Simulation of liquid sloshing in model LNG tank using smoothed particle hydrodynamics

The Smoothed Particle Hydrodynamics (SPH) method is applied to the problem of modeling sloshing in a 2-D water model that is a representation of a scaled LNG tank. Two different tank configurations are considered with 2 fill levels (20%, 70%) and 2 different oscillation amplitudes (10% and 20% of the tank dimension). Predicted pressure signals are compared to experimental measurements. The peak pressure values predicted in the simulations are generally lower than the experimental values, although they are the correct order of magnitude. While the stochastic nature of the oscillations in practice means that an exact match between simulation and experiment is not feasible, the statistics of the pressure signals allow a more meaningful comparison. Also presented are ensemble-averaged pressure traces and standard deviations of the pressure from the simulation results and a sub-set of the experimental results, and these show generally higher variability in the pressure signals for low fill ratios compared to high fill ratios. There is generally good agreement between the simulation and experimental ensemble mean and standard deviation results. The magnitude of fluctuations is also sensitive to the sensor location. SPH is seen to distinguish between the different flow cases; it provides results for the peak pressures that are the correct order of magnitude on average, although the highest peaks are underpredicted. It is a natural numerical technique for coupled fluid-structure problems with large free-surface deformations.