective time length and, if required, the relevant operating parameters of the filling units 10.

In a manner not shown in detail, each filling unit 10 includes a main body, for example with a tubular configuration, having a vertical extension along a longitudinal axis that is substantially parallel to axis A of rotating conveyor 4, and mechanically coupled to the rotating conveyor 4. The main body includes, at a bottom portion thereof, a container receiving part, designed to releasably engage a neck 2' of a respective container 2 that is to be filled during the filling operations.

Operation of the filling units 10 is controlled by a machine control unit 12 (shown schematically) , generally including an industrial PLC (Programmable Logic Controller) or any other suitable digital processing unit, for example a computer running a PLC software application, designed to control general operation of the filling machine 1 according to the desired filling recipe, e.g. providing suitable control signals for the actuation of the flow regulators and the corresponding valves and actuators.

In particular, solutions are known, in which each filling unit 10 is provided with its own intelligence, i.e. with a respective local control unit 10' (shown schematically), e.g. a processor or similar computing unit, programmed to manage the filling operations according to the recipe and operatively coupled to the machine control unit 12.

Typically, expert personnel is needed to set-up a new recipe for the filling machine 1, at its startup, for example to determine the sequence of operating steps and the duration of each operating step envisaged by the recipe, and, if required, any other relevant operating parameter, such as the rotating speed of the rotating conveyor 4, the pressure and flow rate of the filling fluid, and so on.

It is moreover known that processing plants are currently required to frequently change the products or the container type and format in order to meet the rapidly changing consumer needs and market trends; a new filling recipe may be required, at each change of container or product type, so as to properly set-up the filling machine 1 for production.

The Applicant has realized that these set-up operations, however, require a considerable amount of time and may be prone to errors or lack of optimization and/or repeatability.

Moreover, the present Applicant has realized that external conditions may affect the filling operations, while the filling machine is running during production, and in particular may alter the quality of the product and the result of the filling operations. Expert personnel may thus be required to properly adjust the filling recipe in order to cope with the possible changes in the external conditions, again implying time and resource expenditure, the possibility of errors and a lack of optimization .

The aim of the present solution is to solve, at least in part, the problems previously highlighted, and in general to provide improved automated solutions for adjustment of a container filling machine.

According to the present solution, a system and a method are provided, as defined in the appended claims.

For a better understanding of the present invention, preferred embodiments thereof are now described, purely by way of non-limiting examples, with reference to the attached drawings, wherein:

- Figure 1 is a schematic representation of a filling machine of a container processing plant;

- Figure 2 is a diagrammatic representation of a system for automated adjustment of a container filling machine, according to an embodiment the present solution;

- Figure 3 is a diagrammatic representation of a further part of the automated system according to an aspect of the present solution; - Figure 4 is a flow chart of operations of a method for automated recipe calculation and auto-tuning implemented in the automated system;

- Figure 5 shows a plot of monitored filling parameters in the automated system;

- Figure 6 shows an image of a container being filled, captured in the automated system; and

- Figure 7 is a flow chart of operations of a method for automated recipe optimization implemented in the automated system.

In general, the present solution envisages automated adjustment of a filling machine in a processing plant, by means of:

automated calculation of a new, customized, filling recipe based on known features and algorithms relating to the filling machine and stored in a database, e.g. to deal with a new container format and/or a new product. This feature will ease primarily filling machine start-up, since the operator will have to enter only relevant container/product characteristics when a new recipe has to be defined; and

automated tuning of the calculated recipe, while a container is filled in a static manner (i.e. with the rotating conveyor of the filling machine being in a stand-still condition, before start of the production) , at a specific validation station, internal to the filling machine, or external thereto. Using a plurality of feedback information, such as from a smart (or intelligent) camera or other artificial vision means and further process-feedback sensors, such as pressure, flowrate and temperature sensors, the recipe parameters are automatically tuned to achieve a tuned filling recipe; in particular, an iterative action-correction loop is carried out, as long as measured properties are not compliant with desired or theoretical properties.

According to a further aspect, the present solution may further envisage:

- automated optimization of the filling recipe, according to which, by means of a plurality of feedback information, such as from pressure, flowrate and temperature sensors at the various filling units, the recipe parameters are automatically optimized to cope with external variations during actual production (i.e. in dynamic operating conditions, as opposed to the stand-still, or static, operating conditions of the automated tuning phase) .

The present solution is now discussed in more details, referring first to Figure 2, which shows the general architecture of a possible embodiment of a system 20 for automated adjustment of a container filling machine 1 (e.g. of the type shown and discussed in details with reference to Figure 1, thus denoted with the same reference number herein) .

