Application of independent component analysis in vibration

[thumbnail of Applicationofindependentcomponentanalysisinvibration.pdf]
Preview
Available under license: Creative Commons Attribution

Parker, Anthony and Tran, Danh (2014) Application of independent component analysis in vibration. In: 23rd Australasian Conference on the Mechanics of Structures and Materials (ACMSM23), 09 December 2014-12 December 2014, Byron Bay, NSW.

Abstract

This paper investigates the application of Independent Component Analysis (ICA) in vibration of structures to infer excitation forces from vibration response signals only. First, fundamental notions of ICA, statistical independence (SI) of variables and an ICA algorithm for real number, fastICA, were reviewed. The performance of ICA was assessed using harmonic force excitation as source signals and mixture signals are displacement responses from finite element simulation of vibration of a cantilever beam. It was found that fastICA could extract source signals that share most dynamic characteristics of the excitation source signals. Further studies are proposed to tailor ICA better to vibration signals.

Item type Conference or Workshop Item (Paper)
URI https://vuir.vu.edu.au/id/eprint/30167
ISBN 9780994152008
Subjects Historical > FOR Classification > 0913 Mechanical Engineering
Current > Division/Research > College of Science and Engineering
Keywords statistical independence; fastICA; ICA
Download/View statistics View download statistics for this item

Search Google Scholar

Repository staff login