HomeMy WebLinkAboutMarshall Wind Project Wind Power Report - Aug 2010 - REF Grant 7040021 V3 Energy, LLC 1
Marshall Wind Power Report
Report by: Douglas Vaught, P.E., V3 Energy, LLC, Eagle River, Alaska
Date of Report: August 11, 2010 (revision 1)
Contents
Summary Information ...................................................................................................................................2
Wind Turbine Performance ..........................................................................................................................2
Marshall Turbine Performance Estimate......................................................................................................3
Wind Farm .....................................................................................................................................................3
Village Load ...............................................................................................................................................3
Marshall Wind Farm Performance............................................................................................................4
Marshall Wind Resource...............................................................................................................................7
Data Quality Control .................................................................................................................................9
Wind Speed Data Summary......................................................................................................................9
Extreme Winds........................................................................................................................................ 10
Wind Roses.............................................................................................................................................. 11
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 2
Wind Shear .............................................................................................................................................. 12
Probability Distribution Function ............................................................................................................ 13
Air Temperature and Density .................................................................................................................. 13
Speed-Temperature Scatterplot ......................................................................................................... 14
Turbulence.............................................................................................................................................. 15
Summary Information
AVEC has proposed construction of wind turbines near the village of Marshall to augment the diesel ge-
nerators and create a wind-diesel power system. Although abbreviated by a failed anchor that led to
collapse of the met tower, the Marshall wind study indicates a Class 3 to 4 wind resource with accepta-
ble turbulence characteristics and extreme wind probability for wind turbine operation.
Marshall Meteorological Tower Data Synopsis
Data start date December 18, 2008
Data end date October 12, 2009 (9.8 months data)
Wind power class Class 3 (fair) to Class 4 (good)
Wind speed average (30 meters) 5.97 m/s measured, estimate 6.3 m/s annual
Maximum 10-min average wind speed 30.8 m/s
Maximum wind gust 37.8 m/s (January 2009)
IEC 61400-1 3rd ed. extreme winds Class II (note: 10 months data)
Wind power density (30 meters) 343 W/m2
Weibull distribution parameters k = 1.62, c = 6.72 m/s
Roughness Class 0.90 (fallow field)
Power law exponent 0.138 (moderate wind shear)
Frequency of calms (3.5 m/s threshold)29%
Mean Turbulence Intensity 0.095 (IEC 61400-1 3rd ed. turbulence category C)
Wind Turbine Performance
It is perhaps counterintuitive that wind power density and wind class do not correlate linearly with tur-
bine power output. This is due to a number of factors, including theoretical limitations of a lift-
producing aerodynamic device (the turbine rotor) and practical limitations of generator weight and
rated output. For these reasons and others, a wind turbine in a low power class wind regime may still
produce sufficient energy to warrant installation of turbines.
A simplistic consideration of possible turbine output in Marshall is to model power output of a particular
turbine with mean of monthly means data collected and extrapolating to the turbine hub height. Note
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 3
the this analysis is based on raw data with no synthetic data inserted in place of icing data removed for
data quality control or data missing for other reasons. Turbine performance was analyzed with the
HOMER software using the Northern Power Northwind 100 B model (100 kW rated output, 21 meter
rotor diameter).
Marshall Turbine Performance Estimate
100% availability 90% availability 80% availability
Hub
height
Wind
speed
Capacity
factor Production
Capacity
factor Production
Capacity
factor Production
Turbine (m) (m/s) (%) MWh/yr (%) MWh/yr (%) MWh/yr
NW100/21 37 6.42 28.3 254.9 25.5 229.4 22.6 203.9
Wind Farm
Alaska Village Electric Cooperative (AVEC) has proposed installation of two Northern Power Northwind
100 wind turbines (21 meter rotor B model) to create a wind-diesel hybrid power system for the village
of Marshall. HOMER software was used to create simulation model for Marshall.
Village Load
A Marshall hourly load profile was synthesized using the Alaska village load calculator Excel spreadsheet
developed by the Alaska Energy Authority several years ago. The results were adjusted slightly to match
the actual village average and peak loads of Marshall documented by AVEC in their 2008 annual power
generation report. The result is a virtual Marshall village with a 144 kW average load, 232 peak load and
average daily power usage of 3,466 kWh/day. Seasonal, daily and DMap profiles of the Marshall virtual
electric load are shown below.
