Air trapped wave impact in a depressurised tank

Air trapped wave impact in a depressurised tank image

In this test case an air trapped wave impact event is generated in a sloshing tank. The tank has a pure horizontal motion (sway) with a sinusoidal law. After a few cycles a plunging breaker was generated, the wave trapped a large air pocket and impinged at the vertical tank wall. The trapped air pocket underwent alternating compression and expansion, deformation and breakup, producing pulsating positive and negative local gauge pressures and total forces on the tank wall. This test case was numerically studied by Ma et al. (2016) using the compressibleInterFoam solver in OpenFOAM2.3.0.

Contributors

  • Zhihua Ma
Contact person
Zhihua Ma

Description

In this test case an air trapped wave impact event is generated in a sloshing tank. The tank has a pure horizontal motion (sway) with a sinusoidal law. After a few cycles a plunging breaker was generated, the wave trapped a large air pocket and impinged at the vertical tank wall. The trapped air pocket underwent alternating compression and expansion, deformation and breakup, producing pulsating positive and negative local gauge pressures and total forces on the tank wall. This test case was numerically studied by Ma et al. (2016) using the compressibleInterFoam solver in OpenFOAM2.3.0.

Experimental Set-up

The physical experiments were performed by Lugni et al. (2010a, 2010b) in the INSEAN-Italian Ship Model Basin. A square sloshing tank with width of 1 metre underwent pure horizontal motion (sway) with a sinusoidal law x=A sin⁡(2πt/T). The excitation amplitude is A=3 cm and the period of sway motion is T=1.6 s. The tank is closed and partially filled with water of depth 0.125 m. The ullage pressure in the tank is set to 0.75 bar. Five pressure transducers are installed on the left wall of the tank located at y=5, 25, 45, 65 and 85 mm (measured from the free surface).

Experimental Test Program

To precisely capture the important flow features especially the air pocket trapped in the water mass a non-uniform fine mesh is used to discretise the domain. The horizontal direction of the tank is uniformly divided with 500 mesh cells. In the vertical direction, the lower part (40%) and upper part (60%) of the tank are regularly discretised by 200 cells and 60 cells, respectively. The time step is fixed to be Δt = 10 μs throughout the numerical simulation in order accurately to resolve the transient phenomenon of wave impact. All the boundaries are treated with an adiabatic no-slip solid wall condition. Lugni et al. (2010a) did not clarify the environmental temperature for their experiments, while it is initially set to be 15°C everywhere.

Physical Measurement Data

The experimental data of the measured pressures at five gauges is available in the work of Lugni et al. (2010a). Please see the references. The data has been extracted from their work and can be downloaded here: P1, P2, P3, P4 and P5 (gauge pressure).

Numerical Benchmarks

The numerical results have been produced by Ma et al. (2015) using the compressible two-phase code compressibleInterFoam and the incompressible two-phase code interFoam included in OpenFOAM 2.3.0. The computed absolute pressures at the five sensors can be found here.

NOTE: When using this data please be sure to cite the original source of the data appropriately (see full citations in the 'Relevant References' section below).

Relevant References

Ma, Z.H., Causon, D.M., Qian, L., Mingham, C.G., Martinez-Ferrer, P., Numerical investigation of air enclosed wave impacts in a depressurised tank. Ocean Engineering 126 (2016): 15-27. Doi: http://dx.doi.org/10.1016/j.oceaneng.2016.06.044.

Lugni, C., Brocchini, M., Faltinsen, O.M., Evolution of the air cavity during a depressurized wave impact. II. The dynamic field. Phys. Fluids 22 (2010a): 056102. Doi: http://dx.doi.org/10.1063/1.3409491.

Lugni, C., Miozzi, M., Brocchini, M., Faltinsen, O.M., Evolution of the air cavity during a depressurized wave impact. I. The kinematic flow field. Phys. Fluids 22 (2010b): 056101. Doi: http://dx.doi.org/10.1063/1.3407664.