WindPro vs WAsP Wake Effect Difference

[Mert]

Hello all,
We have one Windpro account which utilizes WAsP to calculate the energy production.
Recently we have realized even tough we reach the same Gross Energy Production with using both Windpro and WAsP seperately , we are getting different wake loss values when doing the exact same wind energy assessment.
All the information is exactly same for both softwares(height contours, roughness map, wind data, power curve, same Wake Model).

I am suspicious regarding the wake models. In WindPro we have (N.O. Jensen (RISO/EMD) Park 2 2018 model, in WAsP it says PARK2. Do these have differences?

Thanks and kind regards,
Mert

[Morten]

Hi Mert,

The implementation of the wake models in WAsP and WindPro are independent, so it is quite possible that the programmers made different choices. 

I made a third implementation in WAT and compared it with WAsP. The agreement was reasonable in flat terrain, but the differences were larger than expected in complex terrain. This must be because I handled speedup and deflection between the wind at the wake-shedding and wake-exposed turbine sites differently. 

Other reasons for model deviations could be

• the number of calculated wind directions within each wind sector
• the interpolation in the power and thrust-coefficient curves when looking up data for a speed-up corrected wind speed - especially near the cut-in wind speed
• possible air-density correction of the thrust-coefficient curve in WindPro?  - we decided to ignore this in WAsP and WAT

So you see, even though the PARK 2 model is really simple, there are undocumented details, which programmers might handle differently. 

Kind regards,
Morten

Edited May 13 by Morten

[Morten]

Mert sent us some WindPro and WAsP project files demonstrating these wake loss difference. Thanks for this, it is always easier to discuss possible problems when we have test data!

My former DTU colleague Tobias Ahsbahs, now working for the WindPro developers at EMD, spotted that the Ct curve used in WAsP project was much lower that the Ct curve in the WindPro project. We all agreed that this must be the reason for the different wake loss predictions. 

 

 

Edited May 17 by Morten