Can Floris be used to calculate AEP in 2D? #273
-
|
Hello all! |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 1 reply
-
|
Hi Victor, Indeed, FLORIS by default calculates the wind speed at every rotor for 25 points. We do this to ensure accuracy, since the wind speed may vary significantly across the rotor. However, for your purposes, you can reduce this to a single point to speed up your computations. You can do this by changing the which effectively yields a There are a number of additional steps you can take to speed-up your code. Firstly, you can consider the suggestion made in this discussion post if you are not interested in wake deflection. Secondly, you can also parallelize your code using Also, note that @bayc is currently working on the next iteration of FLORIS, where we specifically target speed improvements for AEP calculations. The aimed release date is some time later this month. |
Beta Was this translation helpful? Give feedback.
-
|
Hello @Bartdoekemeijer
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks so much for fast and kind reply! |
Beta Was this translation helpful? Give feedback.
Hi Victor,
Indeed, FLORIS by default calculates the wind speed at every rotor for 25 points. We do this to ensure accuracy, since the wind speed may vary significantly across the rotor. However, for your purposes, you can reduce this to a single point to speed up your computations. You can do this by changing the
ngridparameter in your .json file:which effectively yields a
1 x 1grid, thus one rotor point.There are a number of additional steps you can take to speed-up your code. Firstly, you can consider the suggestion made in this discussion post if you are not interested in wake deflection. Secondly, you can also parallelize your code using
mpi4pyormultiprocessing.Als…