Sloshing of oil storage tanks is a great threat to industrial complexes when huge earthquakes occur. Overflow of oil due to sloshing could lead to serious tank fire and to oil dispersion due to Tsunamis. Since liquid sloshing is characterized as a highly nonlinear free-surface flow, numerical approaches are more suitable than theoretical ones to estimate the overflow from oil storage tanks caused by sloshing. The moving particle simulation (MPS) method, which is a mesh-free and Lagrangian particle method, is applied to the sloshing simulation. Thanks to the perfect mass conservation of the particle method, the overflow amount of oil can be accurately estimated by simply counting the number of overflowed fluid particles.
There is high-possibility of occurrence of huge earthquakes along the Nankai Trough subduction zone, and the worst scenario is presumed as MW9.0, which is the same as the 2011 Tohoku earthquake. In order to simulate sloshing of oil storage tanks subjected to a possible Nankai Trough earthquake, ground shaking characteristics of the site need to be evaluated. Therefore we carried out microtremor measurements with high density in an industrial complex in Osaka bay area. A strong ground motion during a scenario earthquake with MW9.0 along the Nankai Trough is estimated based on the SMGA models considering the empirical site amplification and phase effects.
Fully nonlinear sloshing simulations, using an in-house GPU computing code based on the explicit MPS method, are performed for a large-size oil storage tank installed with a floating roof by imposing the estimated strong ground motion of Nankai Trough earthquake as the external oscillation. As a result, significant amount of oil flows onto the floating roof and large amount of oil overflow from the tank is confirmed. The numerical result suggests the entire fire is one of major concerns for the assumed earthquake scenario. Finally sloshing simulations are repeated for small- and medium-size tanks, and then the total amount of outflow oil is estimated for the industrial complex of interest in the estimated Nankai Trough earthquake.
[1]
Kojiro Irikura,et al.
Source Characterization for Broadband Ground-Motion Simulation: Kinematic Heterogeneous Source Model and Strong Motion Generation Area
,
2003
.
[2]
Stephen M. Longshaw,et al.
DualSPHysics: Open-source parallel CFD solver based on Smoothed Particle Hydrodynamics (SPH)
,
2015,
Comput. Phys. Commun..
[3]
S. Zama,et al.
Experimental Study on the Estimation of the Amount of Overflowed Oil due to Liquid Sloshing of Oil Storage Tank
,
2008
.
[4]
K. Ichii,et al.
Variation of earthquake ground motions within a very small distance
,
2014
.
[5]
Takashi Nagao,et al.
SIMULATION OF STRONG GROUND MOTIONS BASED ON SITE- SPECIFIC AMPLIFICATION AND PHASE CHARACTERISTICS
,
2006
.
[6]
K. Aki.
Scaling law of seismic spectrum
,
1967
.
[7]
S. Koshizuka,et al.
Moving-Particle Semi-Implicit Method for Fragmentation of Incompressible Fluid
,
1996
.
[8]
Seiichi Koshizuka,et al.
Numerical Simulation of Three-Dimensional Free-Surface Flows with Explicit Moving Particle Simulation Method
,
2011
.
[9]
A. Mussett,et al.
The paleomagnetism of the Turkana lavas
,
1966
.
[10]
Kojiro Irikura,et al.
EVALUATION OF HORIZONTAL AND VERTICAL STRONG GROUND MOTIONS USING EMPIRICAL SITE-SPECIFIC AMPLIFICATION AND PHASE CHARACTERISTICS
,
1998
.
[11]
Masashi Kashiwagi,et al.
Numerical simulation of wave-induced nonlinear motions of a two-dimensional floating body by the moving particle semi-implicit method
,
2008
.