| Texto completo | |
| Autor(es): |
Palate Moyolema, Byron Orlando
;
Tchipongue Kalueyo, Celestino Paulo
;
De Aragao, Pedro Sato
;
Di Santo, Silvio Giuseppe
;
Kagan, Nelson
;
Almeida, Carlos Frederico
Número total de Autores: 6
|
| Tipo de documento: | Artigo Científico |
| Fonte: | 2025 15TH INTERNATIONAL CONFERENCE ON ELECTRICAL ENGINEERING, ICEENG; v. N/A, p. 5-pg., 2025-01-01. |
| Resumo | |
Analyzing faults in power systems is challenging once the current and voltage signals measured may contain several components added to the fundamental one. In addition to the fundamental and the harmonics, the signal may contain non-periodical components that lead the Fourier analysis to fail. In the literature, several methods are proposed to remove non-periodic components; one promising is the Prony's method. This method is based on Prony's filter, which tries to describe a signal as a sum of components, similar to the Fourier series, but the components have exponentials multiplied by the sines and cosines terms. Once the signal is represented as a sum of terms, the user has to provide Prony's method with the number of terms, and the quality of the output depends upon this number. A large number of terms does not always lead to a good signal representation and causes an unnecessary computation burden. Therefore, this work proposes a method to find the optimal number of terms, known as the "p" value, of Prony's method, leading to an effective non-periodical component removal. (AU) | |
| Processo FAPESP: | 20/15230-5 - Centro de Pesquisa e Inovação de Gases de Efeito Estufa - RCG2I |
| Beneficiário: | Julio Romano Meneghini |
| Modalidade de apoio: | Auxílio à Pesquisa - Programa Centros de Pesquisa Aplicada |