ZÁVYSLOST PŘEDPOVĚĎI ŽIVOTNOSTI IZOLAČNÍHO MATERIÁLU NA TYPU FUZZY NEUROVÉ SÍTĚ

DEPENDING OF LIFETIME PREDICTION OF INSULATIVE MATERIAL ON THE TYPE OF FUZZY NEURAL NETWORK

conference paper

en

At present time, due the fast development of computer technology, more and more application of modern control methods in the practice appear. As an example the fuzzy logic, neural systems and hybrid systems like fuzzy neural networks can be mentioned. Those methods can be used for diagnostics. Usage of such instruments can not only evaluate the current insulation condition, but also forecast the certain prognoses of the life of insulating material under the specified operational conditions. This contribution deals with prediction of the residual life of the Relanex insulating material for electrical rotary machine windings by use of fuzzy neural networks.In case of prediction, the insulation condition in the future time k+1 is predicted based on the input quantities in the times k, k-1, k-2, etc. Anyway the prediction means forecasting of quantity in future from N previous measurement this or other quantities in the past. The first part of the paper describes the use of artificial fuzzy neural networks which forecast the life of insulating material for windings in electric rotary machines and setting of fuzzy neural network for prediction. The second part shows the prediction of insulating material behavior with neural networks

Článek pojednává o využití metod umělé inteligence pro diagnostiku životnosti izolačních materiálů vinutí elektrických točivých strojů. Jako vhodný nástroj zde byly použity fuzzy-neuronové sítě jenž spojují výhody fuzzy systémů a neuronových sítí. Predikovaný výstup byl sledován při použití různých učících algoritmů fuzzy-neuronových sítí u nichž byly vhodně nastaveny jejich parametry jako je počet skrytých vrstev, počet neuronů v jednotlivých vrstvách a vhodná metoda defuzzifikace. Tyto sítě byly naprogramovány v programu Matlab verze 6.0, stejně tak výsledky simulací byly získány tímto produktem.

At present time, due the fast development of computer technology, more and more application of modern control methods in the practice appear. As an example the fuzzy logic, neural systems and hybrid systems like fuzzy neural networks can be mentioned. Those methods can be used for diagnostics. Usage of such instruments can not only evaluate the current insulation condition, but also forecast the certain prognoses of the life of insulating material under the specified operational conditions. This contribution deals with prediction of the residual life of the Relanex insulating material for electrical rotary machine windings by use of fuzzy neural networks.In case of prediction, the insulation condition in the future time k+1 is predicted based on the input quantities in the times k, k-1, k-2, etc. Anyway the prediction means forecasting of quantity in future from N previous measurement this or other quantities in the past. The first part of the paper describes the use of artificial fuzzy neural networks which forecast the life of insulating material for windings in electric rotary machines and setting of fuzzy neural network for prediction. The second part shows the prediction of insulating material behavior with neural networks

Fuzzy; Neural sítě; Predikce; Diagnostika

Fuzzy; Neural Network; Prediction; Diagnostics

2005

09.06.2005

Sibiu, Romania

973-718-259-6

UBR-CORR Studa and Control of Corroslon in the Perspective of Sustalnable Development of Urban Distribution Grids

160–162

3