Title :
Separation and recognition of multiple PQ disturbances using independent component analysis and neural networks
Author :
Lima, Marcelo A A ; Ferreira, Danton D. ; Cerqueira, Augusto S. ; Duque, Carlos A. ; Ribeiro, Moises V. ; De Seixas, José M.
Author_Institution :
Dept. of Electr. Eng., Fed. Univ. of Juiz de Fora, Juiz de Fora
fDate :
Sept. 28 2008-Oct. 1 2008
Abstract :
This work presents the development of a system for analysis of power quality multiple disturbances using a recent non-supervised technique called independent component analysis. An application of the technique on an automatic power quality disturbance classification system, based on higher-order cumulants and artificial neural networks, is also presented. Promising results are achieved for simulated data.
Keywords :
neural nets; power engineering computing; power supply quality; artificial neural networks; automatic power quality disturbance classification system; higher-order cumulants; independent component analysis; multiple power quality disturbances; Analytical models; Artificial neural networks; Discrete Fourier transforms; Discrete wavelet transforms; Filtering; Independent component analysis; Neural networks; Power quality; Random variables; Signal analysis; Digital Filtering; Higher-order Cumulants; Independent Component Analysis; Multiple Disturbances; Neural Networks; Power Quality;
Conference_Titel :
Harmonics and Quality of Power, 2008. ICHQP 2008. 13th International Conference on
Conference_Location :
Wollongong, NSW
Print_ISBN :
978-1-4244-1771-1
Electronic_ISBN :
978-1-4244-1770-4
DOI :
10.1109/ICHQP.2008.4668839
