Skip to Main Content
In this paper, a regression technique as the support vector machines (SVM) configured using an optimization technique as the Chu Beasley Genetic Algorithm (CBGA) is proposed to develop a fault location method. As result, a strategy is proposed to relate a set of descriptors obtained from single end measurements of voltage and current (input), to the fault location (output), in a classical regression task. The developed strategy is tested in the selection of the best calibration parameters of a single phase SVM based fault locator where an average error of 5.278% is then obtained. According to the results, the proposed methodology could be applied successfully in power distribution systems.