Weibull Distribution – The Best Way to Calculate Wind Speed


Weibull Distribution – The Best Way to Calculate Wind Speed

There is no more important variable in wind power than the wind itself. The energy contained in the wind increases exponentially as wind speeds increase, and in the wind industry there is no substitute for velocity.
 
The best location for wind power would be somewhere that the wind always blew at the same velocity, from the same direction, at a high enough speed for turbine blades to produce their optimum power as determined by their aerodynamics, but never so high as to require a turbine to turn itself out of the wind or shut down due to over-speed conditions.  Steady wind at a good speed is better than gusty wind, but gusty wind is better than no wind of course.
 
With this knowledge, the average person may go out and look for the average wind speed of a given location, but there’s a problem with using average… the numbers can be very deceiving.
 
Let’s take a look at a couple of examples and you can see the way an average wind speed number can be so misleading.
 
Example site 1:
 
Average wind speed number: 8 m/s
 
Actual wind speed recordings (daily wind speeds):
 
50 days: 1.5m/s

50 days: 2.5m/s

100 days: 6m/s

50 days: 9m/s

50 days: 12m/s

50 days: 19m/s
 
Doing normal averaging math you will find an annual average speed of 8m/s – that sounds like a great number. When you look at the actual values however, 150 days the turbine is out of production due to speeds below the minimum 3m/s startup, or above the 18m/s cutout. The other 200 days power production looks like this:
 
100 days @ 6m/s = 2400 hours x 18(63) = 9331 kWh

50 days @ 9m/s = 1200 hours x 18(93) = 15,746 kWh

50 days @ 12m/s = 1200 hours x 20,000W (turbine at rated max) = 24,000 kWh
 
The annual output of this turbine is 49,000 kWh

 

Now we will look at Site #2, where the average wind speed is again 8m/s.
 
70 days @ 0m/s

100 days @ 6m/s

100 days @ 10m/s

50 days @ 12m/s

30 days @ 20m/s
 
Now at this site, only 100 days the turbine will be offline due to no wind or high wind. The other 250 days power production looks like this:
 
100 days @ 6m/s = 2400 x 18(63) = 9331 kWh

100 days @ 10m/s = 2400 x18(103) = 43,200 kWh

50 days @ 12m/s = 1200 x 20,000W (turbine at rated max) = 24,000 kWh
 
The annual output of this turbine is 76,531 kWh
 
So both sites, with an average wind speed rating of 8m/s, have completely different wind profiles and a turbine located at one location could make $8000 worth of electricity, while the other location that would seem equally as suited, only made around $5000 worth of electricity.
 
Since an average wind speed number doesn’t seem to be as helpful as we would hope for, what better answer is there? The solution is called a Weibull Distribution chart of wind speed. What the Weibull Distribution model gives us is a graphical representation of how often the wind blows at a certain speed. By looking at the bars on the Weibull Distribution, someone investigating a potential wind site can get a much better idea of how well a turbine will perform at that location.
 
By applying the formula above giving each increase in wind speed its proper exponential function, the value of each bar in the graph becomes quite evident.  A Weibull Distribution that is heavily weighted with the majority of the readings falling in the wind speeds where the wind turbine can produce its rated output shows a potential goldmine for wind energy, while a Weibull Distribution that shows most of its values falling where the turbine doesn’t perform as well, could indicate a location where turbine performance would fall well below what you may expect based on the other numbers.
 

WEIBULL WIND SPEED DISTRIBUTION GRAPH

WEIBULL WIND SPEED DISTRIBUTION GRAPH (10,000 hours – 416 days)

 
 

Posted By Sally on July 3, 2013 | 0 Comment

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>