Web tension adjustment in a food packaging system based on reinforcement learning

Abstract

Methods and apparatus for controlling web tension in a food packaging machine (100) are described, including a computer program product, the food packaging machine comprising a plurality of subsystems. One or more local variable values (116) are received indicative of measurements by the food packaging machine (100) of one or more physical parameters of a web tensioning subsystem (200). One or more remote variable values (204) are received indicative of measurements by the food packaging machine (100) of one or more physical parameters of one or more remote subsystems. One or more control parameter values are determined for the web tensioning subsystem (200) by processing the remote and local variable values (204, 116) using a reinforcement learning model (206) and a local control model (210). One or more control parameters of the web tensioning subsystem (200) are adjusted in accordance with the determined control parameter values.

Description

Technical Field

[0001] This invention relates to food packaging systems, and more particularly to adjusting web tension in food packaging systems. Background Technology

[0002] Today, automated control systems are widely used in manufacturing and processing environments, and their complexity is constantly increasing. A common approach to managing this complexity is to divide the system into subsystems and develop appropriate control mechanisms for each subsystem. However, this approach does not always provide the optimal solution for the entire system.

[0003] As systems become increasingly complex and the number of influencing factors grows, capturing these factors from different sources becomes increasingly difficult. This complexity increases further when the relationships between influencing factors, control variables, and the system itself are non-linear and / or difficult to model.

[0004] The levels of abstraction in industrial control can be divided into two main perspectives: low-level control and high-level control. Low-level control refers to the management of individual automation components (e.g., actuators, servo motors, heaters, and many other devices). The level of abstraction in high-level control increases progressively, from the subsystem level to the system level, and further to the coordination of the entire plant with multiple systems and subsystems that need to operate collaboratively.

[0005] As an example, food processing and packaging equipment typically comprises several subsystems, such as filling systems, sterilization systems, and packaging folding systems. Each subsystem contains multiple different components (e.g., pneumatic actuators, servo motors, DC motors, AC motors, sensors, other actuators, etc.). These individual components are typically controlled by a low-level local control system that utilizes conventional control techniques such as proportional-integral-derivative (PID) controllers to control the target variable. Feedback loops are used to keep the controller's error relative to the target operating point of the component, system, or subsystem low.

[0006] However, PID controllers require tuning for their specific applications and are typically optimized for a specific operating range and dynamics. They are also not well-suited to adapting to unforeseen circumstances or operating conditions outside their normal operating range. When such conditions change (e.g., different operating environments, changes in automation components, changes in manufacturing processes, etc.), the parameters of the PID controller usually need to be adjusted and recalibrated. This can be a time-consuming and complex process, requiring significant manual input from experienced personnel, especially when a large number of components and / or subsystems are involved, as is often the case in food processing and packaging plants.

[0007] A filling machine is an example of a complex system that packages liquid, semi-liquid, or pourable foods, such as juice, UHT (ultra-high temperature) milk, wine, and tomato sauce, into composite packages made of multi-layered composite packaging materials for distribution and sale. A typical example is what is called a Tetra Pak aseptic pack. TM This refers to parallelepiped-shaped packaging for pourable food products, manufactured by sealing and folding laminated sheet packaging material. The packaging material has a multi-layered structure comprising cardboard and / or a paper base layer, both sides of which are covered with heat-sealable plastic material layers (e.g., polyethylene layers). For aseptic packaging of products intended for long-term storage, the packaging material also includes an oxygen barrier layer (e.g., aluminum foil) stacked on top of the heat-sealable plastic material layer, which is then covered by another heat-sealable plastic material layer, forming the inner surface of the packaging that ultimately contacts the food.

[0008] The filling machine begins with a web of multi-layer composite packaging material (wound from a spool). The web is fed through the filling machine, where it is formed into a tube by creating a longitudinal seal. Liquid food is fed into the tube through the tube; then the lower end of the tube is fed into a folding device, where a transverse seal is created, and the tube is folded according to a folding line, also known as a weakening line. The tube is then cut, thus forming a composite package filled with liquid food.