Seasonal profile
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann
0
50
100
150
200
250 Seasonal Profile
max
daily high
mean
daily low
min
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 4
Daily profile and DMap
Alaska village calculator load details
Marshall Wind Farm Performance
HOMER software was used to estimate wind turbine production, wind penetration and fuel displace-
ment for wind power input from operation of one and two NW100/21 turbines in the Marshall wind re-
gime measured by the met tower. HOMER modeling may in some cases yield lower production esti-
mates than Alaska Energy Authority (AEA) methods which consider only performance of the turbine dis-
placing fuel usage from diesel generators supplying essentially an infinite load. HOMER considers the
dynamics of the actual village load and the reality that turbines will on occasion generate more power
than can be absorbed by the village electric load. During these times, excess energy must be diverted to
a secondary use, such as a thermal heat load, or turbine operation must be curtailed.
0 6 12 18 24
0
50
100
150
200 Daily Profile
Hour
0
50
100
150
200
250
300
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecMWh/monthOther
Communications
Health Clinic
City/Government
Public Water System
Commercial
School
Residential
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 5
Wind turbine production estimates, 100% turbine availability
No. of
Turbines
System
Wind Pene-
tration
Wind Pro-
duction
Genset
Fuel Dis-
placed
System
Excess
Energy
System
Excess
Energy
Heating
fuel
equiv.
Turbine (%) (MWh/yr) (gal) (MWh/yr) (%) (gal)
NW100/21 1 20.2 254.9
15,614 10.2 0.8 408
2 40.3 509.8
23,976 133.6 9.6 5,344
Notes:
1. Marshall wind resource adjusted to annual
2. Turbine hub height 37 meters
3. HOMER modeling assumes 100% turbine availability
4. Displaced fuel for electrical generation only
5. Excess electricity to secondary thermal heating load
6. Heating fuel equivalent for system excess energy dump; 25 kWh/gal equivalent
7. HOMER model operating reserves: 10% current load, 50% wind power output
8. Diesel generator efficiency data from 2008 AVEC annual generation report
9. Minimum 50 kW diesel generator loading; all excess energy to secondary load
Wind turbine production estimates, 80% turbine availability
No. of
Turbines
System Av-
erage Pe-
netration
Wind Pro-
duction
Genset
Fuel Dis-
placed
System
Excess
Energy
System
Excess
Energy
Heating
fuel
equiv.
Turbine (%) (MWh/yr) (gal) (MWh/yr) (%) (gal)
NW100/21 1 16.2 203.9 12,491 8.2 0.6 326
2 32.2 407.8 19,181 106.9 7.7 4,275
Notes:
1. Marshall wind resource adjusted to annual
2. Turbine hub height 37 meters
3. Turbine availability adjusted to 80%
4. Displaced fuel for electrical generation only
5. Excess electricity to secondary thermal heating load
6. Heating fuel equivalent for system excess energy dump; 25 kWh/gal equivalent
7. HOMER model operating reserves: 10% current load, 50% wind power output
8. Diesel generator efficiency data from 2008 AVEC annual generation report
9. Minimum 50 kW diesel generator loading; all excess energy to secondary load
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 6
HOMER modeled system interaction 1
HOMER modeled system interaction 2
Jul 26 Jul 27
0
50
100
150
200
AC Primary Load
Northwind100/21RevB
Generator 3 Power
Excess Electricity
Jul 28 Jul 29
0
50
100
150
200
AC Primary Load
Northwind100/21RevB
Generator 3 Power
Excess Electricity
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 7
Marshall Wind Resource
A met tower was installed at the proposed wind turbine site in Marshall on December 18, 2008 and was
in continuous operation until October 10, 2009 when an anchor failed during an exceptionally strong
wind storm and the tower collapsed. The tower was not replaced as it is felt that sufficient data was
collected during the ten month data measurement period to adequately characterize the site. With the
data on hand, an average wind speed of 6.0 m/s was measured, with a wind power density of 332 W/m2
(Class 4 wind resource). Because the two missing months are mid-October to mid-December, typically
the windiest period of time of the year, the actual annual wind speed average and wind power density
may well be higher than reported here.
Other aspects of the wind resource also are promising for wind power development. By IEC 61400-1 3rd
edition classification, Marshall is category II-c or III-c, indicating low turbulence (mean TI at 15 m/s =
0.095) and moderate to low 50 year extreme winds. The latter measure is more difficult to quantify
with only ten months of data, but the site clearly is not energetic enough to be IEC extreme wind Class I.
The NW100/21 is designed for IEC II-B sites, so the Marshall site is well within the design parameters of
the turbine. Icing has also not proven to be a significant issue in the met tower data.
