Why “Average Wind Speed” Can Give You the Wrong Answer About Wind Energy

Patel’s treatment of wind-resource analysis makes this point directly: because wind power varies with the cube of speed, the frequency distribution of wind speed must be known to determine the wind power potential of a site.

That is a fundamentally better question than simply asking:

“What’s the average wind speed?”

A Histogram Can Tell a Story an Average Cannot

Imagine two wind-speed histograms.

Both have the same mean.

The first is narrow and concentrated around 6 m/s.

The second is spread widely, with many low-wind observations but also a meaningful number of much stronger winds.

The single-number average can make them look equivalent.

The distributions immediately reveal that they are not.

Advertisement. Scroll to continue reading.

This matters because each wind-speed bin can be treated separately.

For a wind speed v_i occurring with probability or frequency f_i, the contribution to average available wind power is related to:

f_iv_i³

Across all wind-speed intervals, the relevant cubic term becomes:

Σ f_iv_i³

rather than simply:

(Σ f_iv_i)³

Those operations are not interchangeable.

That is the mathematical heart of the issue.

Engineers Don’t Want to Store Thousands of Wind-Speed Numbers Forever

Advertisement. Scroll to continue reading.

A full frequency table is useful, but engineers also need convenient mathematical models that describe how wind speeds are distributed.

This is where probability distributions enter wind engineering.

Two names appear repeatedly:

Weibull distribution

and:

Rayleigh distribution

Patel describes the Weibull distribution as a useful representation of wind-speed variation and presents the Rayleigh distribution as a special case that can be used when only the mean wind speed is available.

The important idea for a non-specialist is not memorizing the probability-density equation.

It is understanding what the distribution gives us.

Instead of saying:

Advertisement. Scroll to continue reading.

“The site’s wind speed is 6 m/s.”

we can describe the probability that the wind will occur at different speeds.

That is much closer to the information a turbine actually experiences.

The Turbine Makes the Average-Speed Problem Even More Interesting

So far, we have discussed power available in the wind.

A real wind turbine adds another layer.

Turbines have power curves.

At very low wind speeds, the machine may produce no useful electricity.

Once wind reaches the cut-in speed, generation begins.

As wind speed rises, electrical output increases.

Advertisement. Scroll to continue reading.

Pages: 1 2 3 4 5 6