[0009] The machine module or subsystem responsible for package formation, lateral sealing, and cutting is called the "gripper system" and consists of a pair of grippers. The synchronized movement of these gripper pairs allows the tube of packaging material to be pulled down and the filled package to be completely sealed. The gripper system is an important component of the filling machine because the coordinated movement of the two gripper pairs is responsible not only for the correct shaping of the package but also for pulling the roll of material through the machine.

[0010] As the gripper system pulls the packaging material, it does so intermittently, rather than continuously, as individual packages are produced. This intermittent pulling of the roll occurs at the end of the packaging machine. Meanwhile, the packaging material is positioned on a large roller at the beginning of the machine. Therefore, the intermittent pulling of the packaging material by the gripper system generates varying forces on the material, partly due to the inertia of the roller and partly due to the material itself. This can be problematic because certain parts of the packaging machine require the roll to move at a constant speed. Furthermore, too much tension in the roll can lead to packaging integrity issues. Moreover, since it pulls the roll down, the tension can be affected by other factors, such as the length of the roll through the packaging machine from start to finish, or the quality of the product being filled into the packaging material tube. The purpose of the roll tensioning system is to maintain the packaging material roll at a tension appropriate to the packaging process in the filling machine. A slack roll results in poor performance, while an excessively taut roll introduces defects and damage into the packaging. Currently, machine technicians typically rely on manual configuration (including trial and error) to obtain appropriate roll tension in roll tensioning systems. Furthermore, there is currently no way to account for the influence of factors such as gripper system movement trajectory, filling conditions (e.g., filler flow rate and product level), and packaging material characteristics (e.g., thickness, mechanical properties) on roll tension.

[0011] Therefore, there is a need to improve the technology used to control roll tension, while taking into account a range of events that may occur in the packaging machine that could affect roll tension, in order to always maintain appropriate roll tension. Summary of the Invention

[0012] One object of the present invention is to overcome, at least in part, one or more limitations of the prior art. Specifically, one object is to provide a method and system that can control the roll tension of a food packaging machine by taking into account parameter values ​​measured not only for the local roll tension subsystem in the food packaging machine but also for other remote subsystems. Therefore, appropriate roll tension can be achieved, which can both speed up the setup process when initially configuring the food packaging machine and enable a more reliable manufacturing process with fewer packages to be discarded.

[0013] In one aspect of the invention, this is achieved by a method for controlling the tension of a roll material in a food packaging machine, wherein the food packaging machine includes multiple subsystems. The method includes:

[0014] • Receive one or more local variable values, which indicate the food packaging machine's measurement of one or more physical parameters of the roll tension subsystem;

[0015] • Receive one or more remote variable values, the one or more remote variable values ​​indicating the food packaging machine's measurement of one or more physical parameters of one or more remote subsystems;

[0016] • By processing the remote variable values ​​and the local variable values ​​using a reinforcement learning model and a local control model, one or more control parameter values ​​are determined for the roll material tensioning subsystem; and

[0017] • Adjust one or more control parameters of the roll tensioning subsystem according to the determined control parameter values.

[0018] Utilizing local variables and inputs from remote subsystems results in more precise control of roll tension and greater resilience in the event of unforeseen events in the food packaging machine. This leads to less waste in packaging (and food), making the food packaging machine more efficient and environmentally friendly to operate. Given the improved control over the packaging formation process, time to market for new products and / or configurations may also be shortened due to the need for less manual testing. This is further enhanced by the ability to learn control strategies in a simulated environment, eliminating the need for manual configuration of the food packaging machine "from scratch."

[0019] In one embodiment, the reinforcement learning model is a deep reinforcement learning model that includes a neural network. Deep reinforcement learning is particularly useful for evolving control strategies for subsystems that must consider a large number of variables (the internal relationships between these variables and their effects on the subsystem may be unknown), and provides a superior method for determining one or more control parameter values ​​for the local roll tensioning subsystem of a food packaging machine compared to what can be achieved using conventional reinforcement learning without neural networks.

[0020] In one embodiment, the roll tensioning subsystem includes two fixed guide rollers and a movable guide roller. By including the movable guide roller, the distance between the movable guide roller and each fixed guide roller can be varied, thereby generating force on the roll when the guide roller moves (typically by using a servo motor), thus creating a simple method for adjusting the roll tension while keeping other parameters of the roll (e.g., speed) constant.

