Fault Location in Electrical Distribution Systems with Distributed Generation Presence Based on Algorithmic Methods

This project focuses on addressing the challenge of fault location in electric distribution systems, a problem exacerbated by the integration of distributed generation (DG). Conventional methods, often relying on visual inspection or simple algorithms dependent on single substation measurements, prove slow and inaccurate due to network topology complexity and inhomogeneity. The proposed solution centers on applying and evaluating algorithmic methods, including Artificial Intelligence techniques like Support Vector Machines (SVM), to enhance the speed and accuracy of fault point identification. The methodology involves a state-of-the-art review, detailed modeling, and simulation of distribution circuits (using specialized software and MATLAB®) to generate robust databases. The accuracy of the methods is assessed under various conditions, such as different fault types, variations in fault resistance, and sensitivity to the amount and location of DG. The expected impact is a significant reduction in service restoration times and improved operational continuity by overcoming the limitations of traditional algorithms when handling the complexity introduced by DG.<br/><br/><b>Goal</b>: <br/>The main objective is to locate faults in electric distribution systems that feature the presence of distributed generation, utilizing advanced algorithmic methods.<br/><br/><b>Research lines</b>: <br/>Transmission and distribution of electrical energy