As previously discussed, filling machine 1 includes a number of filling units 10, coupled to the rotating conveyor 4, each provided with a respective local control unit 10', configured to manage the filling operations being performed.

Each filling unit 10 is further provided with a number of feedback sensor elements 22, operatively coupled to the respective local control unit 10' and operable for monitoring the filling operations being performed on a respective container 2. In particular, feedback sensor elements 22 include one or more of: a pressure sensor 22a, designed to detect a pressure (or vacuum condition) within the container 2 during each step of the filling recipe; a flowmeter 22b, designed to detect the flow rate of the filling fluid; a temperature sensor 22c, designed to detect the temperature of the filling fluid during filling operations.

The presence of the feedback sensor elements 22 allows to collect relevant information for the analysis and monitoring of the filling operations, and to perform control actions and adjustments in real time during the same filling operations. In particular, feedback sensor elements 22 allow to perform a number of operations, among which:

- detecting pressure values during all the steps of the recipe, storing these values, and, e.g. after the filling operations are completed, draw pressure plots, which may be displayed and processed, in order to allow analysis of the performance of the filling operations;

- monitoring the occurrence of failures, such as bursting of the container 2, during execution of the recipe steps, e.g. by checking if the sensed pressure values overcome a preset absolute maximum pressure threshold, or a preset differential pressure threshold, relating to the difference of pressure between the container 2 and a respective feed tank;

- monitoring the correctness of the vacuum level during the recipe steps envisaging vacuum within the containers 2;

- monitoring the correct depressurization of the containers 2 during the recipe steps envisaging depressurization (e.g. during a snift operation) . In this case, the sensed pressure values may be compared to a decompression pressure threshold, and decompression is considered successful only if the sensed pressure is found to be below the decompression pressure threshold at the end of the recipe step.

Local control units 10' are communicatively coupled to the machine control unit 12 of the filling machine 1, e.g. via a data communication bus 23, so as to receive control signals from the machine control unit 12 and provide feedback signals to the same machine control unit 12, while filling operations are performed. Data communication bus 23 may be a real-time bus, in particular an Ethernet-based real-time communication bus, such as the Powerlink bus, Ethercat, Ethernet Realtime, or Profinet, or any other bus capable to offer real-time communication capability (e.g. an optical-fiber based bus) . In the context of the present application, real time data communication denotes the possibility to obtain very fast data refresh values, e.g. lower than five milliseconds.

Local control units 10' may be coupled to the machine control unit 12 according to a master/slave operating relation (the machine control unit 12 of the filling machine 1 acting as "master" and the local control units 10' of the filling machine 1 acting as "slaves") .

The machine control unit 12 is moreover coupled to a central supervising unit 25, including a respective PLC or any other suitable computing and processing unit, e.g. located remotely with respect to the filling machine 1, via a cabled or remote wireless link; central supervising unit 25 may supervise and manage operation of various processing machines in the processing plant, in addition to filling machine 1 (e.g. a container blowing machine, a labelling machine, a capping machine and so on, all cooperating in processing of the containers 2), and may receive process feedbacks from the various machines of the processing plant (as schematically shown) .

Moreover, the machine control unit 12 (and also the central supervising unit 25, in the embodiment shown) are operatively coupled to a HMI (Human Machine Interface) unit 27, via a communication link. HMI unit 27 includes a display device, in order to display a user interface 27' for visualization by a user; HMI unit 27 also includes suitable input devices to allow the user to input data and perform actions in the user interface 27'; advantageously, HMI unit 27 implements a touch-screen display.

According to an aspect of the present solution, machine control unit 12 is provided with a non-volatile memory 28, storing software instructions to implement a method for automated adjustment of the filling machine 1, as will be discussed in details in the following.

Moreover, a storage database 29 is coupled to the same machine control unit 12, and stores know-how data relating to the filling machine 1 and filling units 10, assisting the above method for automated adjustment of the filling machine 1; the storage database 29 may be located internally of, or externally to, the machine control unit 12, e.g. in the form of a flash memory, or hard disk, or other data storage means.

According to a particular aspect of the present solution, automated system 20 envisages a validation station 30 for implementing the above discussed automated tuning of the calculated recipe, during which a container 2 is filled in a static manner (i.e. with the rotating conveyor 4 of the filling machine 1 being still) .

In a possible embodiment, as shown in Figure 2, validation station 30 is implemented internally to the filling machine 1, by a specific one of the filling units 10 thereof.

