Successful applications of neural networks in electrical power systems have demonstrated that this powerful tool can be employed as an alternative method for solving problems accurately and efficiently. The features of neural networks, such as their ability to learn, generalize and parallel processing, among others, have made their applications to many systems ideal. The use of neural networks as pattern classifiers is among their most common and powerful applications.  This thesis presents an artificial neural network approach to detection, classification and isolation (location) of faults in transmission line systems. The objective is to implement a complete scheme for distance protection of a transmission line system. In order to perform this goal, the distance protection task is subdivided into different neural networks for fault detection, fault identification (classification) as well as fault location in different zones. The other purpose of this work is to study and compare the application of three different neural network architectures for the protection of the transmission lines. The considered three approaches are back-propagation, radial basis functions and support vector machines. Simulation results are provided to demonstrate the advantages and disadvantages of these structures when applied to the problem of transmission line protection. (Abstract shortened by UMI.)