This research explores the application of deep learning, specifically artificial neural networks, to predict and determine faults on the 161kV transmission line from Aboadze Thermal Power Station through Takoradi, Tarkwa, and Prestea to New Obuasi in Ghana. This transmission line is critical due to its role in supplying power to New Obuasi, one of the richest gold mining communities in the country. The study aims to use a deep learning approach to detect, classify, and determine the accuracy of fault predictions to ensure uninterrupted power supply. The methodology includes data collection from the transmission line, training an artificial neural network model, and evaluating its performance. Results indicate high accuracy in fault prediction, enabling timely and effective maintenance. The study concludes with recommendations for integrating deep learning models into the power grids fault detection systems to enhance reliability and efficiency