Euclidean Distance-Based Method for Fault Detection and Classification in Transmission Lines
Guilherme T. Alencar,Ricardo C. Santos,Aline Neves
DOI: https://doi.org/10.1007/s40313-022-00918-x
2022-04-29
Journal of Control, Automation and Electrical Systems
Abstract:Overhead transmission lines are the most vulnerable components in electrical power systems. Therefore, to improve the electrical system availability and reliability, fault location should be performed as quickly and accurately as possible. This procedure consists of several steps, including fault detection and classification, which are essential to successfully locate the fault. Due to the importance of these two steps in protection schemes, different approaches have been proposed recently, such as algorithms based on Fourier Transform, Wavelet Transform, Statistical and Artificial Intelligence Techniques. Although they are all promising, they require complex formulations or high computational efforts. This paper proposes a method based on the Euclidean distance measure to perform the fault detection and classification functions, only by using voltage signals from the transmission line ends. The proposed method is simple and robust, not requiring complex formulation or significant computational effort. A large number of tests were performed, considering the power system under different operational conditions and fault characteristics such as fault resistance and fault inception angle, among others. The observed results showed that the proposed method is capable of accurately detecting and classifying faults, regardless of the power system condition or fault characteristic. The comprehensiveness and applicability of the proposed method was confirmed by means of two different transmission lines (line parameters, voltage level, and AC equivalent systems).