Detection of System Changes for a Pneumatic Cylinder Using Self-Organizing Maps
2006 (English)In: IEEE International Symposium on Computer-Aided Control Systems Design, CACSD’06, 2006, 2647-2652 p.Conference paper (Refereed)
Automated monitoring of system is growing in importance as systems become increasingly autonomous and intelligent control is being used. At the same time, component manufacturers' desire to offer components with embedded condition monitoring systems is also increasing. The problem with classical model based monitoring for the component manufacturer is the lack of information about the actual application in which the component is to be used. A general, adaptive method is therefore needed. One such algorithm is the self-organizing (feature) map, which has the desired property of reducing the dimensions of the information space. In this paper, two different measures of divergence from the normal state of operation are discussed: the quantization error and a measure of the neurons' individual training level. The combination of these measures is also briefly discussed.
Place, publisher, year, edition, pages
2006. 2647-2652 p.
control engineering computing, intelligent control, monitoring, pneumatic control equipment, self-organising feature maps, automated system monitoring, autonomous control, condition monitoring systems, intelligent control, pneumatic cylinder, system change detection
Engineering and Technology
IdentifiersURN: urn:nbn:se:liu:diva-12976DOI: 10.1109/CACSD.2006.285524ISBN: 0-7803-9798-3OAI: oai:DiVA.org:liu-12976DiVA: diva2:17572