Redes neurais aplicadas na classificação de distúrbios da rede elétrica

Abstract

Artificial neural networks (ANNs) multilayer perceptron are widely used in complex problems due to their ability to emulate learning and present lower errors in supporting decision making. Therefore, it is feasible to use this computational technique applied to the detection and classification of disturbances in class A consumers, given its ability to model complex electrical data. By being trained with voltage and current information, they can identify oscillations, harmonics, and imbalances, classifying them into specific categories. This automation significantly reduces the time required to identify and resolve electrical problems, ensuring the quality of the energy. This article presents a study with the application of artificial neural networks in the identification and classification of voltage disturbances in the electric power distribution network. The ANN is trained with real data. Test and validation scenarios are presented and discussed.