Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Early detection and identification of dangerous states in chemical plants using neural networks

: Neumann, J.; Deerberg, G.; Schlüter, S.


Journal of loss prevention in the process industries 12 (1999), Nr.6, S.451-453
ISSN: 0950-4230
Fraunhofer UMSICHT Oberhausen ()
early detection; exotherme Reaktion; exothermic reaction; fault diagnosis; Fehlerdiagnose; Früherkennung; safety technology; Sicherheitstechnik

The suitability of pattern recognition for safety diagnosis of chemical plants will be discussed. Therefore, experiments in a miniplant and with a process simulator are carried out. The process characteristics are treated with different recognition methods and classified with the aid of expert know how. Afterwards, the trained system can be used for process diagnosis. The capability of neural networks for this problem could be shown.