[0021] In one embodiment, the movable guide roller is positioned between two fixed guide rollers along the path of the roll through the packaging machine and is movable to increase or decrease the tension of the roll in response to instructions received for control parameter values. This achieves similar advantages to those just described and also ensures that any variation in roll tension is evenly distributed across the two fixed guide rollers.

[0022] In one embodiment, the neural network is a convolutional neural network, a recurrent neural network, a long short-term memory neural network, or a fully connected neural network. These are all different types of neural networks well known to those skilled in the art, and therefore easier to incorporate into existing food packaging machine setups.

[0023] In one embodiment, one or more local variable values ​​include measurements related to the roll tension setpoint or the current position of the roll tensioning system, and one or more remote variable values ​​include measurements related to roll movement control variables, gripper system motion trajectory, packaging material characteristics, or filling status. As achieved by the data-driven methods of the various embodiments described herein, improved control over these parameters significantly enhances the operation of the roll tensioning subsystem, thereby enhancing the overall operation of the packaging machine.

[0024] Other aspects of the invention include systems and computer programs for controlling roll tension in food packaging machines. The features and advantages of these aspects of the invention are substantially the same as those discussed above with respect to the method.

[0025] Other objects, features, aspects and advantages of the invention will become apparent from the following detailed description and the accompanying drawings. Attached Figure Description

[0026] Embodiments of the invention will now be described by way of example with reference to the accompanying schematic diagrams.

[0027] Figure 1 This is a schematic diagram of a portion of a food packaging machine according to one embodiment.

[0028] Figure 2 This is a schematic diagram of a controller in a food packaging machine according to one embodiment. Detailed Implementation

[0029] As described above, the objective of the various embodiments of the present invention is to provide improved control technology for apparatuses and systems relating to food processing and packaging, particularly those related to roll tension. Having appropriate roll tension is important not only from the perspective of machine packaging operation but also from a functional perspective, as inappropriate roll tension can lead to problems with the integrity of the packaging formed by the packaging machine. By applying general concepts of reinforcement learning and / or deep reinforcement learning techniques to control the roll tension system of a food packaging machine, a wider range of factors can be considered compared to those that might be considered in existing systems, and the roll tension can be adjusted very precisely, thereby improving the operability of the food packaging machine and ensuring the integrity, shape, and appearance of the packaging.

[0030] Both reinforcement learning and deep reinforcement learning are examples of machine learning techniques. Generally, reinforcement learning (RL) can be characterized as dynamic learning using either positive or negative rewards. System performance is evaluated based on a desired objective. A positive reward is provided if the objective is achieved, and a negative reward is provided if it is not achieved. As positive and negative rewards accumulate over time, the RL model evolves a control policy for the system, with the goal of maximizing the outcome. Deep reinforcement learning (DRL) can be characterized as an enhancement to RL, where RL is used in conjunction with a neural network as the control policy for the system evolves.

[0031] In the context of food processing and packaging, RL (i.e., agent-environment interaction) can be used to evolve control strategies for food processing and / or packaging machines. DRL (i.e., RL and neural networks) is particularly useful when evolving control strategies for subsystems (e.g., filling subsystems) that must consider a large number of variables whose internal relationships and effects on the subsystem may be unknown. Furthermore, it should be noted that RL and DRL techniques can also be used to improve existing local control techniques, essentially “filling” the “gaps” in traditional control techniques with this data-driven approach. Thus, a DRL algorithm can then directly control actuators (e.g., servo motors, pneumatic actuators, or other actuators) (or indirectly control actuators through other control layers, such as by adjusting the gain of a traditional PID controller to allow the PID controller to operate more efficiently than traditional control techniques) to adjust the roll tension that may occur in a food packaging system.

[0032] To further illustrate these principles, an example will be given of controlling a roll tension subsystem in a food packaging machine to ensure proper roll tension throughout the machine, and various embodiments of the invention will be described more fully with reference to the accompanying drawings (which illustrate some, but not all, embodiments of the invention). The invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

[0033] As mentioned above, the roll tensioning subsystem is an important component of food packaging machines, and its operation requires careful control to ensure that the roll moves smoothly at the required speed throughout the food packaging machine, regardless of the intermittent operation of the aforementioned gripper system, in order to ensure proper packaging integrity, shaping, and appearance.

