Este trabalhoé dedicadò a minha esposa Patrícia, ao meu filho João Pedro, aos meus pais Izabel e Nelson e aos meus irmãos Lívia e Gustavo.iv AGRADECIMENTOS A minha esposa Patrícia pelo amor, companheirismo, dedicação, compreensão, paciência, garra para cuidar do nosso príncipe em minhas ausências, por entender minhas ausências diárias, pelo apoio e carinho essenciaisà minha tranquilidade para mais essa vitória.Ao meu príncipe João Pedro por receber o papai todos os dias com um sorriso lindo e reconfortante, por dar sentido a vida do papai e da mamãe, e por compreender, ainda que com poucos meses de vida, que minhas ausências diárias tinham oúnico propósito de trazer conforto e aprendizadoà nossa família. The proposed methodology was evaluated by two kinds of data. In a first kind, the signals were simulated. For this purpose, it was used the Power System Toolbox software (PST). In the second kind, it was considered field data obtained from PMUs installed at low voltage in Brazilian universities in several states. Considering the stochastic nature of the data, the proposed method identified the dominant modes present in the simulated ambient data of the test system accordingly, both in situations in which the frequency and damping of the mode were fixed, as in situations in which the mode parameters varied throughout the period of simulation. In the case of the field data, the results identified the presence of dominant oscillation mode in the frequency range between 0,2 Hz and 0,4 Hz. This is an inter-area mode, characteristic of the Brazilian National Grid. Therefore, in both kinds of data, the proposed methodology was appropriate to identify dominant modes from ambient data.vii