In particular, at the specific filling unit 10 implementing the validation station 30, a smart (or intelligent) camera 32, or other artificial imaging means, are provided, configured to capture images of the container 2 being filled, e.g. of a neck or bottom area thereof. The smart camera 32 is controlled by, and operatively coupled to, the machine control unit 12, so as to provide acquired image data and receive control signals (e.g. trigger signals) . An illumination unit 33 is further provided at the validation station 30, to provide a source of back-light illumination of the container 2 being filled and to assist the image capturing operations.

As schematically shown in Figure 3, a further smart camera 34 (or other artificial imaging means) is provided at the output of the rotating conveyor 4 of the filling machine 1, configured to image the filled containers 2, exiting the same rotating conveyor 4 towards the output wheel 8 and the further processing machines of the container-processing plant.

Referring now to Figure 4, the recipe calculation and auto- tuning (validation) for the automated adjustment of the filling machine 1 are now discussed in more details.

In a first step 40, an operator of the filling machine 1 enters, via the HMI unit 27, input information relating to the recipe to be performed, in particular including container and product characteristics, among which (purely by way of example) : format of the container 2; material of the container 2; type of the product (e.g. still or carbonated); density of the product; temperature of the product; CO2 content of the product. User may also input the speed of the filling machine 1 and fluid dynamics characteristics and/or other features of the filling units 10.

Based on the input information, the automated system 20 is configured to automatically determine, at step 42, a starting filling recipe. As previously discussed, filling recipe includes the relevant parameters of the filling process, for example the sequence of the filling steps to be performed (e.g. pressurization, filling, depressurization) and the duration of each filling step; the filling recipe may include further parameters (such as the rotating speed of the rotating conveyor 4, the flow rate and pressure of the filling fluid, and so on, depending on the particular application) .

This calculation by the automated system 20 is based on the know-how data stored in the storage database 29, relating to the known physics of the filling process and the known characteristics of the filling machine 1 and filling units 10, among which (purely by way of example) : the basement type; the number of filling units 10; the operating characteristics of the same filling units 10, such as the flow coefficient K_v of the respective valves; the equilibrium pressure; and so on.

In particular, suitable calculation algorithms may be implemented to calculate a customized recipe, suitable for the input information relating to the product and container. As an alternative, a known recipe, stored in the storage database 29 may be retrieved based on the input information, which may represent a suitable starting point for the processing of the containers, possibly being modified according to the same input information .

In any case, advantageously, start-up of the filling machine 1 is eased with respect to traditional solutions, since it does not require intervention of trained personnel, the starting recipe being automatically determined. The operator has in this case only to input simple information about the recipe that has to be implemented, without requiring any substantial skill or experience.

Afterwards, at step 44, the automated system 20 envisages carrying out container filling operations based on the determined starting recipe, at the validation station 30, with the rotating conveyor 4 at stand-still (i.e. with null rotation speed of the same rotating conveyor 4) . In particular, filling operations at the validation station 30 may be manually started and managed by the operator, via the HMI unit 27.

Performance of the filling operations is then evaluated, at step 45, at each operating phase of the starting filling recipe, by means of the feedback sensor elements 22 and moreover by means of the smart camera 32, which acquires images of the container 2 being filled. In particular, actual parameters of the filling process, including temperature, pressure, CO2 content and flowrate, are detected to provide feedback information to the automated system 20, for correction and auto- tuning of the filling recipe.

At step 46, the recipe is automatically corrected based on the detected information and the difference with respect to a desired filling performance; the measured parameters may conveniently be compared with corresponding predefined theoretical values. Know-how algorithms and strategies (that may also be stored in the storage database 29) may be used for tuning of the filling recipe.

In more details, as shown in Figure 5, in a possible embodiment the system 20 is configured to evaluate the pressure (P) and flowrate (F) profiles determined by the feedback sensor elements 22 during a filling cycle, and in particular one or more of the following parameters: the flow rate value F(t); minimum and maximum pressure values ΡΜΙΠ, PMHX (after a depressurization or, respectively, a pressurization phase of the container 2) ; pressure variation dP/dt (and its comparison with at least a variation coefficient ki) ; duration of the time intervals ti and t2, corresponding to the rising portion of the pressure profile after depressurization and the subsequent flat portion of the same pressure profile.

Moreover, as shown in Figure 6, the automated system 20 may be configured to process and evaluate the images acquired by the smart camera 32, for example to evaluate the thickness of the foam formed at the neck 2 ' of the container 2 during or after the filling operations, and in particular the total resulting thickness hi of the foam and the height 12 of the void space between the foam top level and the top of the container (other parameters may also be evaluated via processing of the acquired images, such as the entry angle of the product within the same container 2 ) .

