Application of Artificial Neural Networks for electrical losses estimation in three-phase transformer

This paper proposes an application of Artificial Neural Networks (ANN) for estimation of electrical losses in the three-phase distribution transformer during construction stages. The Artificial Neural Networks (ANN) is employed as an estimator in order to identify the electrical loss of the distribution transformer during design process. The related parameters such as input current, core loss, copper loss, resistance of transformer windings, and ambient temperature were collected from the measuring of 100 transformers. Some of these data are used to train ANN and test. The trained ANN is then tested by 20 data sets from the collected data. The simulations which are compared to the measured values of the test sets provide satisfactory estimation of electrical loss with an acceptable error.