Abstract

The paper deals with a segment of research examining the possibility of application of the theory of recognizing samples as the core of a diagnosing model. System diagnosing is based upon the signals of operation process. Diagnosing rules are determined based upon the sample of diagnosing signals produced by different simulated aspects of system technical condition. The model has been verified in the process of Diesel engine fuel injection system diagnosing. The fuel injection pressure signal in front of the injector has been used as a diagnosing signal.

References

E.BAZIJANAC: Istraživanje mogućnosti automatskog dijagnosticiranja sistema za ubrizgavanje goriva dizel motora. Doktorska disertacija, VVTŠ, Zagreb, 1991.

J.T.TOU, R.C.GONZALES: Pattern recognition principles. Addison- Wesley Publishing Company, London, 1984.

L.MONOSTORI: Pattern recognition based learning and decision making in complex machine tool monitoring systems. International Federation for Information Processing, Dubrovnik, 1987