Reduced kernel recursive least squares algorithm for aero-engine degradation prediction

•The novel method considers all the constraints generated by the whole training set.•The redundant data is used to modify the coefficients related to each kernel unit.•The structure update and coefficient adjustment are carried out independently.•A more compact network without accuracy deterioration...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Mechanical systems and signal processing Ročník 95; s. 446 - 467
Hlavní autoři:	Zhou, Haowen, Huang, Jinquan, Lu, Feng
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Berlin Elsevier Ltd 01.10.2017 Elsevier BV
Témata:	Accuracy Adaptive algorithms Adaptive filters Aero-engine Basis functions Degradation Kernel recursive least squares Least squares method Markov chains Mathematical models Prognostics Radial basis function Recursive algorithms Reduced technique Redundancy Remaining useful life Signal processing Sparse kernel method Time series Training Reduced technique Aero-engine Remaining useful life Kernel recursive least squares Prognostics Sparse kernel method
ISSN:	0888-3270, 1096-1216
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Buďte první, kdo okomentuje tento záznam!