Fault Feature Extraction Based on Improved EEMD and Hilbert Transform

Autor:	Qian Zhou, Pei Guo Hou, Zhong Dong Wang
Rok vydání:	2011
Předmět:	business.industry Feature extraction General Engineering Pattern recognition Fault (power engineering) Instantaneous phase Signal Hilbert–Huang transform Support vector machine symbols.namesake Mixing (mathematics) symbols Artificial intelligence Hilbert transform business Mathematics
Zdroj:	Advanced Materials Research. :1126-1130
ISSN:	1662-8985
DOI:	10.4028/www.scientific.net/amr.314-316.1126
Popis:	Ensemble Empirical Mode Decomposition (EEMD) can overcome the mode mixing problem in Empirical Mode Decomposition (EMD) effectively. The Hilbert-Huang transform still exists end effect in applications, in order to improve the end effect, this paper put forward a method of fault feature extraction based on improved EEMD and Hilbert transform which combines support vector regression (SVR) machine with mirror extension to continue the signal. The analysis on simulation experiments results show that the method can restrain the end effect effectively, get a more accurate instantaneous frequency and instantaneous amplitude.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::efcf83b67e26d93e8e81c4938bc9caaa https://doi.org/10.4028/www.scientific.net/amr.314-316.1126 Zobrazit plný text záznamu