Publication detail

Neural networks for the life prognosis of insulating materials for electrical machines

HAMMER, M.

Czech title

Neuronová sít pro předpověď životnosti izolačního materiálu elektrických strojů

English title

Neural networks for the life prognosis of insulating materials for electrical machines

Type

conference paper

Language

en

Original abstract

The life of the insulating systems of electric rotary machines is strongly dependent upon electrical and thermal features of the insulating material used. The subject of the diagnostic prediction is to specify the condition of insulation used. At present days, the most popular prediction tools are the methods of artificial intelligence, and one method is the neural networks. This paper is concentrated on the use of neural networks in the life prediction of Relanex insulating material that is applied as insulation of electrical machine windings. In this case the condition of insulating in a time step k+1 is predicated from input quantity in time steps 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 neural networks which forecast the life of insulating material for windings in electric rotary, the description of train and test data and setting of neural network for prediction. The second part shows (in the figures) the curves of prediction and the simulation of insulating material behavior with neural networks. Tables evaluate the application of this tool for the solution of the problems investigated. We have used the above-mentioned neural networks for the prediction of insulating materials that were programmed in Matlab 6 environment. All simulations and the values calculated were also obtained by means of this product.

Czech abstract

Článek se zabývá diagnostikou izolačního materiálu elektrických strojů.

English abstract

The life of the insulating systems of electric rotary machines is strongly dependent upon electrical and thermal features of the insulating material used. The subject of the diagnostic prediction is to specify the condition of insulation used. At present days, the most popular prediction tools are the methods of artificial intelligence, and one method is the neural networks. This paper is concentrated on the use of neural networks in the life prediction of Relanex insulating material that is applied as insulation of electrical machine windings. In this case the condition of insulating in a time step k+1 is predicated from input quantity in time steps 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 neural networks which forecast the life of insulating material for windings in electric rotary, the description of train and test data and setting of neural network for prediction. The second part shows (in the figures) the curves of prediction and the simulation of insulating material behavior with neural networks. Tables evaluate the application of this tool for the solution of the problems investigated. We have used the above-mentioned neural networks for the prediction of insulating materials that were programmed in Matlab 6 environment. All simulations and the values calculated were also obtained by means of this product.

Keywords in Czech

neuronové sítě

Keywords in English

Neural networks

Released

01.01.2003

Publisher

Publishing House PRINTECH

Location

Miercurea Ciuc - Romania

Book

Study and Control of Corrosion in the Perspective Sustainable Development of Urban Distribution Grids

Pages count

10

BIBTEX


@inproceedings{BUT13524,
  author="Miloš {Hammer},
  title="Neural networks for the life prognosis of insulating materials for electrical machines",
  booktitle="Study and Control of Corrosion in the Perspective Sustainable Development of Urban Distribution Grids",
  year="2003",
  month="January",
  publisher="Publishing House PRINTECH",
  address="Miercurea Ciuc - Romania"
}