IEEE: About Wind Turbines

Mike O'Dell mo at
Sat Feb 6 12:00:28 CST 2010

generators don't have KWh meters on them the same
way your house does. there is certainly instrumentation
from which one can (and does) get that kind of data,
but that's not the point.

firstly power production agreements are bilateral
contracts between two parties. there are legal reporting
requirements to the FERC but there are almost certainly
limitations on the detail the FERC can disclose about
a private contract between two parties.

most importantly, though, precision beyond a certain point
is pretty much useless. as you say, just because there are
digits on the calculator doesn't mean they are meaningful.

the real question is "why is it important to know precisely
how much was delivered?"  knowing precisely how much load
was carried last year in the aggregate is not nearly as
important as understanding on a per-farm basis the ratio
between theoretical capacity and actual delivered capacity,
and that is monitored with keen interest. the range of
ratios, however, was a useful item to include in the
article because it shows the significant variation between
the realized performance of various wind farms. 

it is critically important for people, especially 
non-technically-minded policy wonks, to understand
one simply cannot take a number and assume it's some
kind of reliable constant.

equally important is understanding that even if there
was an effective capacity of 3.3GW last year, the
weather next year might yeild more, and it might yeild
less. all of these alternative sources are still
operating in the realm of small numbers. that's why
it was so easy to increase wind capacity by 40% in
one year. that's because it was so small!

there is one wind field in development in Canada that
will single-handedly double the wind generation in
the province of Ontario. the prime contractor on that
project is Samsung, by the way. 

that's the perspective issue: the current numbers are
a tiny fraction of total capacity. that will probably
change over time, but a lot of things must happen
for that to change significantly, over and above
"just plant more windmills."

it's not commonly known that one of the largest
expenses in constructing the wind farm is the cost
of renting the cranes which can erect the turbines.
CAPEX for the turbines is first, but the construction
costs are usually second, and of that number, 
crane rental is a huge fraction.

consider the scale of the problem:

a 1.5-2 MW turbine has a rotor that's approximately
100 meters in diameter. yes, 100 meters. so the
pylon must be some significant fraction of 100 meters
assuming you'd like to keep the blade tips well
above the ground. a typical nacelle with all the
stuff inside weighs between 75 and 100 tons without
the rotor. once the pylon is assembled, mounting the
nacelle requires a direct vertical lift of 100 meters
of that 100 ton nacelle. cranes which can perform 
this feat rent for about $100,000 per day, including
transit, erection, and teardown. the crane arrives
on 5 semi trailers, takes two days to erect, a
day or two to do its work, and another two days
do disassemble. without road time, that's half
a million dollars just for the crane. 

this is why farms are assembled en masse,
and why maintenance of turbines is done in
groups. if a turbine goes down, you lose its
output until a few more fail to make it economical
to bring in the crane and repair them. this assumes
you have parts, etc, to do the work.

so when contemplating effective capacity of a wind farm
for planning purposes, a utility is forced to
use estimates, and those estimates vary widely
based on the track record of the wind farm operator,
the turbines chosen, and a number of other factors.
and the wind farm operator derates the output as well
so he can meet the contractual requirements of
the PPA because there are performance penalties.

as to knowing the exact numbers, you can probably 
get as good a number as you can get starting 
from the FERC and following the strings from there.


More information about the Tacos mailing list