One of the key factors in simulating realistic wall-bounded flows at high Reynolds numbers is the selection of an appropriate turbulence model for the steady Reynolds Averaged Navier–Stokes equations (RANS) equations. In this investigation, the performance of several turbulence models was explored for the simulation of steady, compressible, turbulent flow on complex geometries (concave and convex surface curvatures) and unstructured grids. The turbulence models considered were the Spalart–Allmaras model, the Wilcox k- ω model and the Menter shear stress transport (SST) model. The FLITE3D flow solver was employed, which utilizes a stabilized finite volume method with discontinuity capturing. A numerical benchmarking of the different models was performed for classical Computational Fluid Dynamic (CFD) cases, such as supersonic flow over an isothermal flat plate, transonic flow over the RAE2822 airfoil, the ONERA M6 wing and a generic F15 aircraft configuration. Validation was performed by means of available experimental data from the literature as well as high spatial/temporal resolution Direct Numerical Simulation (DNS). For attached or mildly separated flows, the performance of all turbulence models was consistent. However, the contrary was observed in separated flows with recirculation zones. Particularly, the Menter SST model showed the best compromise between accurately describing the physics of the flow and numerical stability.
more »
« less
Computational fluid dynamics-based modeling of liquefied soils
The residual shear strength of liquefied soil is a key parameter in evaluating liquefaction flow failures. Results from a series of dynamic centrifuge experiments where the shear strength of liquefied soil was inferred by measuring the force required to pull a thin metal plate (coupon) horizontally through the liquefied soil are assessed here using a computational fluid dynamics (CFD) based model. Viscosity is a key parameter for the Newtonian fluid constitutive model used in the simulations, and apparent viscosities of liquefied soil in the range of about 5,800 – 13,300 Pa·s were obtained when the CFD model was calibrated against coupons pulled through liquefied soil in dynamic centrifuge tests. These computational values agree reasonably with apparent viscosities of liquefied soil reported in the literature when the Reynold’s numbers exceeded 1.0. Importantly, the CFD simulations illustrated that in cases where Reynold’s numbers are < 1.0, apparent viscosities of liquefied soil back-calculated using simplistic closed-form solutions commonly applied in geotechnical literature are several orders of magnitude too large; and therefore, such closed-form solutions should not be used for these cases.
more »
« less
- Award ID(s):
- 1728172
- NSF-PAR ID:
- 10104124
- Date Published:
- Journal Name:
- 7th International Conference on Earthquake Geotechnical Engineering
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract Seismic design of water retaining structures relies heavily on the response of the retained water to shaking. The water dynamic response has been evaluated by means of analytical, numerical, and experimental approaches. In practice, it is common to use simplified code‐based methods to evaluate the added demands imposed by water sloshing. Yet, such methods were developed with an inherent set of assumptions that might limit their application. Alternatively, numerical modeling methods offer a more accurate way of quantifying the water response and have been commonly validated using 1 g shake table experiments. In this study, a unique series of five centrifuge tests was conducted with the goal of investigating the hydrodynamic behavior of water by varying its height and length. Moreover, sine wave and earthquake motions were applied to examine the water response at different types and levels of excitation. Arbitrary Lagrangian‐Eulerian finite element models were then developed to reproduce 1 g shake table experiments available in the literature in addition to the centrifuge tests conducted in this study. The results of the numerical simulations as well as the simplified and analytical methods were compared to the experimental measurements, in terms of free surface elevation and hydrodynamic pressures, to evaluate their applicability and limitations. The comparison showed that the numerical models were able to reasonably capture the water response of all configurations both under earthquake and sine wave motions. The analytical solutions performed well except for cases with resonance under harmonic motions. As for the simplified methods, they provided acceptable results for the peak responses under earthquake motions. However, under sine wave motions, where convective sloshing is significant, they underpredict the response. Also, beyond peak ground accelerations of 0.5 g., a mild nonlinear increase in peak dynamic pressures was measured which deviates from assumed linear response in the simplified methods. The study confirmed the reliability of numerical models in capturing water dynamic responses, demonstrating their broad applicability for use in complex problems of fluid‐structure‐soil interaction.
