Development of Power Transformer Health Index Assessment Using Feedforward Neural Network

The role of a power transformer is to convert the electrical level and send it consumer, making an essential component system. In addition, asset management for monitoring functioning transformers in system prevent failure anticipating health state transformers, using technique known as index (HI). However, calculation computation determine HI based on scoring ranking complex required expert validation. Therefore, this paper presents prediction feedforward neural network (FFNN) improve existing technique. Levenberg–Marquardt (LM), Bayesian Regularized (BR), Scaled Conjugate Gradient (SCG) are FFNN training techniques presented study forecast HI. To validate techniques, values generated by different were compared Then, performance proposed ANN was evaluated correlation coefficient mean square error (MSE). As result, successfully predicted employing three namely LM, BR, SCG which able whether transformer's condition very good, fair, or poor. conclusion, suggested has also been validated with approach, provides high similarity score comparison index.