Estimação robusta de modelo funcional tridimensional (3D), para identificação de faltas em linhas de transmissão

Abstract

This thesis presents a novel method for detection and fault classification 2D and its extension to 3D, that is, taking into account the three phases of the transmission lines (LT) based on the functional form of the ellipse. the methodology implementation of Algorithms are presented, as well as tests in real and simulated data. The results were satisfactory indices with similar accuracy to those found in the literature which uses the wavelet transform pattern. It is proposed construction elliptical confidence limits through quantile regression, which are used for monitoring the transmission line to the fault detection purpose. Thus upon detection of the fault characteristics of the signals are extracted faulted via distance metric as input and used in binders in order to determine the type of fault that has occurred. Thus, the simplicity of the methodology and implementation are intelligible inherent characteristics, which can make your most extensive use.