Thanks Jérôme for sharing your views and insights. I'm working engineering tasks related to design (loads on support structures) but also revenue (yield assessments) sides of projects, and I wonder:
With a nominal 10-year standard error (uncertainty) of 6% for a wind farm with a best estimate (p50) net capacity factor of 46%, the corresponding 90%-exceedance threshold (p90) is 42.5%. How do these uncertainty estimate compare with actual revenue uncertainty? My own analysis (using publicly-available data) shows that most wind farm do achieve much larger long-term net yield than their original p90 estimates.
I guess it is worth making a difference between production- and revenue uncertainty, as there are multiple risk-mitigation mechanisms (guaranteed availability, pay-back mechanism when grid is down etc). What do you think, is the revenue risk still at high as it used to be? How could this risk be reduced further?
Hi Rémi, thanks for your comments. From my experience, wind assessments have been generally quite correct; the reason actual revenues have been higher is that other factors than wind have been conservatively assessed, such as the various loss factors (in particular for transmission), and losses end up being less than feared, allowing net production to be higher. These items are slowly absorbed into more bullish assumptions by the industry, and revenue risk is indeed seen as lower than in the past.
Thanks, it's great to hear this. State-of-the art yield assessments still include hefty uncertainties related to the measurements data and the analyses results, see for instance the DNV-GL study for the NY Bight here: https://oswbuoysny.resourcepanorama.dnvgl.com/ "New York Bight Preliminary Energy Assessment", and the its uncertainties summary table https://bit.ly/3L3rPSX. Maybe too hefty sometimes, if you ask me ^^.
I wonder whether these could be reduced further, and more emphasis be put on the "wind-to-power" (power performance) and general operability issues (like grid loss revenue compensations).
If one day you fancy a discussion on the topic of yield assessment bankability nowadays, with a wide audience of Wind & Site specialists, maybe would the WRAG https://groups.io/g/wrag online group be the right forum. People there would be happy to hear your thoughts and give you feedback.
I'm easy to reach if you want. But note that lenders and most financial investors are happy to see that actual production numbers are consistent (or slightly above) estimates and don't need more than that. Good resource assessment is critical in only a few highly competitive tenders (which NY Bight will not be, given that only a handful of projects will be able to compete, for instance)
Thanks Jérôme for sharing your views and insights. I'm working engineering tasks related to design (loads on support structures) but also revenue (yield assessments) sides of projects, and I wonder:
With a nominal 10-year standard error (uncertainty) of 6% for a wind farm with a best estimate (p50) net capacity factor of 46%, the corresponding 90%-exceedance threshold (p90) is 42.5%. How do these uncertainty estimate compare with actual revenue uncertainty? My own analysis (using publicly-available data) shows that most wind farm do achieve much larger long-term net yield than their original p90 estimates.
I guess it is worth making a difference between production- and revenue uncertainty, as there are multiple risk-mitigation mechanisms (guaranteed availability, pay-back mechanism when grid is down etc). What do you think, is the revenue risk still at high as it used to be? How could this risk be reduced further?
All the best. Rémi (C2Wind, DK).
Hi Rémi, thanks for your comments. From my experience, wind assessments have been generally quite correct; the reason actual revenues have been higher is that other factors than wind have been conservatively assessed, such as the various loss factors (in particular for transmission), and losses end up being less than feared, allowing net production to be higher. These items are slowly absorbed into more bullish assumptions by the industry, and revenue risk is indeed seen as lower than in the past.
Thanks, it's great to hear this. State-of-the art yield assessments still include hefty uncertainties related to the measurements data and the analyses results, see for instance the DNV-GL study for the NY Bight here: https://oswbuoysny.resourcepanorama.dnvgl.com/ "New York Bight Preliminary Energy Assessment", and the its uncertainties summary table https://bit.ly/3L3rPSX. Maybe too hefty sometimes, if you ask me ^^.
I wonder whether these could be reduced further, and more emphasis be put on the "wind-to-power" (power performance) and general operability issues (like grid loss revenue compensations).
If one day you fancy a discussion on the topic of yield assessment bankability nowadays, with a wide audience of Wind & Site specialists, maybe would the WRAG https://groups.io/g/wrag online group be the right forum. People there would be happy to hear your thoughts and give you feedback.
Bonne continuation :)
I'm easy to reach if you want. But note that lenders and most financial investors are happy to see that actual production numbers are consistent (or slightly above) estimates and don't need more than that. Good resource assessment is critical in only a few highly competitive tenders (which NY Bight will not be, given that only a handful of projects will be able to compete, for instance)