Openwind contains a wind-flow model based on the NOABL code (Phillips, 1979)3. The model attempts to conserve momentum while minimizing divergence. It takes account of terrain, roughness and atmospheric stability and has been modified in Openwind to be able to model displacement heights.
Users who wish to create an Openwind wind map should right-click in the tree-view and choose New Layer->Wind Map. This will create an empty, uninitialized WindMapLayer. They can then add one or more terrain layers and roughness layers by dragging and dropping them onto the WindMapLayer. If they do not want to use the roughness data, they can set the default roughness in the Options dialog later. Lastly, they will need to add one or more MetMastLayers.
Now that the WindMapLayer has access to all the information it needs to calculate a wind flow, the user can right-click on the WindMapLayer and choose Calculate. At this point, the cursor becomes a cross with a box made up of dotted lines. Use this drag-box cursor to define the area in the map-view that should be covered by the new WRG by pressing the left mouse button at the top left of the desired area and then holding the button down as you drag the cursor to the bottom right of the desired area.
The dialog shown in figure 86 appears once the user has let go of the mouse button. This dialog allows the user to edit the selected area manually, and to set the spacing between adjacent grid nodes. The model works by iterating a 3-dimensional grid, making it possible to extract the mean wind speeds at any height above ground level. The dialog allows the user to specify up to six output heights although of course it is not recommended to use wind flow models to shear data from met height to hub height. The memory requirements are there as a rough guide.
Of the Model Options, the default roughness length may be of the greatest interest. This is the roughness length assigned to any area in the wind map calculation area that does not have a roughness already defined, either via a RasterLayer or a PolygonLayer.
Displacement height works by elevating the terrain by the height of the vegetation (accessed from a RasterLayer or a PolygonLayer) for the duration of the calculation and then subtracting the height of the vegetation again before the wind speed grids are output. Areas with a displacement height should show up as areas of slower wind-speed in the finished WRG. The modification of the terrain includes a ramp up and a ramp down, both in the same direction as the wind for this calculation step. The slope and extent of these ramps are determined either by the user setting the reciprocal of the recovery grade which is the same as the recovery length in terms of the number of displacement heights that it takes to recovery. Some people prefer to think of these ramps in terms of a fixed gradient and others prefer to think in terms of the distance in multiples of the obstacle height.
One issue with using displacement height is that you need to rely on the software to shear the wind speeds according to the roughness length and that may not be in agreement with what is seen at the met masts. To attempt to address this we have added an option to “Correct initial grid to match met mast vertical profiles”. This option is applied at the end of step 3 below and applies a correction across the wind map at every height such that the shear profiles in the initial wind field match those at the multi-height met masts used in the wind map calculation. It is important not to use this option is you do not have full 10-200m met mast objects.
Displacement height is linked to vegetation height but is not always equal to the vegetation height. For this reason, it is possible to set a scale factor. By default, it is set to 1.0 but is quite often set to something around 0.7 or 0.8.
The calculation itself consists of several steps which are repeated for each direction:
1.First, the roughness around each mast is used to shear the met mast data up to the top of the at the atmospheric boundary layer which, for the purposes of this model is set at 200 m above ground level by default.
2.The values at 200 m height at the met masts are then blended across all nodes in the wind map at 200 m a.g.l. using the inverse-distance squared weighted sum of all the met masts.
3.These 200m values are then sheared down so an entire column of values are found for each node in the wind map. The top height is the height of the atmospheric boundary layer set in the Wind Map Options (default = 200 m). The bottom height is approximately 10m above ground level depending on the complexity of the terrain used. This is then the initial wind field for a given wind direction.
4.Then, the model is iterated for each direction until the desired convergence level is reached or until it is clear that the model is no longer improving (this will result in warnings).
5.Finally, output WRGs are extracted from the 3-D grid along with XYDRasterLayers containing estimates of average inflow angles for each calculation height specified in the Wind Map Options dialog.
Once the calculation is complete, each WindMapLayer can be used in place of a WRGLayer for energy capture calculations as long as it has access to a MetMastLayer.