Uma nova proposta para detecção de faltas de alta impedância envolvendo a ruptura do condutor

Abstract

Overhead lines of the power distribution networks are subject to climate, human and animal actions. These actions may cause the power supply to be interrupted by the protection system acting. In this context, conventional protections act whenever they detect an overcurrent or a voltage surge in the system, which may be caused by a short circuit, leakage current in isolators or overcharge, among other factors. However, there are situations where these conventional devices do not work. One of these occurs, for example, when a conductor of the primary network breaks and comes into contact with a surface that has a high electrical resistance. This event is called High Impedance Fault (HIF). In this type of fault, the resulting electric current, depending on the terrain, can reach values close to the nominal, so that the protection devices are not sensitized. This problem is very important as the fallen driver remains energized, endangering the lives of people and animals passing by. Given this, many studies are being conducted to develop methodologies capable of detecting the HIFs. However, there is no robust method yet to detect high impedance faults under all possible operating conditions. In addition, most existing methodologies are economically unviable for large scale deployment. Thus, this work proposes a new technique for detecting FAIs that uses only the current values sampled at the power substation, which has a transient detector that, in turn, serves as a trigger for the phase identification process. of buildup, a unique feature of current waveforms in this fault type. The proposed algorithm was implemented in the MATLAB R environment and tested based on simulations performed in the EMTP-ATP software to obtain current signals using computer simulations of FAIs found in the literature. For purposes of comparison and proving the effectiveness of the proposed method, its response was compared to that obtained through a methodology based on the analysis of the negative sequence current of the network for different fault situations. The results show that the proposed technique is robust and effective for detecting FAIs, since it correctly classified the occurrence of high impedance faults in the vast majority of cases analyzed and did not act falsely in the event of transient events common to the system, such as transformer energization and capacitor bank switching, input and output of loads.