Yaw error
Nacelle LiDARUp to 60% of turbines run misaligned (industry estimate)
The nacelle points degrees off the true wind — and holds it for years. cos³·⁵ of that angle comes straight off the output.
Service 04 — The recovered margin
From diagnosis to measured gain.
Nacelle LiDAR diagnostics converted into yaw correction and wake steering — the gain verified by re-measurement, never assumed. Finding and recovering lost energy across operating wind fleets.
Campaign verdict · 6.3 MW fleet · re-measured
+3.36%
Verified
AEP gain on the fleet’s best machine, re-measured after correction — the worst still gained +1.66%.
€145,180Total gain / yr
€24,197Per turbine / yr
1.6 moPayback
As found · static misalignment−5.6° to −9.0°
After correction · re-measured+1.66% to +3.36%
The fleet · wakes written in the mist
6Failure modes, one instrument
Up to 60%Turbines run misaligned · industry estimate
Up to 90%Sit in disturbed flow · industry estimate
Re-measuredGain verified, never modelled
Wake steering
Steer the wake, free the row behind.
A turbine’s wake can cost the machine behind it double-digit output. A small, deliberate yaw offset on the front row pushes that wake aside — trading a sliver on the leaders for a lot on the followers. Flip the switch, drag to orbit — and click any machine for its measured record.
WAKE STEERING · LIVE 3D · DRAG TO ORBIT
STEERING OFF · BACK ROW IN WAKE
Front row0° yaw · full output
Back rowIn wake · losing hard
Whole farmBaseline
MachineHover a turbine · click to pin its record
On a real 6.3 MW fleet we corrected −5.6° to −9.0° of static misalignment and converted the diagnosis into exactly this strategy — the gain re-measured at +1.66% to +3.36% AEP, payback inside two months.
The cost, live
What is a crooked nacelle costing you?
Power falls with cos³·⁵ of yaw error — the exponent our campaigns fit — so a handful of quiet degrees is real money, every year. Set your fleet and see. Then remember: on a campaign, we don’t model this number — we measure it.
Static yaw misalignment in degrees
Fleet capacity in megawatts
Energy tariff in euros per megawatt hour
AEP lost—
Energy—
Left on the table, every year€0
Illustrative — cos³·⁵ model as fitted on campaign data, capacity factor assumed 40%. On the campaign above, −5.6° to −9.0° of misalignment came back as +1.66% to +3.36% AEP once corrected — re-measured, not modelled.
The diagnostics
Six ways a turbine quietly underperforms.
Yaw is only the loudest one. The same instrumented campaign reads five more failure modes — each one measured on the operating machine, each one a number before it becomes a work order.
Turbulence & wake
Nacelle LiDARUp to 90% of turbines sit in disturbed flow (industry estimate)
Wake-churned inflow, read as turbulence intensity per direction — the measurement the wake-steering strategy is built from.
Pitch error
AccelerometersUp to 25% of turbines pitch-misaligned (industry estimate)
Blade-angle differences between the three blades bend the power curve and load the drivetrain on every rotation.
True north
MagnetometerHeading vs true north, per machine
A heading error shifts every directional curtailment sector — noise, shadow and load sectors all fire in the wrong wind.
Mass imbalance
AccelerometersBlade-to-blade weight difference
Grams off on one blade become vibration at the hub — fatigue in the bearings and gearbox long before it shows in AEP.
Tower frequency
StructuralTower & foundation condition
The structure’s natural frequency is a fingerprint — when it drifts, the tower or the foundation is telling you something.
One campaign · one instrument set on the operating machine · six verdicts, each a number · industry estimates above; your fleet is measured, not assumed
The method
Diagnose the inflow, don’t guess it.
A yaw controller can drift several degrees off true over time, quietly shaving AEP. Nacelle LiDAR looks upwind and measures the real inflow, so the misalignment is a number, not a hunch — turbine by turbine.
01
Diagnose
Nacelle LiDAR looks upwind and measures the true inflow ahead of the rotor — static yaw misalignment, veer and turbulence, per turbine.
- Nacelle LiDAR · WindEagle (Epsiline)
- Per-turbine yaw & inflow, measured upwind
- Misalignment as a number, not a hunch
02
Correct
Static yaw offsets and a wake-steering strategy, applied across the fleet — the diagnosis converted into an operating change.
- Static yaw offsets, per turbine
- Wake-steering strategy across the fleet
- Signed off by the OEM before it goes live
03
Verify
The AEP gain is re-measured after the change — verified, never modelled. The number you act on is the number the fleet actually produced.
- Gain re-measured after correction
- Verified, never modelled
- Per-turbine figures on request
On the curve
Where the recovered energy actually lives.
A yaw correction doesn’t move the rated plateau — it lifts the knee. The gain lives between cut-in and rated, exactly where a fleet spends most of its operating hours. The shaded band is the +1.66% to +3.36% per machine that the corrected fleet was re-measured at.
As found−5.6° to −9.0° of yaw — the knee sags, hours are lost
CorrectedThe knee lifts · +1.66% to +3.36% AEP, per machine
Above ratedThe curves converge — correction costs nothing at full power
This is why yaw is the cheapest energy a fleet can buy: no new hardware on the rotor, no loads penalty at rated — just the machine finally facing the wind it was sold to face.
The instrument programme behind it (Epsiline)
One corrected fleet is a result. Three hundred instrumented farms are a method.
300+Epsiline fleet: wind farms on WindEagle
16Epsiline fleet: regions · Europe · America · Asia
~2%Epsiline fleet: average AEP improvement
0Gains assumed, not measured
Where is the site — and what must the data prove?
Tell us where the site is and what decision the data has to support. We’ll come back with a measurement scope, a programme and a price.