JKAU: Eng. Sci., Vol. 18 No. 2, pp: 73-97 (2007 A.D. /1428 A.H.) Fault Diagnosis System for Power Transformers E. A. Mohamed*, A.Y. Abdelaziz and A. S. Mostafa ** Dept. of Elect. Power & Machines, Faculty of Eng., Ain Shams University, Cairo, **Ministry of Irrigation, Egypt, and * Qassim Univ., Saudi Arabia Abstract. This paper introduces an artificial neural network (ANN) based fault diagnosis system (FDS) for power transformers. The system is designed to detect, localize and finally classify faults. The proposed FDS consists of three hierarchical levels. In the first level, a preprocessing procedure for input data is performed. In the second level, an ANN is designed to detect the fault and localize its side. In the third level, there are two sub-diagnosis systems. Each system is dedicated to one side and consists of one ANN designed to classify the fault. This ANN is also cascaded with four parallel ANN's utilized to identify the faulted phase. The performance of FDS is evaluated using samples from local measurements (three-phase primary voltage and primary & secondary currents). These samples were generated using the EMTP simulation of the High-Dam 15.75/500 kV transformer substation in the 500 kV Upper Egypt network. Different fault types were simulated. Fault location and incipience time were also considered. Evaluation results proved that the performance of the proposed FDS is promising. 1. Introduction Fault diagnosis (FD) plays a crucial role in power system monitoring and control that ensures a stable electrical power supply to consumers. FD involves identifying the location and nature of faults occurring on power system due to different disturbances [1,2] . FD function is the most basic fault handling function of power system supervisory control and data acquisition (SCADA) systems. Figure 1 shows the different blocks that configure a protective relay, based on which the FDS module is the most important one. Fault type classification is an essential protective relaying feature due to its significant effect on the enhancement of relaying operation. Correct operation of major protective relays may be depending on fault 73