Method and device for monitoring a yarn tension of a running yarn

Patent metadata

Field	Value
Publication number	US-11305960-B2
Application number	US-201816617395-A
Country	US
Kind code	B2
Filing date	Jun 1, 2018
Priority date	Jun 7, 2017
Publication date	Apr 19, 2022
Grant date	Apr 19, 2022

Abstract

Techniques are directed to a method and a device for monitoring a yarn tension of a running yarn in a yarn treatment process. To this end, the yarn tension of the yarn is continuously measured and the measurement signals for the yarn tension are compared with a threshold value of an admissible yarn tension. In the event of an inadmissible tolerance deviation of the measurement signals, a short-term signal path of the yarn tension is detected as a fault graph. In order to enable a fault diagnosis, the fault graph of the yarn tension is analyzed using a machine learning program. The fault graph is then allocated to one of the existing fault categories or to a new fault category. A device for this purpose may include a diagnosis unit, which cooperates accordingly with the yarn tension evaluation unit.

First claim

The invention claimed is:

1. A method for monitoring a yarn tension of a running yarn in a yarn treatment process, in which the yarn tension of the yarn is progressively measured, in which measurement signals of the yarn tension are compared to at least one limiting value of a permissible yarn tension and in which in the event of an impermissible tolerance deviation of the measurement signals, a short-term signal profile of the yarn tension is acquired as a fault graph,
wherein the fault graph of the yarn tension is analyzed using a machine learning program,
wherein the fault graph is assigned to a known fault graph category or a new fault graph category, and
wherein after assignment of one of the fault graphs to one of the fault graph categories, a control command relating to the fault graph category effectuates a direct intervention into the yarn treatment process.

2. The method as claimed in claim 1 , wherein the fault graph categories are each specified by a fault pattern of one of the fault graphs and/or a group of fault graphs.

3. The method as claimed in claim 1 , wherein a specific process disturbance and/or a specific operating fault and/or a specific disturbance parameter and/or a specific product fault is/are assigned to each of the fault graph categories.

4. The method as claimed in claim 1 , wherein the analysis of the fault graphs is executed by at least one machine learning algorithm of the machine learning program.

5. The method as claimed in claim 4 , wherein at least one of the fault graph categories is defined solely by the machine learning algorithm from analyzed fault graphs.

6. A device for monitoring a yarn tension of a running yarn in a yarn treatment process, comprising:
a yarn tension measuring unit having a yarn tension sensor and having a measurement signal pickup, and
a yarn tension analysis unit having a fault graph generator,
wherein the yarn tension analysis unit interacts with a diagnostic unit in such a way that a fault graph is analyzable using a machine learning program,
wherein a known fault graph category or a new fault graph category is assigned to the fault graph, and
wherein the diagnostic unit is connected to a machine control unit, by which a control command effectuates a direct intervention into the yarn treatment process.

7. The device as claimed in claim 6 , wherein the diagnostic unit comprises a storage unit and a programmable learning processor for executing the machine learning program.

8. The device as claimed in claim 7 , wherein the learning processor is coupled to an input unit, by which one or more ascertained fault graphs can be input.

9. The device as claimed in claim 7 , wherein the learning processor is coupled to an output unit, by which an assignment of the analyzed fault graphs to one of the fault graph categories can be visualized.

10. The device as claimed in claim 7 , wherein the learning processor comprises a neural network for executing the machine learning program.

11. A method for monitoring a yarn tension of a running yarn in a yarn treatment process, the method comprising:
progressively measuring the yarn tension of the running yarn in the yarn treatment process to provide measurement signals identifying the yarn tension,
comparing the measurement signals to at least one limiting value of a permissible yarn tension to detect an event of an impermissible tolerance deviation of the measurement signals,
in response to the event of the impermissible tolerance deviation of the measurement signals, acquiring a short-term signal profile of the yarn tension as a fault graph of the yarn tension,
analyzing the fault graph of the yarn tension using a machine learning program,
based on analyzing the fault graph of the yarn tension, assigning the fault graph to one of a known fault graph category and a new fault graph category, and
wherein after assignment of one of the fault graphs to one of the fault graph categories, a control command relating to the fault graph category effectuates a direct intervention into the yarn treatment process.

Assignees

• Oerlikon Textile Gmbh & Co Kg

Inventors

• Huthmacher Jörg

Classifications

• B65H59/40 Primary

  Applications of tension indicators · CPC title

• G05B23/0229

  knowledge based, e.g. expert systems; genetic algorithms · CPC title

• G01L5/102 Primary

  using sensors located at a non-interrupted part of the flexible member · CPC title

• G01N33/365

  Filiform textiles, e.g. yarns · CPC title

• B65H63/02

  responsive to reduction in material tension, failure of supply, or breakage, of material · CPC title