Pracovní životnost izolačních materiálů vinutí elektrických strojů

A working life diagnostic of insulating materials for electric machines winding

conference paper

Czech

The contribution deals with the use of artificalintelligence methods in the life diagnostics of Relanex insulating material thet is applied as insulation of elestric machine windings. For example, neural networks are one applicable method. The method belong to the appropriate tools that provide the modeling, identification and simulation of technological systems and units. In such a case, the insulating material is used as a modelled and identified system. The first part of the paper describes the archutecture and the setting of neural networks, the description of input modelled and identified data. The second part shows (in the figures) the curves of identification, modelling and the simulation of insulating material behaviour with neural networks. Tables evaluate the application of this tool for solution of thr problems investigated. We have used the abovementioned neural networks for the diagnosticsof insulating materials that were programmed in Matlab 6 environment. All simulations and the values calculated were also obtained by means of this product.

The contribution deals with the use of artificalintelligence methods in the life diagnostics of Relanex insulating material thet is applied as insulation of elestric machine windings. For example, neural networks are one applicable method. The method belong to the appropriate tools that provide the modeling, identification and simulation of technological systems and units. In such a case, the insulating material is used as a modelled and identified system. The first part of the paper describes the archutecture and the setting of neural networks, the description of input modelled and identified data. The second part shows (in the figures) the curves of identification, modelling and the simulation of insulating material behaviour with neural networks. Tables evaluate the application of this tool for solution of thr problems investigated. We have used the abovementioned neural networks for the diagnosticsof insulating materials that were programmed in Matlab 6 environment. All simulations and the values calculated were also obtained by means of this product.

diagnostic, insulating materials, electric machines, neural networks

HAMMER, M.

1. 1. 2002

EDIS- vadavateľstvo ŽU

Žilina, Slovenská republika

80-7100-960-1

Nové smery v diagnostike a opravách elektrických strojov a zariadení

179

184

6