Results of AWE simulations
Archived runs of the 150 m reel-out optimizer scenario from
SimpleKiteControllers.jl
(examples/simple_opt_reelout.jl), one subfolder per wind speed: vNN is a mean
wind speed at 6m height of NN m/s, not a version counter. Each folder holds the latest run at that
wind speed. SimpleKiteControllers.jl/output/scenarios is a symlink into this folder,
so the two paths refer to the same data.
| folder | wind speed | mean reel-out power | peak | run date |
|---|---|---|---|---|
v03 |
3 m/s | 463 W | 1.2 kW | 2026-08-21 |
v04 |
4 m/s | 1.6 kW | 2.5 kW | 2026-08-21 |
v05 |
5 m/s | 3.9 kW | 4.6 kW | 2026-08-21 |
v06 |
6 m/s | 7.3 kW | 8.2 kW | 2026-08-20 |
v07 |
7 m/s | 12.0 kW | 13.2 kW | 2026-08-21 |
v08 |
8 m/s | 17.7 kW | 21.4 kW | 2026-08-21 |
v09 |
9 m/s | 26.0 kW | 31.5 kW | 2026-08-20 |
v10 |
10 m/s | 35.0 kW | 53.8 kW | 2026-08-20 |
Each folder is a self-contained record of one run:
reelout_150m_opt.arrow— the flight log (the simulation trace).reelout_150m_opt.yaml— a run summary written at the end of the run: git hash/dirty status, wind/turbulence settings, wall-clock time,fig8_metrics(cross-track error, elevation, tether force, turn rate, lap count, pattern extent vs. commanded amplitude, etc.) and measured vs. predicted power — read this first instead of loading the Arrow file.fc_settings_reelout.yaml,settings_reelout_150m.yaml,system_reelout_150m.yaml,traj_opt.yaml,wc_settings.yaml,gui.yaml— copies of every settings file that fed the run, so each archive is reproducible independent of whateverdata/in SimpleKiteControllers.jl currently holds.
The runs come from different commits of SimpleKiteControllers.jl (git_hash in the
summary), so a comparison across wind speeds is not necessarily a comparison at one
fixed tuning — check the settings copies before reading a trend into the table.
SimpleKiteControllers.jl's output/ is overwritten by every new run, so these folders
are the durable record — read a run's numbers from its archive here, not from output/.