[0034] Figure 1A schematic diagram of a sub-section of a food packaging machine 100 is shown, wherein a roll 102 of packaging material (preferably including at least one sealable surface) is fed forward through a roll feeder in an S-shaped pattern by guide rollers 106, 108, 110 and forms a tube 112. The longitudinally overlapping side edges of the roll 102 are sealed to close the tube along the longitudinal edges. These side edges may overlap with their lower sides against each other, or with their lower sides facing the same direction. A tape may also be provided along one or both longitudinal edges to aid in tube formation.

[0035] After the roll is formed into a tube, the food is fed into the formed tube at a filling station (not shown). In this context, food refers to anything ingested, eaten, and / or drunk by humans or animals, or absorbed by plants, including but not limited to liquid foods, semi-liquid foods, viscous foods, dry foods, powdered foods, and solid foods, beverage products, and water. For the avoidance of doubt, food also includes ingredients used to prepare the food. Some examples of food include milk, water, and fruit juice. The filled tube is then conveyed to a gripper system that laterally seals the filled tube and laterally cuts the sealed tube along its length within the laterally sealed area to form individual packages filled with the product. Thus, the gripper system generates a pulling force on the roll 102 in the forward direction. As described above, a corresponding counterforce is generated on the roll 102 due to the inertia of the large roller holding the roll 102 from the starting end of the packaging machine.

[0036] Figure 1 The right side shows a more detailed view of the roll tensioning subsystem 200 of the food packaging machine 100. In the embodiment described below, to adjust the tension on the roll 102, the intermediate guide roller 108 can move vertically, thereby allowing the tension on the roll 102 to be reduced during those times when the gripper system actively pulls the roll 102 in the forward direction, and the tension on the roll 102 to be increased during those times when the gripper system does not actively pull the roll 102. It should be noted that... Figure 1 Only one possible embodiment of the roll tensioning subsystem 200 is shown; other embodiments may have a greater number of guide rollers, multiple of which are movable. Similarly, the reference to the vertical movement of the intermediate guide roller 108 is for illustrative purposes only. In other embodiments, with... Figure 1 Compared to the example shown, the guide roller can rotate, for example, 90 degrees, so that the movement of the intermediate guide roller 108 is changed to a left-right movement. Therefore, many variations can be conceived by those skilled in the art. Due to its movement, the intermediate guide roller 108 is often also referred to as a swing roller.

[0037] exist Figure 1In the illustrated embodiment, the movement of the intermediate guide roller 108 is performed in response to a signal from the controller 114. The controller 114 receives input 116 from a sensor in the roll tension subsystem 200 that measures the current tension of the roll.

[0038] In addition, the controller receives input from one or more remote subsystems of the food packaging machine 100, which may experience events that also affect the operation of the roll tensioning subsystem 200. Some examples of such events may include splicing events (i.e., when the tail end of the packaging roll on the used packaging roll roller at the beginning of the food packaging machine is joined to the front end of the packaging roll on the new packaging roll roller to form a continuous packaging roll, thereby producing a roll portion with two layers of thickness instead of a single layer of thickness); acceleration, deceleration, or stopping of the roll 102 due to gripper movement or other reasons; packaging material change; product change; packaging fill status; roll length, etc.

[0039] These events can be represented by a set of variables, the values ​​of which indicate various states at different subsystems of the food packaging machine. This is in Figure 2 The diagram illustrates that, Figure 2 This demonstrates how input 116 from a local sensor in the roll tension subsystem 200, along with input values ​​204 from other subsystems of the food packaging machine, are fed into the controller 114.

[0040] In one embodiment, some examples of variables representing physical parameters from the local roll tensioning subsystem 200 include: • Roll tension set point (i.e., the desired roll tension for a specific type of roll being used in a food packaging machine). • Current position of the roll tensioning system (e.g., represented by the physical displacement of the movable guide roller from its intermediate position).

[0041] In one embodiment, some examples of variables from other subsystems include: • Roll material movement and control variables, represented (e.g., detected splice or package dimensions, speed, etc.) • The movement trajectory of the gripper system (e.g., the frequency and force with which the grippers pull the roll material). • Packaging material characteristic variables (e.g., packaging material stiffness, presence of a seal, packaging volume, roll length, etc.) • Filling status (e.g., fill flow rate, product level, etc.)

