Small-scale wind farms
Posted in Research by Kate Archdeacon on December 1st, 2009
Source: Environmental Research Web
Image via physicsworld
From “Fish inspire wind farm configuration”, Edwin Cartlidge
Conventional wind turbines work best when located as far as possible from the destructive vortices of neighbouring turbines. However, a pair of scientists in the US have worked out that the performance of other kinds of turbine actually improves when they are placed close to one another, concluding that wind farms could therefore be made much smaller than they are today. The familiar propeller-like turbine with a horizontal axis of rotation can convert 50% or more of the energy from the wind that it is exposed to. In a wind farm, however, the wake from one turbine will disturb the air reaching the blades of its neighbours meaning that turbines must be placed far apart.
A less familiar family of turbines have a vertical axis of rotation. Individually, these vertical-axis turbines are less efficient than the horizontal-axis devices because only part of the turbine can be pushed by the wind at any one time, and they have therefore proven far less popular. However, these turbines have a significant advantage over the horizontal-axis variety – their power output can be increased when they are placed very close to one another. Now, Robert Whittlesey and John Dabiri of the California Institute of Technology have worked out how best to arrange such closely spaced turbines by drawing on the work of aeronautical engineer Daniel Weihs, who showed in the 1970s how fish save on energy by swimming within schools. Such fish form a series of offset rows, and Weihs found that fish get carried forward by the vortices created by the swimming motion of their two closest companions in the row immediately in front of them. Whittlesey and Dabiri wondered whether the relative spacing of vortices produced by an individual fish might serve as a good template for the arrangement of vertical-axis turbines within a wind farm and set up a computer model to test this idea.
Read the full article by Edwin Cartlidge.
