Prospects of Optimization of Energy Production by LiDAR Assisted Control of Wind Turbines

1
Prospects of Optimization of Energy Production by
LiDAR Assisted Control of Wind Turbines

• D. Schlipf1, S. Kapp1, J. Anger1, O. Bischoff1,
  M. Hofsäß1, A. Rettenmeier1, U. Smolka1 and M.
  Kühn2
• 1Endowed Chair of Wind Energy, Universität
  Stuttgart
• 2AG Wind Energy Systems, Universität Oldenburg

2
Motivation

• nacelle-mounted LiDAR systems provide preview
  information of incoming wind
• previous work shows promising load reduction

• Estimation of energy yield gained by
• LiDAR assisted speed control (based on
  simulations)
• LiDAR assisted yaw control (based on
  measurements)

SWE
3
Wind Reconstruction of 3D Wind

• LiDAR measures the velocity of aerosols by
  backscattered light.
• Cyclops- Dilemma
• 3 LiDAR from independent directions needed for
  real 3D vector
• missing LiDAR substituted by assumptions
• no vertical horizontal wind component
• no vertical component homogenous flow
• possible solution
• collective / cyclic pitch / speed control (high
  frequency)
• yaw control (low frequency)

 
 
SWE
4
LiDAR Assisted Speed Control
 
 
 
NREL 5 MW
SWE
5
LiDAR Assisted Speed Control
Indirect Speed Control (ISC)
 
 
Direct Speed Control with Feedforward (DSC)
 
other feedforward based controllers possible
(higher order error dynamics)
SWE
6
LiDAR Assisted Speed ControlTheoretical
Considerations
 
 
SWE
7
LiDAR Assisted Speed ControlConnection
Simulation ? Measurements
 
 
Filter of simulated data
Real inflow measurements on a 5MW turbine with
SWE nacelle-based scanning LiDAR
SWE
8
LiDAR Assisted Speed ControlDemonstration
Simulation with NREL 5MW and realistic LiDAR
simulator
AEP GWh
ISC 0.18 458.7 2.6
DSC 0.05 459.1 2.9
DSC/ISC -74 0.085 8.9
 

• Not a good idea!

SWE
9
LiDAR Assisted Yaw Control

• Yaw Control normally by
• nacelle sonic/wind vane.
• disturbed by blades
• only point measurement

• LiDAR based yaw control
• undisturbed inflow
• measurement over rotor area

SWE
10
LiDAR Assisted Yaw Control
 
 
SWE
11
LiDAR Assisted Yaw ControlSimulated Measurements
 
 
 
 
SWE
12
LiDAR Assisted Yaw ControlSimulated Measurements

• Full simulations
• absolute error lt1 for 10 min
• depending on turbulence
• In simulations LiDAR can measure the 10 min rotor
  averaged wind direction!
• But we have no model for
• anemometer disturbance
• inhomogeneous inflow
• Consider real data!

SWE
13
LiDAR Assisted Yaw Control

• 5 month of inflow measurement were analyzed,
• filtered for
• trajectory
• turbine status
• LiDAR quality

• Analysis
• 10 min LiDAR wind direction assumed as perfect
• compared to sonic
• same control strategy is assumed for LiDAR and
  sonic turbine yaws if 10 min average gt 10

SWE
14
LiDAR Assisted Yaw Control
 

• Maximal 2 more energy output!
• With standard control maximal 1!

SWE
15
Conclusions

• Wind Reconstruction
• with one nacelle based LiDAR no 3D wind
  measurement possible based on line-of-sight
• estimation using assumptions / fit to model

•  

SWE
16
Outlook

• including other spectral information of LiDAR
  measurements or combination with blade root
  bending moment data
• validation of wind characteristics from LiDAR
  with those estimated from turbines
• improve yaw control strategy
• LiDAR measurements atalpha ventus in RAVE -
  OWEA
• development of robust LiDAR and test in RAVE -
  LIDAR II

Oelker

• Focus on Look-ahead Collective Pitch Control
• Feedforward and Gust Detection Carcangiu et al.
  EWEA 2011
• Nonlinear Model Predictive Control Schlipf et
  al. AWEA 2011

16
17
Thanks for you attention!

• AcknowledgementThe RAVE project Development of
  LiDAR technologies for the German offshore test
  field is funded by the German Federal Ministry
  for the Environment, Nature Conservation and
  Nuclear Safety (BMU). Thanks to AREVA Wind GmbH.