[0042] It should be recognized that these are merely a few examples of possible influencing factors from other subsystems and should not be considered an exhaustive list. However, they do represent influencing factors that conventional roll tension control systems cannot take into account, as it is difficult or impossible to determine how the various possible combinations of these factors should affect the operation of the roll tension subsystem 200.

[0043] According to the various embodiments described herein, controller 114 uses local control model 210 to process local subsystem input variables 116 and combines it with reinforcement learning model 206 to process input values ​​from other subsystems to determine how all measured variables as a whole collectively affect the operation of the roll tensioning subsystem 200. Local control model 206 may be an algorithm executed by a PID controller. Reinforcement learning model 206 may be a deep reinforcement learning model comprising one or more neural networks, as described above. In some embodiments, local subsystem input variables 116 may be processed by reinforcement learning model 206. In some embodiments, reinforcement learning model 206 may be used to calculate how different combinations of local and remote variables should affect the roll tensioning subsystem and use this perspective to improve local control model 210. Based on the processing and determination results, controller 114 generates a set of output control signals 208 for the local roll tensioning subsystem 200, which control the position of intermediate guide roller 108 to achieve appropriate roll tension.

[0044] Examples of neural networks that can be used in embodiments employing deep reinforcement learning model 206 include, for example, convolutional neural networks (CNNs) trained using reinforcement learning and deep reinforcement learning, recurrent neural networks (RNNs), long short-term memory (LSTM) neural networks such as those commonly used in the field of deep learning, or fully connected neural networks. LSTM networks can be particularly useful because, unlike standard feedforward neural networks, LSTMs have feedback connections. This allows LSTMs to process not only single data points but also entire sequences of data, which can be particularly useful in the context of designing food packaging machines for producing large quantities of packages.

[0045] Therefore, if the tension of the roll 102 changes, for example, due to variations in the movement of the grippers, or due to inaccurate function of one or more mechanical components of the filling machine, the data-driven method allows the controller 114 to detect such changes in roll tension and adjust the position of the intermediate guide roller 108 to always maintain appropriate roll tension, thereby avoiding potential damage to the packaging material and ensuring sealing and forming quality. Furthermore, conventional control techniques typically require manual calibration for each different operating setting. In contrast, this embodiment of the invention allows for the provision of a training environment, enabling the controller 114 to learn the optimal control strategy given the objectives of the roll tension subsystem 200. This can save significant time setting up the packaging machine, thereby also shortening the time to market for new packaging and products. Moreover, in some embodiments, the output from the reinforcement learning model can be used to adjust the gain of a conventional PID controller, making it more efficient than conventional control techniques that rely solely on local variable values. Therefore, embodiments of the invention can still be advantageous even when the only means of controlling the roll tension subsystem 200 is a PID controller.

[0046] It should be noted that even though a subsystem has been referred to above as a roll tensioning system, filling system, sterilization system, packaging folding system, etc., it can also refer to a part of the aforementioned subsystem, system, or individual component.

[0047] It should be noted that in some embodiments, the control model of the controller 140 may reside within the controller 140 itself, such as... Figure 1 As shown. In other embodiments, they may reside in and be operated by external hardware / software (e.g., an external computer or similar processing device) to further accelerate the required computation, and the controller 140 in the food packaging machine may be a simpler controller that only performs the functions as determined by the external hardware / software.

[0048] The systems and methods disclosed herein can be implemented as software, firmware, hardware, or a combination thereof. In a hardware implementation, the task division among functional units or components described above does not necessarily correspond to the division of physical units; rather, a physical component can perform multiple functions, and a task can be completed collaboratively by multiple physical components.

[0049] Some or all of the components may be implemented as software executed by a digital signal processor or microprocessor, or as hardware or application-specific integrated circuits. Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is well known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other storage technologies, optical or magnetic storage devices, or any other medium that can be used to store desired information and is accessible by a computer.

[0050] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a portion of a module, segment, or instruction, which includes one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may not appear in the order indicated in the figures. For example, two blocks shown consecutively may actually be executed substantially simultaneously, or these blocks may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs a specific function or action or executes a combination of dedicated hardware and computer instructions.

