DNV GL © 2014 SAFER, SMARTER, GREENER DNV GL © 2014
Redefinition of Power Curves: Electric vs. Kinetic Power
10 December 2014
1
Erik Tüxen, Vineet Parkhe, Paul Lawson
DNV GL © 2014
Starting point: What does a wind turbine do?
10 December 2014 2
• Conversion of kinetic power of wind (into mechanical power) into
electric power.
• Efficiency is output (electric power) divided by input (kinetic power).
(In other words: Output depends on input and efficiency.)
• Input and efficiency are both affected by several parameters,
additional to the 10 minute average wind speed at hub height.
• Knowledge and understanding about input and efficiency is the key to
avoid expensive surprises, e.g. regarding energy yield or with respect
to the results of a power performance verification test.
DNV GL © 2014
Shear and turbulence intensity (TI)
10 December 2014 3
• Shear causes a non-ideal angle of flow (towards blade) during the
rotation (e.g. at lower and/or upper tip height).
• IEC 61400-12-1 Ed. 1 (2005) does not consider the effect of shear on
average kinetic power of a 10-minute period.
• TI causes a non-ideal angle of flow (towards blade) after every change
of the wind speed until the pitch angle is adjusted to the new wind
speed. The rate of this adjustment is subject to physical limitations.
• IEC 61400-12-1 Ed. 1 (2005) does not consider the effect of TI on
average kinetic power of a 10-minute period.
kin
etic p
ow
er
[kW
]
DNV GL © 2014
Reconsideration of approaches: Hub Height Power Curve
10 December 2014 4
measured wind speed centre of rotor area
3
2
1vAPkin
-200
0
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
2600
0 2 4 6 8 10 12 14 16 18 20
wind speed [m/s]
pow
er
outp
ut
[kW
]
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
cp
DNV GL © 2014
Reconsideration of approaches: Presently available suggestions
10 December 2014 5
Rotor-equivalent wind speed (REWS) and TI renormalisation
DNV GL © 2014
Are we on the right track?
10 December 2014 6
• Rotor-equivalent wind speed (REWS) has gained a significant amount
of support, but a certain degree of opposition has also been observed.
This can similarly be said about the turbulence renormalisation method
which is included as Annex M (informative) in the latest CD of IEC
61400-12-1 Ed. 2.
• Both methods are indirect. (Kinetic power, which is a central element in
the theory, is still not shown in the power curve.)
Can we apply the useful ideas behind this in a simpler way?
DNV GL © 2014
Proposal for a new approach
7
• Let us try to create a new kind of power curve:
electric power vs. kinetic power (instead of applying various corrections
to wind speed and/or electric power, in order to get artificial data which
is supposedly more “representative” than the measured data).
• “High Resolution Kinetic Power” (in terms of space and time) can be
measured nowadays, due to new types of remote sensing devices and
other innovative technology. (Further improvement can be expected.)
• This new method does not mean that the principle of 10 minute
averaging will have to be discontinued.
ele
ctr
ic p
ow
er
kinetic power
10 December 2014
Bin Kinetic
Power
Kinetic
Power No. of
Kinetic
Power
Electric
Power Cp
Number From To Datasets Mean Mean Mean
[-] [kW] [kW] [-] [kW] [kW] [-]
1
2
3
4
DNV GL © 2014
Case 1: PPM with ground-based lidar
10 December 2014 8
Electric power (y-axis) vs. kinetic power (x-axis)
(data filtered on valid sector and turbine availability, then split into
subsets for TI ranges)
DNV GL © 2014
Case 2: PPM with 3D nacelle-based lidar (Slide 1 of 2)
10 December 2014 9
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
0.025
0.05
0.075
0.1
0.125
0.15
0.175
0.2
0.225
0.25
0.275
0.3
Tu
rbu
len
ce I
nte
ns
ity
Kinetic Power (kW)0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
3000
Ele
ctr
ical
Po
wer
(kW
)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
0.025
0.05
0.075
0.1
0.125
0.15
0.175
0.2
0.225
0.25
0.275
0.3
Tu
rbu
len
ce I
nte
ns
ity
Kinetic Power (kW)0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
3000
Ele
ctr
ical
Po
wer
(kW
)
0 500 1000 1500 2000 2500 3000 3500 4000 4500 50000
500
1000
1500
2000
2500
Ele
ctr
ical
Po
wer
(kW
)
Kinetic Power (kW)
TI 0.15-0.20
0 500 1000 1500 2000 2500 3000 3500 4000 4500 50000
500
1000
1500
2000
2500
Ele
ctr
ical
Po
wer
(kW
)
Kinetic Power (kW)
TI 0.05-0.10
DNV GL © 2014
Case 2: PPM with 3D nacelle-based lidar (Slide 2 of 2)
10 December 2014 10
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Tu
rbu
len
ce I
nte
ns
ity
Kinetic Power (kW)
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
x 104
0
250
500
750
1000
1250
1500
1750
2000
2250
2500
2750
3000
Ele
ctr
ical
Po
wer
(kW
)
TI
Electrical Power
0 0.2 0.4 0.6 0.8 1 1.20
0.2
0.4
0.6
0.8
1
1.2
Kinetic Power
Ele
ctr
ical
Po
wer
TI 0.00-0.05
TI 0.05-0.10
TI 0.10-0.15
TI 0.15-0.20
TI 0.20-0.25
c
c
c
c
CP
CP
DNV GL © 2014
Summary and Outlook
10 December 2014 11
• A new power curve definition is being proposed, focusing directly on
input, efficiency, and output of the conversion from kinetic power to
electric power.
• Promising starting points have been made regarding the experimental
application of the method. (Due to limited amount of data, not all of
the intended evaluations have been completed yet – further progress
can be expected very soon.)
• The capability of this method to reduce existing problems in energy
assessment and power performance verification will be demonstrated,
as soon as comprehensive amounts of data are existent.
• Certain aspects of the established method are easily adjusted (e.g. the
method of bins), while others will be challenging (e.g. site calibration),
but no insolvable issue has been identified so far.
DNV GL © 2014
SAFER, SMARTER, GREENER
www.dnvgl.com
Thank you!
10 December 2014 12
+49 4856 901 24