# How to estimate wind generation using power curve?

I am trying to verify windfarmer generation estimates manually. To do so I am trying to manually calculate the estimated energy production (kWh) of a wind turbine. The information I have available is the

1. Hourly weather data (wind speed, direction & air density at Hub Height, and temperature)
2. Power Curve of wind Turbine
3. Thrust Coefficient of wind turbine
4. Temperature de-ration curve of wind turbine
5. Turbine specifications (blade radius, swept area, cut in & cut out speeds, Cp, Rated Power, etc)

How would I use this data to estimate the potential annual generation?

I tried multiplying hourly wind speeds * the specific value in the power curve, then de-rating the result based on the given temperature, However it seems to significantly overestimate the answer.

Unfortunately I cannot share the data, but all I am looking for is a formula or calculation methodology. Any help be greatly appreciated!

• To get a meaningful result you will need 10 years of wind speed data... and even then the weather can come up with extremes... Using your 1 & 2 should give a good enough estimate for a comparison. Aug 7 '19 at 6:05
• Also, you probably want tip to toe system efficiency figured in. The number you want is grid supplied power, so you need all the conversion losses and downtime figured in. Aug 7 '19 at 9:29

For an approximate estimate, use

$$P = \frac{1}{2}\rho C_p A u^3$$

where $$\rho$$ is the air density, $$C_p$$ is the power coefficient (read off the power curve that you have, where it will depend on the wind speed), $$A$$ is the swept area of the rotor, and $$u$$ is the wind speed.

If you have hourly data, calculate this for each hour. That will give you a power in Watts (or kW, or MW), so you can then assume that that is the average output over the hour centred on the time of the weather measurements, and hence that the energy generated in that hour is the same number of Wh (or kWh, or MWh).

You could try to manually include temperature de-rating (I don't know exactly how this works, but if you have a curve you could apply it), but if your aim is to check that the Windfarmer result is sensible, this ought to get you pretty close.

• Including the temperature correction is easy if you use a spreadsheet... a vlookup(). Aug 9 '19 at 21:16
• thanks mate, perfect exactly what I was after Aug 16 '19 at 0:29
• @BobbyHeyer at the risk of sounding mercenary, how about an upvote then? :-) Aug 16 '19 at 20:14
• @flyto hey mate, I did but don't have enough reputation for it to be visible Aug 26 '19 at 3:52
• Ah, I see. Oh well ;) Aug 26 '19 at 4:08