Because excellent properties of the fixed-point iteration algorithm of fast independent component analysis algorithm (FastICA) and better statistical properties of negative entropy and Newton iteration, a FastICA effectively combining negative entropy and Newton iteration for OMA is put forward in this paper. Modal parameter identification results on simulation data of beam show that this method could identify the main modal vibration modes and natural frequencies correctly and efficiently only from measured signals of stationary random vibration response and more robustness than gradient descent and information maximization based ICA for OMA.
ReyndersE., System identification methods for (operational) modal analysis: Review and comparison [J], Archives of Computational Methods in Engineering19(1) (2012), 51-124.
2.
RainieriC. and FabbrocinoG., Operational modal analysis of civil engineering structures [M]. Springer, New York, 2014.
3.
WangC., GuanW., GouJ. et al., Principal component analysis based three-dimensional operational modal analysis [J], International Journal of Applied Electromagnetics and Mechanics45(1) (2014), 137-144.
4.
SwaminathanB., SharmaB. and ChauhanS., Utilization of blind source separation techniques for modal analysis[M]//Structural Dynamics, Volume 3. Springer New York, 2011, 189-206.
5.
PonceletF., KerschenG., GolinvalJ.C. et al., Output-only modal analysis using blind source separation techniques [J], Mechanical Systems and Signal Processing21(3) (2007), 2335-2358.
6.
HyvärinenA., Independent component analysis: recent advances [J], Philosophical Transactions of the Royal Society of London A: Mathematical, Physical and Engineering Sciences371(1984) (2013), 20110534.
7.
Durán-DíazI., CrucesS., Sarmiento-VegaM.A. et al., Cyclic maximization of non-Gaussianity for blind signal extraction of complex-valued sources [J], Neurocomputing74(17) (2011), 2867-2873.
8.
J.`Vía, PalomarD.P., VielvaL. et al., Maximum likelihood ICA of quaternion Gaussian vectors [C] Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on. IEEE, 2011, 4260-4263.
9.
LangloisD., ChartierS. and GosselinD., An introduction to independent component analysis: InfoMax and FastICA algorithms [J], Tutorials in Quantitative Methods for Psychology6(1) (2010), 31-38.
HyvärinenA., The fixed-point algorithm and maximum likelihood estimation for independent component analysis [J], Neural Processing Letters10(1) (1999), 1-5.
12.
HyvärinenA., Fast and robust fixed-point algorithms for independent component analysis [J], Neural Networks, IEEE Transactions on10(3) (1999), 626-634.
13.
XuB., JinH., TanX. et al., Extracting Fetal Electrocardiogram based on a modified fast independent component analysis[C]//Fuzzy Systems and Knowledge Discovery (FSKD), 2012 9th International Conference on. IEEE, 2012, 1787-1791.
14.
HyvärinenA. and OjaE., Independent component analysis: Algorithms and applications [J], Neural Networks13(1) (2000), 411-430.
15.
DeuflhardP., Newton methods for nonlinear problems: Affine invariance and adaptive algorithms [M], Springer Science & Business Media, 2011.
16.
LuenbergerD.G., Optimization by vector space methods [M], John Wiley & Sons, 1997.
