Performing web tensioning adjustments in a food packaging system based on reinforcement learning

Patent metadata

Field	Value
Publication number	US-12420968-B2
Application number	US-202118257854-A
Country	US
Kind code	B2
Filing date	Dec 17, 2021
Priority date	Dec 17, 2020
Publication date	Sep 23, 2025
Grant date	Sep 23, 2025

Abstract

Methods and apparatus, including computer program products, are described for controlling web tensioning in a food packaging machine comprising a plurality of sub-systems. One or more local variable value are received, which indicate measurements by the food packaging machine of one or more physical parameters for a web tensioning sub-system. One or more remote variable values are received, which indicate measurements by the food packaging machine of one or more physical parameters for one or remote sub-systems. One or more control parameter values are determined for the web tensioning sub-system, by processing the remote and the local variable values using a reinforcement learning model and a local control model. One or more control parameters of the web tensioning sub-system are adjusted in accordance with the determined control parameter values.

First claim

The invention claimed is:

1. A method for controlling web tensioning in a food packaging machine, the method comprising:
receiving one or more local variable values indicating measurements by the food packaging machine of one or more physical parameters for a web tensioning sub-system of a plurality of local sub-systems, wherein the one or more local variable values include measurements relating to one or more of: a web tension set point or a current web tensioning system position;
receiving one or more remote variable values indicating measurements by the food packaging machine of one or more physical parameters for one or more remote sub-systems, wherein the one or more remote variable values include measurements relating to one or more of: web movement control variables, a jaw motion profile, packaging material characteristics, or a filling status;
determining one or more control parameter values for the web tensioning sub-system, by processing the remote variable values and the local variable values using a reinforcement learning model and a local control model; and
adjusting the one or more control parameters of the web tensioning sub-system in accordance with the determined control parameter values.

2. The method according to claim 1 , wherein the reinforcement learning model comprises a deep reinforcement learning model including a neural network.

3. The method according to claim 2 , wherein the neural network comprises one of: a convolution neural network, a recurrent neural network, a Long Short-Term Memory neural network, or a fully connected neural network.

4. The method according to claim 1 , wherein the web tensioning sub-system includes two stationary guide rolls and a movable guide roll.

5. The method according to claim 4 , wherein the movable guide roll is located between the two stationary guide rolls along a path traversed by the web through the packaging machine, and is movable so as to increase or decrease the tension of the web in response to instructions received from the control parameter values.

6. A non-transitory computer readable storage medium storing instructions that, when executed by a processor, cause the processor to carry out the method according to claim 1 .

7. A food packaging machine comprising:
a plurality of local sub-systems configured to control web tensioning;
a memory; and
a processor,
wherein the memory stores instructions that, when executed by the processor, cause the processor to perform a method comprising:
receiving one or more local variable values indicating measurements by the food packaging machine of one or more physical parameters for a web tensioning sub-system of the plurality of local sub-systems, wherein the one or more local variable values include measurements relating to one or more of: a web tension set point or a current web tensioning system position;
receiving one or more remote variable values indicating measurements by the food packaging machine of one or more physical parameters for one or remote sub-systems, wherein the one or more remote variable values include measurements relating to one or more of: web movement control variables, a jaw motion profile, packaging material characteristics, or a filling status;
determining one or more control parameter values for the web tensioning sub-system, by processing the remote variable values and the local variable values using a reinforcement learning model and a local control model; and
adjusting one or more control parameters of the web tensioning sub-system in accordance with the determined control parameter values.

8. The food packaging machine according to claim 7 , wherein the reinforcement learning model comprises a deep reinforcement learning model including a neural network.

9. The food packaging machine according to claim 8 , wherein the neural network comprises one of: a convolution neural network, a recurrent neural network, a Long Short-Term Memory neural network, or a fully connected neural network.

10. The food packaging machine according to claim 7 , wherein the web tensioning sub-system includes two stationary guide rolls and a movable guide roll.

11. The food packaging machine according to claim 10 , wherein the movable guide roll is located between the two stationary guide rolls along a path traversed by the web through the packaging machine, and is movable so as to increase or decrease the tension of the web in response to instructions from the received control parameter values.

Assignees

• Tetra Laval Holdings & Finance

Inventors

• Borelli Gabriele
• Fornasari Stefano

Classifications

• B65B57/04

  and operating to control, or to stop, the feed of such material, containers, or packages · CPC title

• B65B43/42

  Feeding or positioning bags, boxes, or cartons in the distended, opened, or set-up state; Feeding preformed rigid containers, e.g. tins, capsules, glass tubes, glasses, to the packaging position; Locating containers or receptacles at the filling position (by means of filling-nozzles B65B39/00); Supporting containers or receptacles during the filling operation (by filling-nozzles B65B39/00) · CPC title

• G05B13/024

  in which a parameter or coefficient is automatically adjusted to optimise the performance · CPC title

• B65B57/145

  for fluent material · CPC title

• B65B43/08

  Forming three-dimensional [3D] containers from sheet material · CPC title