Diagnóstico automático de faltas em subestações de transmissão de energia elétrica utilizando sequência de eventos

Abstract

This work presents a general view of power systems operation considering technological evolution of information systems occurred during the last few years. It also show a conceptual review of sequence of events recorded during power system faults and its use by the real time operation crew. The consequences of the large amount of data generated due to fault and the difficulties in analyzing it are also discussed. Some recent studies related with automated analysis of faults in power systems are reviewed, leading to a verification of the computational systems capable of performing an automated analysis of sequence of events recorded during faults. These faults may result in power outages and have to be analyzed as fast as possible in order to provide the operators with enough information to take corrective actions. In order to help the operators with their task this work propose the development of an expert system, which is based on artificial intelligence techniques, to be employed as an auxiliary tool. The proposed expert system is tested with sequence of events recorded during real fault cases provided by a power system utility. The comparison of the results obtained by the proposed scheme with the manual analyses from the operators lead to the verification of the viability of implementing such system in automated fault analysis. This work also shows the limitations and benefits of using automated fault analysis schemes.