For power plants where the energy that could be generated is driven by exogenously variable factors, such as wind farms, PV power plants, and run-of-river hydro, which are all driven by weather, how accurate are the forecasts that are commissioned by the developer at the design phase, before construction starts?

What does the error distribution of the forecasts look like?

  • PV estimates should be fairly accurate using a tool such as PVWatts. Run-of-river should be pretty straightforward with a few measurements and some historical rainfall data. I would imagine wind is the wild-card... What scale are you thinking?
    – LShaver
    Oct 10, 2016 at 15:39
  • 1
    For commercial scale stuff I suspect this may be tricky to find data on - since both the pre-development studies and the eventual performance data are likely to be confidential (though for big enough developments that connect directly to transmission, data may be available on their output)
    – Flyto
    Oct 11, 2016 at 6:56

1 Answer 1


Usually the estimates are pretty good. Constructing plants is very expensive, so companies choose good consultants, and because good consultants have already estimated energy for a lot of plants, they can estimate better generation (and other local problems).

Additionally, for wind turbines, a measurement station is set-up for several months, which measure the wind at the expected high, in order to validate the models. For photovoltaic and hydro, usually good (and less expensive) data is already available.

In any case, it is better to have several sites in several places, in order to reduce risks.

Note: companies are interested on yearly average. Estimate on short term is not really taken into account.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.