LIFETIME PREDICTION OF INSULATING MATERIALS BY FUZZY-NEURAL NETWORKS AND EXPERT SYSTEMS

conference paper

English

The paper deals with searching for new methods of diagnostics and lifetime prediction of insulating materials of electric rotary machines windings. The subject of the diagnostics is to specify the condition of insulation used. In this time, the most popular diagnostics tools are the methods of artificial intelligence like expert systems and fuzzy neural networks. In our research we are using expert system for diagnostics of insulating material and fuzzy neural network for lifetime prediction of this material. Input of those systems is Bv (activation energy) which was obtained by non-destructive measurement method. The determination of this quantity is the main prerequisite for the determination of output coefficient Up (breakdown voltage) characterizing the lifetime of the insulating system and, consequently, the whole electric machine.

Neural Networks, Expert Systems, Diagnostics, Insulating Material, Artificial Intelligence

HAMMER, M.; ŘÍHA, Z.; LATINA, P.

2005

6. 9. 2005

University of Salerno

Fisciano, Italy

88-87030-96-0

ICPR-18

-

-

4