The smart camera 32 may advantageously operate in the Near InfraRed (NIR) interval of the electromagnetic spectrum (e.g. at a wavelength of 850 nm) , in which visibility is maximized through PET and other fluid filling products. The Applicant has indeed realized that, using NIR light, it is possible to see through PET containers, irrespective of the color thereof (i.e. from the containers being transparent up to containers being black) . Moreover, use of backlight illumination by means of the illumination unit 33 may be advantageous to improve differentiating the foam from the liquid in the container 2.

The smart camera 32 may be configured to process the acquired images and determine the relevant information, such as the thickness hi of the foam and the height 12 of the void space between the foam top level and the top of the container, and to transmit the determined information to the machine control unit 12 of the filling machine 1.

In particular, and returning now to the flow-chart of Figure 4, the automated system 20 is configured to perform an iterative correction process of the filling recipe, therefore iteratively repeating steps 44-46 as long as the actual parameters are not compliant to the desired, predefined, theoretical properties and until it is determined that the filling recipe achieves the desired filling performance.

Various algorithms and methodologies may be implemented to perform this iterative correction; for example, statistical analysis and correlations (e.g. with respect to average values of the detected pressure and flowrate parameters), or machine- learning algorithms may be implemented.

Afterwards, as shown in the same Figure 4 at step 48, e.g. when each measured parameter is determined to be compliant with the corresponding predefined theoretical values, the tuned recipe is finalized to start the production and filling of the containers 2 in dynamic condition (i.e. with the rotating conveyor 4 rotating at the desired production speed) , by all the filling units 10 of the filling machine 1.

The finalized recipe may conveniently be stored in the storage database 29, for future use in subsequent filling operations, possibly after being tuned each time to cope with different operating requirements of the filling machine 1.

HMI unit 27 may conveniently be configured to signal to the user, with suitable notifications on the corresponding display, that the recipe has been finalized and is ready for start of the production.

Referring now to Figure 7, the optimization of the filling recipe during production, performed by the automated system 20, is now discussed in more details.

As previously discussed, optimization allows to automatically adjust the filling recipe to maintain standard compliance to the desired filling performance, even in the presence of changes in the operating conditions of the filling machine 1. For example, optimization allows to cope with modifications of one or more of: product temperature in the tank; CO2 purity; product mixing; pressure profile.

In detail, optimization is performed during dynamic operation of the filling machine 1, as shown at step 50. Filling operations are evaluated, at step 51, by means of the feedback sensor elements 22, at each filling unit 10, and moreover by means of the further smart camera 34 provided at the output of the rotating conveyor 4, which acquires images of the filled container 2 exiting the same filling machine 1.

In particular, in a substantially corresponding manner as discussed in connection with the static automated tuning of the filling recipe, detected information about pressure, flowrate, temperature and possibly further parameters (such as CO2 content) are provided to the automated system 20, for optimization of the filling recipe. Moreover, images of the filled containers 2 are acquired and processed by the further smart camera 34, in order to identify possible defects, such as excessive foam formation, spilling of the product, and similar.

At step 52, the recipe is automatically corrected based on the feedback information and the difference with respect to desired performance of the filling recipe. Suitable algorithms and strategies (stored in the storage database 29), for example statistical or machine-learning algorithms, may again be implemented by the automated system 20 (in the example at the machine control unit 12) for correction of the filling recipe.

In particular, the automated system 20 is configured to perform a continuous optimization of the filling recipe, therefore executing steps 50-52, until the end of the filling operations, thus continuously coping with possible external variation during production. HMI unit 27 may conveniently be configured to signal to the user, with suitable notifications on the corresponding display, that the recipe has been optimized to cope with one or more variations in the operating conditions.

The advantages that the described solution allows to achieve are clear from the foregoing description.

In particular, it is again underlined that the proposed solution allows to automatically calculate a new, customized, filling recipe, e.g. for a new product and/or a new container, and validate the same filling recipe in a stand-still condition, to automatically set the best filing parameters.

The solution therefore allows to minimize the skill and experience required to the operator in order to set up a new production; HMI unit 27 also guides the operator to create a new filling recipe, thus further improving ease of use of the filling machine 1, in particular with respect to setting-up of a new filling procedure.

Moreover, the discussed solution allows to optimize the filling recipe dynamically during production, and cope with possible variations of the external conditions (that are not controlled by the filling machine 1), again in an automated manner, without requiring intervention by expert personnel.

Therefore, downtime is reduced, as well as non-conformity of the resulting products, efficiency and flexibility are increased and costs and wastes reduced.

