Greetings sir,
Congratulations on your interesting project. What is the location, if you don't mind saying?
The following is a description of the general procedures available to you to do a preliminary, and then rigorous, annual energy estimate for an anticipated facility of the size you anticipate.
===
THUMBNAIL initial feasibility estimate of local wind resource
====
You can make a thumbnail estimate of annual energy by taking an annual average windspeed number and generating a statistical distribution of annual hours the wind is in windspeed BINS from cut-in to cut-out of the anticipated machine. An then use a spreadsheet with the certified POWER CURVE (windspeed vs power in KW) of the anticipated machine (in tabular form). You will also need the air density multiplier of the site in question as comparative to the air density of the site where the power curve was measured (the latter should be readily available from the machine manufacturer; the multiplier of your site is dependent on elevation). In the spreadsheet you set up, simply multiply the number of estimated annual hours in each of the windspeed bins by the power output of the machine at the center windspeed of each bin; multiplied by the air density multiplier you have derived. That will provide at least a thumbnail "first look" by which you can, using utility buyback rates, determine if a rigorous wind measurement program is warranted. You say you have "30 mph winds" If that is an annual average (almost impossible unless you are on a high mountain peak), then you have one of the best wind sites in the world. The best commercial wind site in California (in Tehachapi, California) has an annual average of about 21 mph. So you are wise to be careful about "sensible estimates" and definitely do a measurement program.
In order to determine an annual average at your site, you will need to correlate your site with a local and reliable point of measurement which has published or available long-term data. If you are near an existing commercial wind power facility, you will have to purchase that data. If you are near an airport without much topography in between you and the airport, then that data is available, usually from NCDC or a similar government agency (eg NOAA). The windrose (direction of winds) from the two sites should be almost identical. It also helps to call the airport in question and see how well their anemometer is maintained. Sometimes the data will be questionable. If importance also is the difference in height between the hub-height of your anticpated machine and the height of the airport anemometer. You may have to ask a meterologist to provide an estimate of that difference. Jack Kline of Brentwood, Calif. ( a friend of mine who does such consulting) can provide such calculations, or even do the entire measurement program and calculations for you). If the site with long-term data is extremely close, then you will have some idea already of the annual average, and can use that number provisionally to plug into either a Rayleigh or Weibull distribution to obtain your number of hours annually in each windspeed bin (estimated).
In order to make up an initial preliminary correlation between your site and the site with long-term data, you will need at least a month of data during the high-wind season and the low wind season on your site. Then you need to do a linear regression analysis (using SAS or other stat program) between two concurrent data sets. From that equation you can then predict the difference between your site and the longterm site, and derive an estimated annual average using the long term data (the longer-term the better). This average can then be plugged into a Rayleigh or Weibull distrbution equation. Be forewarned, however, that for most very windy sites, these distributions will not be very accurate, especially where local factors (canyons, peaks, trade winds, etc) exist.
Multiply the thumbnail initial estimate of annual hours (8760 total hrs if I recall--but you can calculate this easily) in each of all bins from cut-in to cut-out; times the power in KW for that bin, times air density multiplier--and sum up all those for the annual energy from the machine. Subtract line losses. Use the calendar monthly estimates you derive and the local commercial buyback rates to figure your preliminary estimate of income annually. If this thumbnail and prelim. analysis indicates economical viability for commercial electric production using local utility published buyback rates extending out into the future, then you can go for the real thing--the better estimate of wind resource using a rigorous measurement program.
=
===
RIGOROUS "micrositing" calculation of annual energy resource from actual site data of at LEAST a year duration
=====
For this you will need a computer data-logger (approx $500) which collects and stores HOURLY AVERAGES. Hourly average wind direction and gust and lull measurements are nice features also. A vertical anemometer to give an indication of turbulence is indicated if the wind on the site is not laminar in flow at the hub-height of the planned machine. Alternatively, you can fly a kite in the area during several high wind periods and get an idea if more detailed turbulence measurements are needed. Turbulence can affect blade and bearing longevity, and thus the anticipated service costs of maintaining the machine over its lifetime of service.
You will need to measure the whole year, dumping the data to your computer via the memory device provided or by direct or modem connection. The sensors should be on a tower at HUB-HEIGHT of the planned wind turbine, on the chosen anticipated SITE if at all possible. If the hub height is 60 ft, then your current plans are sufficient. NRG systems of Vermont makes tubular guyed anemometer towers up to about 100-120 feet tall. Otherwise you can use a truss-type communications tower of some sort. If hub-height measurement is not possible, then you will need to use projection methods to estimate the windspeed at the higher height using the data collected at the lower height.
The monthly hourly average data of speed and direction should be compiled in a database. Gust and lull information is also of interest.
With each monthly data set, the ACTUAL power from an anticipated machine can be calculated using the hourly average, the power in KW from the power curve of the machine, and the density multiplier. This yields an infinitely higher resolution picture of daily, monthly, and annual projected energy production from a machine with a good power curve (and the Danish, or other commercial machines of 100KW will have very well measured and certified power curves). Once a year of data has been collected, then a decent picture of annual variation will be collected, (although unless you have a nearby site with longterm data, even a year of data can be skewed by interannual variations). With this actual onsite data, you can, after the first few collected months of data, make the following two calculations:
--use the hourly averages and power curve data to determine ACTUAL energy production during that measured period
--use the actual data collected on site and the actual concurrent data collected at the nearby long-term site (corrected for height difference by calculations); to create a more accurate correlation between the long term data and your site. IF the correlation coefficient is high, then you can study inter-annual variation historically at your site, to determine the low and high bounds of likely deficit and "extra" as compared to your actual year of collected data. This information will allow your more certain financial calculations to be done for inclusion in your actual business plan. If the correlation coefficient is poor, then you will have to use the long-term data only very carefully in attempting to determine inter-annual variation.
Be forewarned that if you have material partners in the project, then it is wise to have a professional meteorologist do the measurements and calculations, to prevent legal harangues if the site has some bad wind years with revenue shortfalls that no one wants to see. If the site is really an obvious "screamer", then with the information given here, study of other commercial windfarm methods, and perhaps some piecemeal help from a meteorologist, you should be able to do the calculations yourself if you are inclined toward such calculations. But you will have to examine production estimates based on your measurement for as long a period as is practical (at the least one year, unless there is nearby highly correlated reliable long-term data) --against buyback rates; state, federal, and local tax or rebate benefits; and careful estimates of project and maintenance costs in order to really know if the project will ever be lucrative.
I suggest a low-rpm, stall regulated turbine such as a Danish design to avoid any opposition on the grounds of raptor mortality. One of the best manufacturers is Vestas, who I think has a US subsidiary and probably a US manufacturing plant. Most other Danish manufacturers are comparable, but measure them against Vestas in terms of reliability.
I hope this has been helpful. Good luck with your project.
Cypress
