This paper presents the development of an algorithm for identifying the saturation and locating the fault in relation to the protected object (internal or external fault). The methodology applied was based on the ability of detecting sudden changes of the signal through the Discrete Wavelet Transform and on the learning capacity of the Artificial Neural Network. In the localization stage, a decision criterion was used in the time for the corresponding current shifting calculation within a frame of a cycle. The results show that the protection device may take this methodology as an alternative for faster decisions even under eventual CT saturation conditions.