View More Content

Modelling P50 versus P90 wind profiles

Modelling P50 versus P90 wind profiles

Modelling P50 and P90 wind profiles is generally poorly done in the models I am sent. This is maybe one reason we are producing so many financial models for wind projects at the moment!

Depending on your budget and purpose there are quite a lot of resources on the internet for understanding wind economics. The template model from Windustry has some interesting features for a quick evaluation, but I would not recommend this for any form of investment decision.

Banker-friendly wind project financial model

A wind model needs the functionality to switch between different wind profiles, say P90 (debt) and P50 (equity). I thought I would share with you how we do it as it seems to work well.

What does P50 and P90 mean in wind financial models?

Without turning this into a statistics forum, we can broadly say that P50 and P90 represent two probabilistic comparisons of the expected wind available to drive the turbines, compared to historic records. P90 broadly represents the probability that the wind (fuel) experienced by the project will on average be in the 90% confidence range.

How do banks use P90 and P99 profiles?

P90 is often used for the ‘bank base case’. A P99 profile is even more conservative, and often used as the ‘bank downside case’. Because of the different risk profiles of different parties involved in the same transaction, it is a useful feature for the financial model in a wind project to have the ability to switch between a P50 to a P90 with ease. I recommend setting the profile up as a scenario field – it’s important to make your bankers life as easy as possible!

How are wind probability profiles modelled?

It is easier said than done, but let’s take a look at how it could be modelled.

To model a P90 versus P50 case, a NORMINV function is used. For a P50 scenario, the formula would be:

= NORMINV (1 - 50/100, net production, uncertainty * net production)

Similarly for P90, the 50 would just be replaced by 90. Therefore, it’s easy to see that scenarios can easily be run by just controlling one parameter in the above formula.

Advanced tip for modelling multiple wind turbines  

This computation can slow Excel down, so rather than calculating this for each turbine, the formula can be used on the aggregated amount. So, instead of having, say, 40 lots of NORMINV functions for 40 turbines, you would only have just one NORMINV line, which speeds up the model tremendously! If you find yourself in this situation and not sure how to solve it, then let me know!

Rickard Wärnelid
by Rickard Wärnelid

Contact Rickard Wärnelid

view all