High voltage transmission lines (HVTL) are susceptible to numerous faults. In fact, the percentage of faults occurring on a transmission line (TL) is typically in the range of 80-85% when compared to total faults in the entire power system. If a prompt and efficient countermeasure is not adopted once a fault occurs, it may propagate to other equipment and, in the worst-case scenario, lead to a system-wide blackout. Therefore, fault diagnosis in HVTLs remains a challenging problem, and there is a need for a reliable and efficient transmission line distance protection scheme, as well as an exceptional fault diagnosis method to handle this challenging task. In this paper, a viable fault diagnosis approach using artificial neural networks (ANN) is proposed to achieve an efficient, reliable, and secure TL distance protection scheme. The Rogowski coil is utilized as an alternative to conventional current transformers due to its built-in linear working characteristics. Ten different fault types are created on a typical 200 km, 220 kV TL, each with different fault inception angles and resistances at five various locations, for experimentation purposes. A conventional tool is used to generate the data set containing the apparent impedance values necessary for the proposed model. The presented method establishes a good relationship between apparent impedances, fault types, and fault locations. Finally, the results show that the proposed model is more accurate than existing approaches.