Finally, it is clear that modifications and variations may be applied to the solution described and shown, without departing from the scope of the appended claims.

In particular, it is underlined that the discussed system 20 for automated adjustment of the container filling machine 1 envisages a processing module that may be implemented in the machine control unit 12 of the same filling machine 1, as in the example shown and previously discussed, or by means of any other processing unit, for example the central supervising unit 25 of the processing plant or a further, dedicated, processing unit properly arranged at the filling machine 1 or externally thereto .

Claims

1. A system (20), for automated adjustment of a container filling machine (1) comprising a number of filling units (10) for filling of containers (2) during a production phase, coupled to a rotating conveyor (4) designed to rotate around a rotation axis (A) , and a control unit designed to manage filling operations by the filling units (10) according to a filling recipe,

wherein the system (20) comprises a processing module (12, 28), coupled to a HMI module (27) and to a storage database (29) and configured to determine a new filling recipe, based on user input information received via the HMI module (27) and known characteristics of said filling machine (1) stored in the storage database (29),

the processing module being further configured to: control execution of filling operations at a validation filling station (30), during a recipe-validation phase preceding the production phase; receive feedback information about measured parameters related to the filling operations being performed; and iteratively adjust the new filling recipe until the measured parameters satisfy a desired relation with desired filling parameters .

2. The system according to claim 1, wherein the rotating conveyor (4) is maintained in a stand-still condition during the recipe-validation phase.

3. The system according to claim 1 or 2, wherein the validation filling station (30) is implemented at one of the filling units (10) of the filling machine (1) .

4. The system according to any of the preceding claims, wherein the validation filling station (30) includes feedback sensor elements (22) coupled to the processing module (12, 28) and configured to provide the feedback information about the measured parameters; the feedback sensor elements (22) comprising one or more of: a pressure sensor (22a), designed to detect the pressure within a container (2) during execution of the filling recipe; a flowmeter (22b) , designed to detect the flow rate of the filling fluid; a temperature sensor (22c) , designed to detect the temperature of the filling fluid.

5. The system according to claim 4, wherein the measured parameters comprise a pressure (P) and a flowrate (F) of the filling fluid during filling operations.

6. The system according to claim 4 or 5, wherein the feedback sensor elements (22) further comprise an imaging sensor (32), coupled to the processing module (12, 28) and configured to capture images of the container (2) being filled.

7. The system according to claim 6, wherein the measured parameters comprise a height (hi) of a foam formed at a neck portion (2') of the container (2) .

8. The system according to any of the preceding claims, wherein the processing module (12, 28) is further configured to: evaluate the filling operations being performed during the production phase using the recipe validated after the recipe- validation phase, in dynamic conditions, with the rotating conveyor (4) rotating around its axis (A) at a production speed; and provide adjustments to the filling recipe to optimize the filling recipe with respect to external variations occurring during the production phase.

9. The system according to claim 8, wherein each filling unit (10) is provided with respective feedback sensor elements (22), coupled to the processing module (12, 28) to provide feedback information about measured parameters relating to the filling operations being performed, based on which the processing module (12, 28) is configured to perform the adjustments to the filling recipe.

10. The system according to claim 9, wherein the measured parameters comprise at least one of a pressure (P) and a flowrate (F) of the filling fluid during filling operations.

11. The system according to claim 9 or 10, further comprising a camera (34), arranged at an output of said rotating conveyor (4), configured to provide images of the filled containers (2) to the processing module (12, 28), based on which the processing module (12, 28) is configured to perform the adjustments to the filling recipe.

12. The system according to any of the preceding claims, wherein the processing module (12, 28) is implemented in the control unit (12) of the filling machine (1) .

13. A container filling machine (1), provided with the system (20) according to any of the preceding claims.

14. A method for automated adjustment of a container filling machine (1) comprising: a number of filling units (10) for filling of containers (2) during a production phase, coupled to a rotating conveyor (4) designed to rotate around a rotation axis (A) ; and a control unit designed to manage filling operations by the filling units (10) according to a filling recipe,

the method comprising determining a new filling recipe, based on user input information received via a HMI module (27) and known characteristics of said filling machine (1) stored in a storage database (29),

wherein the method further comprises:

controlling execution of filling operations at a validation filling station (30), during a recipe-validation phase preceding the production phase;

receiving feedback information about measured parameters related to the filling operations being performed; and

iteratively adjusting the new filling recipe until the measured parameters satisfy a desired relation with desired filling parameters.

15. A computer program, configured to implement, when run on a processing module (12, 28), the method according to claim 14.