Verbesserte Webmaschineneinstellungen mittels Simulationsrechnungen

The weaving process is substantially influenced by the warp thread tension. In order to obtain good fabric quality at a smooth run of the loom, the optimisation of the warp thread tension is necessary. The tension is influenced by a multitude of parameters, especially the back shedding geometry. The pure analytical determination of the best back shedding geometry is not possible. The experimental determination of the best loom setting takes up a great deal of time and is only an approximation. This complex task occurs particularly with new products, for which the user has not established machine settings yet. Within the framework of the presented thesis the development of a technical system is presented that provides optimised machine settings based on machine and fabric specific data and with help of mathematic methods and simulations. Neural networks have been trained and tested for the prediction of ideal article and machine specific warp thread tension sequences (comparison tension sequences) based on empiric data of weaving mills. For that the necessary product, machine and operational data was collected on very well adjusted weaving machines with satisfactory processing performance and fabric quality over a long time period. This data has been analysed by special mathematic methods. The determination of the machine setting from the tension sequence is not possible directly. Because of that the optimisation algorithm of the AUTO-WARP-concept, that has been developed at the Institut für Textiltechnik der RWTH Aachen (ITA), is used. According to the principle of evolution strategy and based on fabric and machine data, the comparison tension sequence and by the help of a special simulation program of the ITA for warp thread tension sequences the fabric and machine specific optimised machine setting is calculated. For an easy operation the various described components have been put together to one program and a common user interface was created to achieve a simple data input according to the demand of a weaving mill. The developed system was tested in practise for the product family of aramid yarns. In the practise test the developed adjustment device was able to determine good and very good machine settings for new products and machines, although the test fabric was far out of the training product spectrum. With the recommended loom setting the warp thread tension sequence, the run of the loom and the fabric quality were improved clearly. The adjustment device is a tool for the weaver that makes a good and quick fabric specific adjustment of the back shedding possible even for new fabrics. By this it enable the weaver to reduce material, machine and personal costs and to improve the fabric quality. The program is simple to operate and runs on usual personal computers. In order to use the system for new product families the neural networks have to be trained with corresponding data. This further conversion and marketing strategies are viewed finally.

Grantor dc:publisher

Publikationsserver der RWTH Aachen University

Year dc:date

2003

Author dc:creator

• Wolters, Thomas

Contributors dc:contributor

• Wulfhorst, Burkhard

• info:eu-repo/classification/ddc/660
• Webmaschine
• Kettfaden
• Zugkraft
• Prozesssimulation
• Neuronales Netz
• Einrichten
• Weben
• Prozessoptimierung
• Computerunterstütztes Verfahren
• Technische Chemie
• Weberei
• Webmaschineneinstellung
• Simulationsrechnungen
• Evolutionsprozess
• neuronale Netze
• artikelspezifisch
• maschinenspezifisch
• Kettfadenzugkraft

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

ger

Identifier

https://publications.rwth-aachen.de/search?p=id:%22RWTH-CONV-120933%22

OAI identifier oai:identifier

oai:publications.rwth-aachen.de:59121