Diagnosis assistance for process performance, dynamic plant condition and periodic quality defects in the cold rolling area
The aim of the project was to better meet strip quality demands with regard to periodic defects through intelligent and automated monitoring and diagnosis support. To analyse DWI periodic quality defects an FFT and inverse FFT-guided diagnosis tool was generated, to determine online whether the material rolled is good for customers or to indicate causes of bad final thickness. One quality criterion was found in quick thickness variations within small sections of the strip. The tool was extended from tandem mill for hot strip and skin pass mill purposes and implemented at corresponding lines. Furthermore algorithms and procedures were developed to automatically check periodicity and sources/causes of the excitation of vibration, affecting quality signals. A core part of these results was realised in a prototype diagnosis system KOPF at a two stand temper mill. KOPF detects critical quality oscillation and corresponding cause frequencies and thus locates the defect sources automatically. Fuzzy logic rules of diagnosis assistance were developed to be implemented in a knowledge-based system built on Internet technology. This system was tailored to a cold reversing mill for magnetic steel production and tested by real data simulation. It overcomes the boundaries of the specific plant since a preventive diagnosis of the incoming trip is carried out, thus involving upstream processing lines enabling operators to make actions before cold rolling operations. Due to similar problems, plants and processes, the project findings are directly transferable to other rolling processes. Even some correspondent applications outside the rolling sector are conceivable.
Bibliographic Reference: EUR 21413 EN (2005), 104 pp. Euro: 25
Availability: Katalogue Number: KI-NA-21413-EN-S The paper version can be ordered online and the PDF version downloaded at: http://bookshop.eu/
ISBN: ISBN: 92-894-9692-4
Record Number: 200518376 / Last updated on: 2005-10-05
Original language: en
Available languages: en