[0051] As can be seen from the above description, although various embodiments of the present invention have been described and illustrated, the present invention is not limited thereto, but may be embodied in other ways within the scope of the subject matter defined in the following claims.

Claims

1. A method for controlling roll tension in a food packaging machine (100), wherein the food packaging machine (100) includes a plurality of subsystems, the food packaging machine (100) including a gripper system responsible for package formation, lateral sealing and cutting, the method comprising: Receive one or more local variable values ​​(116), which indicate the food packaging machine (100) measurements of one or more physical parameters of the roll tension subsystem, including measurements related to one or more of the following: roll tension setpoint and current roll tension system position; Receive one or more remote variable values ​​(204), the one or more remote variable values ​​(204) indicating the measurement values ​​of one or more physical parameters of one or more remote subsystems of the food packaging machine (100), the one or more remote variable values ​​(204) including measurements related to one or more of the following: roll material movement and control variables, gripper system motion trajectory, packaging material characteristic variables, and filling state, the gripper system motion trajectory including the frequency and force of the grippers pulling the roll material; One or more control parameter values ​​are determined for the roll tension subsystem (200) by processing the remote variable values ​​(204) and the local variable values ​​(116) using a reinforcement learning model (206) and a local control model (210); as well as Based on the determined control parameter values, one or more control parameters of the roll tensioning subsystem (200) are adjusted.

2. The method according to claim 1, wherein the reinforcement learning model (206) is a deep reinforcement learning model including a neural network.

3. The method according to claim 1 or 2, wherein the roll tensioning subsystem (200) comprises two fixed guide rollers (106, 110) and a movable guide roller (108).

4. The method according to claim 3, wherein the movable guide roller (108) is located between the two fixed guide rollers (106, 110) along the path of the roll (102) through the packaging machine (100) and is movable to increase or decrease the tension of the roll (102) in response to receiving an instruction of a control parameter value.

5. The method according to claim 2, wherein the neural network is one of the following: a convolutional neural network, a recurrent neural network, a long short-term memory neural network, and a fully connected neural network.

6. A system for controlling roll tension in a food packaging machine (100), the food packaging machine having multiple subsystems, the food packaging machine (100) including a gripper system responsible for package formation, lateral sealing and cutting, the system comprising: Memory; as well as processor, The memory contains instructions that, when executed by the processor, cause the processor to perform a method comprising: Receive one or more local variable values ​​(116), which indicate the food packaging machine (100) measurements of one or more physical parameters of the roll tension subsystem, including measurements related to one or more of the following: roll tension setpoint and current roll tension system position; Receive one or more remote variable values ​​(204), the one or more remote variable values ​​(204) indicating the measurement values ​​of one or more physical parameters of one or more remote subsystems of the food packaging machine (100), the one or more remote variable values ​​(204) including measurements related to one or more of the following: roll material movement and control variables, gripper system motion trajectory, packaging material characteristic variables, and filling state, the gripper system motion trajectory including the frequency and force of the grippers pulling the roll material; One or more control parameter values ​​are determined for the roll tension subsystem (200) by processing the remote variable values ​​(204) and the local variable values ​​(116) using a reinforcement learning model (206) and a local control model (210); and Based on the determined control parameter values, one or more control parameters of the roll tensioning subsystem (200) are adjusted.

7. The system according to claim 6, wherein the reinforcement learning model (206) is a deep reinforcement learning model including a neural network.

8. The system according to any one of claims 6 or 7, wherein the roll tensioning subsystem (200) comprises two fixed guide rollers (106, 110) and a movable guide roller (108).

9. The system of claim 8, wherein the movable guide roller (108) is located between the two fixed guide rollers (106, 110) along the path of the roll (102) through the packaging machine (100) and is movable to increase or decrease the tension of the roll (102) in response to receiving a command of a control parameter value.

10. The system of claim 7, wherein the neural network is one of the following: a convolutional neural network, a recurrent neural network, a long short-term memory neural network, and a fully connected neural network.

11. A computer program product comprising a computer-readable storage medium having instructions adapted, when executed by a processor, to perform the method according to any one of claims 1-5.

Images