Figure 1
- Immersed geometry on a staggered finite volume computational grid. Interpolation of the Eulerian velocity field to a Lagrangian point.
Figure 2
- Immersed geometry on a staggered finite volume computational grid. Distribution of the Lagrangian forces to the Eulerian field.
Figure 4
- Correct distribution of Lagrangian forces in a parallel approach. Step 2.
Figure 3
- Correct distribution of Lagrangian forces in a parallel approach. Step 1.
Figure 5
- Correct distribution of Lagrangian forces in a parallel approach. Step 3.
Figure 7
-L2 norm of the Zero Mach Number manufactured solution. ■: u, ▲: L2 of the immersed boundary. The solid line stands for the second order decay.
Figure 6
- Immersed geometry in the computational domain. Parallel computing is used. Each one of the Cartesian block is a subdomain assigned to a different CPU or processor core.
Figure 8
- Scalar variable c for the cold jet: instantaneous (a) and averaged (b) fields.
Figure 9
- Streamwise evolution of average velocity and turbulent kinetic energy profiles for the cold jet. The experimental results are extracted from Chen et al. (1996). 1In the present work, a simulation is said to diverge when the size of the time step continuously decreases from several orders of magnitude.
Figure 10
- Radial variation of average velocity and kinetic energy profiles for the cold-flow F3 jet. The solid lines are results of the present work, the plus, (+), symbols are numerical results obtained by Yilmaz (2008) and the circles are experimental results of Chen et al. (1996).
Figure 11
- Cross section showing the velocity vectors in a Poiseuille flow.
Figure 13
- Quantitative comparison of the velocity profile for a circular Poiseuille flow, obtained with the Immersed Boundary Method against an analytical solution.
Figure 12
- Poiseuille flow. Computational domain and u-component of the velocity.
Figure 14
- Eulerian and Lagrangian mesh and control volumes.
Figure 16
- Velocity vectors and isovalue of Q=300 for three distinct Reynolds numbers: (a)Re=200, (b)Re=400 and (c)Re=1000.
Figure 15
- Velocity field and streamlines for the flow over the sphere, for three values of the Reynolds numbers: (a)Re=200, (b)Re=400 and (c)Re=1000.
Figure 17
-L2 norm of the velocity at the immersed boundary for different Reynolds numbers.
Figure 19
- Drag coefficient as a function of the Reynolds number: Re = 10 up to Re = 103 (Fornberg 1988, White 1999, Subramanian 2003, Campegher 2005).
Figure 18
- Time distribution of the drag coefficient for four values of the Reynolds number.
Figure 21
- Original rotor design.
Figure 20
- Drag coefficient as a function of the Reynolds number: Re = 10 up to Re = 106; present work and White (1999).
Figure 22
- Three-dimensional views of the fan; surface triangular mesh used. 2 We may recall that this pressure variation acting on a surface will impose a fluctuating force on the structure with a characteristic frequency. If this frequency is the same of any of the natural frequencies of the structure we may have the development of cracks or even catastrophic failure.
Figure 23
- Original (yellow) and modified (green) rotor design.
Figure 25
- Instantaneous field of velocity magnitude and isosurfaces of the criteria Q = 106 at t = 0:34s. Left: original rotor; right: modified rotor. 3 It is important to keep in mind that this flow is not permanent and it will not achieve a permanent regime, however, we may achieve a regime statistically permanent.
Figure 24
- Mean pressure and mean velocity fields magnitude.
Figure 27
- Averaged field of pressure; left: original rotor; right: modified rotor.
Figure 26
- Instantaneous field of pressure at t = 0:34s; left: original rotor; right: modified rotor.
Figure 29
- Averaged field of turbulent kinetic energy; left: original rotor; right: modified rotor.
Figure 28
- Variations in the mean pressure along the rotor for the original and modified geometries.
Figure 30
- Positions of the numerical probes in the computational domain.
Figure 32
- Energy spectra for probe 2; (a): in linear scale; (b): log scale.
Figure 31
- Energy spectra for probe 1. Left: in linear scale; right: log scale.
Figure 34
- Energy spectra for probe 5; (a): in linear scale; (b): log scale.
Figure 33
- Energy spectra for probe 3; (a): in linear scale; (b): log scale.
Figure 36
- Energy spectra for probe 4; (a): in linear scale; (b): log scale.
Figure 35
- Energy spectra for probe 6; (a): in linear scale; (b): log scale.
