Back to Search
Start Over
Predicting flow-induced vibrations of tandem square cylinders using finite element simulations and data-driven neural network model.
- Source :
-
Ocean Engineering . Sep2024, Vol. 308, pN.PAG-N.PAG. 1p. - Publication Year :
- 2024
-
Abstract
- This study investigates the dynamic responses and associated wake structures of three identical tandem square cylinders oscillating in the cross-flow direction at a Reynolds number, R e = 150. The reduced velocity is varied within the range U ∗ = 3 − 14 , with a mass ratio of m ∗ = 2 considered. Finite element simulations are conducted for four streamwise gap configurations: L x = 3 B , 5 B , 8 B and 10 B , where B represents the cylinder's side length. Using the r m s values of amplitude and lift coefficient obtained for the above-mentioned streamwise gaps, a data-driven artificial numerical model is created to predict the amplitudes and lift coefficients for L x = 4 B , 6 B , 7 B and 9 B. For smaller streamwise gaps ( L x ≤ 4 B), all three cylinders exhibit four distinct oscillation regimes. However, the larger gaps promote vortex interaction with the downstream cylinders, leading to the emergence of different vibration regions for each body. The upstream cylinder experiences higher amplitudes for L x = 3 B compared to the L x = 10 B case, resulting in an upper response branch for smaller gap spacings. In contrast, the larger gaps result in less intense interference effects on the upstream cylinder, causing periodic oscillations across the entire U ∗ regime considered. Nevertheless, the two downstream bodies undergo frequency modulation in their vibration responses for larger gaps. These observations align with the periodic shedding of the upstream cylinder and modulations in the Strouhal frequency of the downstream bodies observed in the static system for L x > 4 B. • Three tandem square cylinders exhibit four oscillation regions with the gap of 3B • Vibrations are influenced by the flow characteristics of stationary system for larger gaps • FE simulations and a data-driven ANN model predict higher amplitudes for smaller gaps [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 00298018
- Volume :
- 308
- Database :
- Academic Search Index
- Journal :
- Ocean Engineering
- Publication Type :
- Academic Journal
- Accession number :
- 177908869
- Full Text :
- https://doi.org/10.1016/j.oceaneng.2024.118237