Power Transformer Fault Diagnosis using Som-Based RBF Neural Networks

A radial basis function (RBF) neural network used in fault diagnosis system is developed for power transformer fault analysis. The Gas extracted from transformer oil is the input of RBF-type neural network architecture. Our proposed cell-splitting grid algorithm determines the optimal network architecture of the RBF network automatically. This facilitates the conventional laborious trail-and-error procedure in establishing an optimal architecture. In this paper, the proposed RBF machine fault diagnostic system has been intensively tested with the overheating faults and discharging faults of power transformer