Nonstationary Processes Studying Based on “Caterpillar”—SSA Method

Vibration Engineering and Technology of Machinery: Proceedings of VETOMAC X 2014. Mechanisms and Machine Science. Vol.23 2015
Svetlana Polukoshko, G Hilkevica, Vladimirs Gonca

This paper deals with “Caterpillar”-SSA method—a new powerful model-free method of time series analysis and forecasting. It allows to decompose the nonstationary time series into trend, periodic components (sum of elementary harmonic), noise (random components) and to forecast subsequent behavior of system, if separability condition of time series and possibility of reverse synthesis of individual elements are fulfilled. This paper shows application of this method for random and nonlinear vibrations study on the example of building element vibration under seismic action.

Atslēgas vārdi
Forecasting, Harmonic component, Noise, Nonstationary time series, Singular spectrum analysis, Trend
DOI
10.1007/978-3-319-09918-7_88
Hipersaite
https://link.springer.com/chapter/10.1007%2F978-3-319-09918-7_88

Polukoshko, S., Hilkevica, G., Gonca, V. Nonstationary Processes Studying Based on “Caterpillar”—SSA Method. No: Vibration Engineering and Technology of Machinery: Proceedings of VETOMAC X 2014. Mechanisms and Machine Science. Vol.23, Angilja, Manchester, 9.-11. septembris, 2014. Manchester: Kluwer Academic Publishers, 2015, 999.-1008.lpp. ISBN 9783319099170. ISSN 2211-0984. Pieejams: doi:10.1007/978-3-319-09918-7_88

Publikācijas valoda
English (en)