Input files for heterogeneous inflow.

[triantafillosdazelidis]

Hello. I’m using long-term measurement sites with hourly wind speed and direction data to estimate energy production, optimize farm layout etc. The problem is that I can't find a way to input multiple datasets for heterogeneous inflow and using wind speed multipliers seems impossible in my case. What can I do? Am I missing something? Thank you in advance.

[misi9170]

Hi @triantafillosdazelidis , thanks for reaching out! Could you explain the format of your site data a little further? For instance, do you have something that looks like:

timestep  wind_speed  wind_direction
xxx.xx0   8.0               270.0
xxx.xx1   8.5               280.0
xxx.xx2   9.2               350.2

or do you have wind speed and wind directions at multiple locations within the area of interest?

Better yet, are you able to upload a subset of your data here to clarify?

[triantafillosdazelidis]

Thank you for your swift reply @misi9170. Indeed I have many datasets, each corresponding to a different location within the area of interest. The datasets look like the attached picture. Every dataset includes hourly observations for many years.

[Bartdoekemeijer]

@triantafillosdazelidis would it be possible to assign your first dataset (location 1) as the "reference" dataset (which is used to set the floris wind_speeds and wind_directions), and all the other datasets are converted into a relative speed-up compared to dataset 1/location 1? That should then allow you to use the speed multiplier functionality, no?

[triantafillosdazelidis]

I have thought of that, to be honest I'm not sure how to apply it, can I have an example?

[Bartdoekemeijer]

You have to start by dividing the wind speed timeseries of every Dataset by the wind speed timeseries of Dataset 1. That will give you the speed-up timeseries for every Dataset. Second step is to collect all those together.

So roughly, just looking at the first timestep now:

• The FLORIS inflow wind speed: wind speed of Dataset 1 for the first timestep, so 11.05004983.
• Speedup multiplier: the x and y locations are the (x, y) locations corresponding to Datasets 2 until N. The actual speed-up array would be [ws_dataset2/ws_dataset1, ws_dataset3/ws_dataset1, ws_dataset4/ws_dataset1, ws_dataset5/ws_dataset1, ... ws_datasetN/ws_dataset1].

Does that make sense?

[triantafillosdazelidis]

Thank you for your time, I will give it a try and come back here if I run into a dead end.