Using artificial neural network for electrical losses analysis in 3-phase transformers

This paper proposes the use of artificial neural network (ANN) for electrical losses analysis in 3-phase distribution transformers during the stages of transformer construction. An ANN is employed as a predictor in order to identify electrical loss of the distribution transformer. The data of related parameters such as core loss, copper loss and the winding resistance at different temperature levels, collected from 100 transformers 160 kVA 3 ph 50 Hz 33 kV-400/230 V Dyn11, are used to train ANN. The trained ANN is employed to identify electrical losses. Satisfactory results were obtained from the simulations which were compared to the measured values.