-
more » « less
Abstract The susceptibility of a granular soil to suffusion is strongly dependent on its grain size distribution (GSD) and the mechanical and hydraulic conditions it is subjected to. This study investigates the onset of suffusion considering the effect of confining pressure and stress anisotropy using a fully resolved computational fluid dynamics and discrete element method (CFD–DEM). Three benchmarks, including the sedimentations of single and two adjacent spheres and the classic one‐dimensional (1D) consolidation are performed to demonstrate the capability of this method for high‐fidelity particle‐fluid simulations. A modified hydraulic criterion for the onset of suffusion considering stress anisotropy is presented. The microstructural changes of soil specimens before and during global suffusion are inspected, with emphasis on the evolutions of particle kinetic energy and displacements, force chain networks, and stress anisotropy. We found that the critical hydraulic gradient is negatively correlated with the confining pressure and the degree of stress anisotropy. Fine particles in the soil matrix are locally detached at small hydraulic gradients before the apparent global suffusion, as manifested by the variation of particle kinetic energy and coordination numbers. The roles of different contact types on force transmission and stress anisotropy in eroded specimens are also examined.
-
more » « less
Abstract Background Hemocompatibility‐related adverse events (HRAE) occur commonly in patients with left ventricular assist devices (LVADs) and add to morbidity and mortality. It is unclear whether the outflow graft orientation can impact flow conditions leading to HRAE. This study presents a simulation‐based approach using exact patient anatomy from medical images to investigate the influence of outflow cannula orientation in modulating flow conditions leading to HRAEs.
Methods A 3D model of a proximal aorta and outflow graft was reconstructed from a computed tomography (CT) scan of an LVAD patient and virtually modified to model multiple cannula orientations (
n = 10) by varying polar (cranio‐caudal) (n = 5) and off‐set (anterior–posterior) (n = 2) angles. Time‐dependent computational flow simulations were then performed for each anatomical orientation. Qualitative and quantitative hemodynamics metrics of thrombogenicity including time‐averaged wall shear stress (TAWSS), oscillatory shear index (OSI), endothelial cell platelet activation potential (ECAP), particle residence time (PRT), and platelet activation potential (PLAP) were analyzed.Results Within the simulations performed, endothelial cell activation potential (ECAP) and particle residence time (PRT) were found to be lowest with a polar angle of 85°, regardless of offset angle. However, polar angles that produced parameters at levels least associated with thrombosis varied when the offset angle was changed from 0° to 12°. For offset angles of 0° and 12° respectively, flow shear was lowest at 65° and 75°, time averaged wall shear stress (TAWSS) was highest at 85° and 35°, and platelet activation potential (PLAP) was lowest at 65° and 45°.
Conclusion This study suggests that computational fluid dynamic modeling based on patient‐specific anatomy can be a powerful analytical tool when identifying optimal positioning of an LVAD. Contrary to previous work, our findings suggest that there may be an “ideal” outflow cannula for each individual patient based on a CFD‐based hemocompatibility profile.
-
The standard of practice when assessing the liquefaction susceptibility of geosystems uses an empirical case history database that was primarily developed for clean, poorly graded sands. However, many geosystems in the built environment are either constructed with or founded on well graded soils, creating a disconnect between the sand encountered in practice and the sand used as the basis of knowledge. Using the 9-m centrifuge at the University of California Davis’s Center for Geotechnical Modeling a centrifuge experiment was designed to test the dynamic response of embankments constructed poorly graded and well graded sands at the system level scale. The experiment consisted of two 10-degree slopes, one constructed with a poorly graded sand and the other with a well graded sand positioned side by side in the same model container. Each slope was dry pluviated to the same relative density of Dr=63%, while the absolute densities were different. The slopes were instrumented with dense arrays of pore pressure transducers and accelerometers in the level ground at the head of the slope. The stress-strain behavior between accelerometers was calculated using inverse analysis techniques, providing a 1-D shear-beam soil response at the sensor array location. Liquefaction was triggered, as defined by an excess porewater pressure ratio (ru) of 1.0, but the shear strains at triggering in the well graded sand were significantly less than the strains in the poorly graded sand. During cyclic mobility, strain accumulation in the well graded sand occurred at a slower rate. This study demonstrates that liquefaction triggering and the post-triggering response for saturated sands needs to consider gradation characteristics and clean poorly graded sands cannot act as a single predictor of dynamic response for all sand gradations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
