Generally Extra High Voltage (EHV) transmission substations in power systems are connected with multiple transmission lines to neighboring substations. The transmission lines are exposed to faults, which are equipped with protective devices such as relays and circuit breakers. Power system disturbance are often caused by a faults on transmission lines. When a fault occurs on one of the transmission lines, the protective relays expected to detect the fault and give trip signal to circuit breaker to clear the fault. In these cases over tripping of relays can happen because of inappropriate coordination of relay settings. Due to this condition the power system margins for contingencies are decreasing. With this situation, power system protective relaying reliability be-comes increasingly important. In this paper an approach is presented using Support Vector Machine (SVM) as an artificial intelligent tool for identifying the faulted feeder that is emanating from a substation and finding the distance from the substation. Support vector Machines have well-established advantages over other methods. This approach is particularly important to avoid mal- operation of relays following a disturbance in the neighboring feeder connected to the same substation and assuring secure operation of the power systems. Copyright © IICAI 2005.