Please use this identifier to cite or link to this item:
- Using artificial neural networks for identification of electrical losses in transformers during the manufacturing phase
- Universidade Estadual Paulista (UNESP)
- The paper describes a novel neural model to estimate electrical losses in transformer during the manufacturing phase. The network acts as an identifier of structural features on electrical loss process, so that output parameters can be estimated and generalized from an input parameter set. The model was trained and assessed through experimental data taking into account core losses, copper losses, resistance, current and temperature. The results obtained in the simulations have shown that the developed technique can be used as an alternative tool to make the analysis of electrical losses on distribution transformer more appropriate regarding to manufacturing process. Thus, this research has led to an improvement on the rational use of energy.
- Proceeding of the 2002 International Joint Conference on Neural Networks, Vols 1-3. New York: IEEE, p. 1346-1350, 2002.
- Institute of Electrical and Electronics Engineers (IEEE)
- Acesso restrito
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.