ASK THE EXPERTS: Average Wind Speed

Rayleigh distribution curve
Wind speed distributions can be used to approximate real-world conditions (and therefore, wind system performance) at a site.
Rayleigh distribution curve

It’s common to read about “average wind speed” when deciding whether or not it is worth installing a wind generator at a given site. But it seems unrealistic to talk about average wind speeds, as such. Obviously, if you had winds of, say, 25 mph for half a given amount of time, and 5 mph the other half, the power output would not be the same as if the wind were a constant 15 mph—there would be 125 times as much (theoretical) power generated during the time the wind was at 25 mph than during the time it was blowing at only 5 mph. So shouldn’t this be taken into account when analyzing a potential wind site?

Furthermore, does it make sense to you that a wind system at my home could theoretically generate as much power in one or two days of 40 to 50 mph winds than could be generated over the rest of the year’s average winds of less than 5 mph? After all, a 50 mph wind would generate (again, theoretically) 1,000 times the power that a 5 mph wind would generate over the same time span.

Malcom Drake • via email

You are absolutely correct that average wind speed alone does not provide enough information to determine energy potential. What is required is the distribution, which specifies the fraction of the time the wind speed spends in a particular range. Because it takes years of recorded data to provide an accurate distribution for a site, assumptions are often made. The typical assumption (based on much experience and measurement) is that the wind speed follows a predictable distribution—a common one that is fairly accurate for most sites is the Rayleigh distribution. For small wind turbines, a Rayleigh distribution is what the Small Wind Certification Council uses for the certified annual energy production curve based on the certified power curve. If an average wind speed is provided with no other context, the assumption is most likely a Rayleigh distribution.

As for building a turbine for rare, high winds—the economics are likely unworkable. The unit would have to be very stout—and hence prohibitively expensive—and it would sit idle most of the time, as the startup speed would be high due to its mass and inefficiency in light wind. When the winds did blow fast enough, it would indeed produce a lot of power, but then you’d have to find a way to store the energy. Energy is power multiplied by time: a large amount of power for a small period of time could be equal to small amounts of power for longer periods of time, and a wind turbine designed for the latter scenario would be much cheaper.

David Laino • Cofounder, Endurance Wind Power

Comments (0)