In modern era, ensuring a stable and continuous power supply is essential. Fault detection in power system are critical in isolating faulty sections to minimize power loss and ensure safety. This study proposes an advanced Fault Detection system based on machine learning (ML). MATLAB Simulink was used to simulate two Transmission Lines TL systems—one with a single generator and load (TL-1) and another with two generators and three loads (TL-2). The simulations generated both normal and fault data, representing different fault types. The data were normalized and analysis using a Regression Machine Learning tool, leads to the development of distinct Regression Machine Learning models for Fault Detection. Compared to traditional neural networks (NN), the Regression Machine Learning model demonstrated shorter processing times and reduced computational complexity, outperforming state-of-the-art methods.