wind energy resource evaluation in a site of central italy ... - WindSim

More documents

Recommendations

Info

Figure 3.15: Mean velocity profile over the flat floor at inlet position.3.5.2 Numerical methodThe WindSim software solves the RANS equations (Reynolds Averaged NavierStokes), using the finite volume method through a graphical interface and the CFDsolver Phoenics. The continuity equation and the three momentum equations aresolved, using the standard k-ε turbulence model to introduce the equationsneeded to solve the Reynolds stress tensor.Also the “Yap” correction for k-ε turbulence model is needed to help theconvergence process in the separated flow. In addition to the turbulence model, thewall function is imposed to the ground.The fluid blown by the wind tunnel is air, in incompressible flow condition: in thisway, no thermal effects are present. The characteristic of this flow is expressed bythe Reynolds Number, defined as3.5.4) = ∙ =1,87×10where:U ∞[m/s] free stream velocity39
H [m] hill heightν[m 2 /s] air cinematic viscosityThe numerical grid is built, in both S3H4 and S5H4 cases, with a non-orthogonal gridas shown in the further paragraphs.3.5.3 Case S3H4The case S3H4 has been studied in a geometrical model extending from streamwisecoordinates -1,5 m to 1,5 m, for a total length of 3 m. The width is 0,1 m while thetotal height is 2 m; thus, the blockage factor is 2%.The grid contains 130 cells in x-directions, 3 cells in y-direction and 60 cells invertical direction; the horizontal resolution has been increased in the hill profile, inorder to describe in a sufficient accurate way the behavior of the flow in the criticalzone. The height of the first cell is 2 mm, while the roughness height imposed in thewall function formula is 0,05 mm. Then, the condition stating that the first cellheight must be at least double of the roughness value is satisfied [1].The following figure 3.16 shows a particular of the numerical grid.Figure 1.16: Particular of the non-orthogonal regular grid for case S3H4.The boundary condition imposed on the top of the model is “wall with no-friction”,while at the bottom the terrain contains the roughness information all over thesurface.40
Page 1 and 2: UNIVERSITÀ DEGLI STUDI DI CAGLIARI
Page 3 and 4: IndexIndex page IAcknowledgmentsVII
Page 5 and 6: CHAPTER 4: Characteristics of the s
Page 7 and 8: 6.7.1 Mean wind velocity page 1096.
Page 9 and 10: This thesis is the result of a prac
Page 11 and 12: WindSim is a software developed by
Page 13 and 14: CHAPTER 1. Atmospheric Boundary Lay
Page 15 and 16: Figure 1.1: Example of an atmospher
Page 17 and 18: 1.5 The WindSim numerical codeWindS
Page 19 and 20: (κ−ε RNG or the Low Reynolds),
Page 21 and 22: 1.5.4 RESULTSThe Results module all
Page 23 and 24: Figure 1.2: Example of TKE vertical
Page 25 and 26: CHAPTER 2. Simulations over flat te
Page 27 and 28: inlet position (0 meters), station
Page 29 and 30: 2.4 Height distribution factor sens
Page 31 and 32: e decreased, to avoid to build thin
Page 33 and 34: Figure 2.10: Vertical profiles for
Page 35 and 36: CHAPTER 3. Simulations over 2D hill
Page 37 and 38: Table 3.1: Measurement coordinatesS
Page 39 and 40: The results shown above demonstrate
Page 41 and 42: Figure 3.6: Sketch of the 15 km lon
Page 43 and 44: Figure 3.9: Influence of the height
Page 45 and 46: Figure 3.13: Zoom of the hill top r
Page 47: 3.5.2) = ∙.The two models for att
Page 51 and 52: Figure 3.18: Particular of the nume
Page 53 and 54: Finally, the horizontal x velocity
Page 55 and 56: Figure 3.25: Z velocity componentFi
Page 57 and 58: 3.6 RemarksA CFD study of a boundar
Page 59 and 60: Some experimental wind data of this
Page 61 and 62: Both the pictures 4.3 and 4.4 have
Page 63 and 64: Table 4.2: Description of the color
Page 65 and 66: In this periods, data are recorded
Page 67 and 68: In the following pages, a short des
Page 69 and 70: 4.3.2 Mast 5424-15 m Period April-M
Page 71 and 72: 4.3.4 Mast 5424-15 m Period Novembe
Page 73 and 74: 4.3.6 Mast 5428-40 m Period April-M
Page 75 and 76: 4.3.8 Mast 5428-50 m Period Novembe
Page 77 and 78: 4.3.10 Mast 5428-30 m Period Novemb
Page 79 and 80: 4.4.2 Mast 542870
Page 81 and 82: 4.5.2 Mast 542872
Page 83 and 84: CHAPTER 5. Test of the Free Stream
Page 85 and 86: case the wind is blowing from the e
Page 87 and 88: Geometrical modelTable 5.3: Input p
Page 89 and 90: Figure 5.6: Vertical profiles in po
Page 91 and 92: The orthogonalization of the grid h
Page 93 and 94: Figure 5.9: Map of speed-ups for di
Page 95 and 96: Figure 5.30: Speed-2D at 30 m heigh
Page 97 and 98: Figure 5.12: Speed-2D at 30 m heigh
Page 99 and 100:
CHAPTER 6. Analysis of the wind fie
Page 101 and 102:
Table 6.2: Boundary conditions and
Page 103 and 104:
6.5 Vertical profiles of VelocityTo
Page 105 and 106:
characterized by high wind speed (5
Page 107 and 108:
From the error analysis, it is foun
Page 109 and 110:
Anyhow, the trend of the linear int
Page 111 and 112:
Concerning the CFD simulations, val
Page 113 and 114:
Analyzing RMS error table for tower
Page 115 and 116:
A general overview of the figure 6.
Page 117 and 118:
6.7 Cross Checking of wind dataIn t
Page 119 and 120:
Figure 6.8 a/bFigure 6.8 c/dFigure
Page 121 and 122:
Table 6.11: RMS error for cross che
Page 123 and 124:
Figure 6.9 e/fFigure 6.9 g/hFigure
Page 125 and 126:
Figure 6.10 g/hFigure 6.10 i/lFigur
Page 127 and 128:
to have a better agreement during t
Page 129 and 130:
Figure 7.2: Weibull fitting of the
Page 131 and 132:
Figure 7.5: Perspective view of the
Page 133 and 134:
The calculated AEP has been reporte
Page 135 and 136:
ConclusionsIn this thesis the evalu
Page 137 and 138:
Bibliography[1] CRASTO G. - Numeric
show all

wind energy resource evaluation in a site of central italy ... - WindSim

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?