General Site Information
Site name/number Marshall/datalogger site 0050
Site Description East of village on prominent ridge along road to airport
Latitude/longitude N 61° 52’ 32”, W 162° 3’ 58” (WGS 84)
Site elevation 63 meters (200 ft)
Datalogger/modem type NRG Symphonie/no modem
Tower type NRG 30-meter tall tower, 152 mm (6 in) diameter
Tower Sensor Information
Channel Sensor type Height Multiplier Offset Orientation
1 NRG #40C anemometer 30 m (A) 0.765 0.35 270°, west
2 NRG #40C anemometer 30 m (B)0.765 0.35 180°, south
3 NRG #40C anemometer 22 m 0.765 0.35 270°, west
7 NRG #200P wind vane 29 m 0.351 023 270°, west
8 NRG #200P wind vane 20 m 0.351 0 180°, south
9 NRG #110S Temp C 2 m 0.136 -86.383 north
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 8
Test Site Location
Topographic Map
Google Earth Map
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 9
Data Quality Control
Data was filtered manually to remove obvious icing data. Typically, anemometer icing is identified by
non-variant data readings at the sensor offset values, a standard deviation of zero, and temperature
near or below freezing. Wind vane icing is identified by non-variant readings, a standard deviation of
zero, and almost always occurs in conjunction with anemometer icing. The wind data from Marshall
indicates surprisingly few icing events, especially compared to the heavier icing measured down river on
the met towers installed in Pitka’s Point and Saint Mary’s.
Sensor Units Height
Possible
Records
Valid
Records
Recovery
Rate (%)
Speed 30 m A m/s 30 m 42,990 41,855 97.4
Speed 30 m B m/s 30 m 42,990 41,830 97.3
Speed 22 m m/s 22 m 42,990 41,962 97.6
Direction 29 m ° 29 m 42,990 41,017 95.4
Direction 20 m ° 20 m 42,990 40,937 95.2
Temperature °C 42,990 42,752 99.5
Wind Speed Data Summary
Measured wind speeds in Marshall are rather high for an inland site and are promising for wind power
development.
Variable
Speed 30 m
A
Speed 30 m
B
Speed 22
m
Measurement height (m) 30.0 30.0 22.0
Mean wind speed (m/s)5.98 6.02 5.75
MMM wind speed (m/s) 6.16 6.21 5.91
Median wind speed (m/s) 5.40 5.40 5.20
Min wind speed (m/s) 0.4 0.4 0.4
Max 10 min avg wind speed (m/s) 28.8 30.8 26.6
Max gust wind speed (m/s) 35.2 37.8 34.8
Weibull k 1.63 1.62 1.64
Weibull c (m/s) 6.68 6.72 6.42
Mean power density (W/m²) 332 343 294
MMM power density (W/m²) 352 366 309
Mean energy content (kWh/m²/yr) 2,908 3,003 2,572
MMM energy content (kWh/m²/yr) 3,081 3,203 2,708
Energy pattern factor 2.40 2.44 2.39
Frequency of calms (%) 29.1 28.7 31.0
1-hr autocorrelation coefficient 0.898 0.902 0.898
Diurnal pattern strength 0.075 0.078 0.086
Hour of peak wind speed 17 17 16
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 10
Monthly time series
Wind speed daily profiles
Extreme Winds
The calculation of extreme wind probability generally requires more than ten months of data to be con-
sidered acceptable, but nonetheless, extreme wind predictions are shown below. Note that the pre-
dicted 50 year ten-minute average wind speed of 38.2 m/s classifies the site as IEC Class II (International
Electrotechnical Commission 61400-1, 3rd edition), although with more data a calculation of Class III
probability is likely (based on data from other similar Alaska sites with Class 4 winds).
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
2
4
6
8
10 Seasonal Wind Speed Profile
Speed 30 m A
Speed 30 m B
Speed 22 m
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 11
Marshall RETURN YR 10 min average, m/s 3 sec gust, m/s
30 meter B anem. 2 27.5 34.5
10 32.9 41.3
15 34.2 43.0
30 36.5 45.9
50 38.2 48.1
100 40.5 51.0
Wind Roses
Winds at the Marshall met tower test site are primarily east-northeast, north-northwest with occasional
winds from south-southeast (wind frequency rose), with the strongest winds east-northeast (mean value
rose). The power density rose indicates that the power producing winds at the site are very predomi-
nately east-northeast. Should multiple wind turbines be sited, they should be oriented approximately
north-northeast to south-southwest to provide good exposure to ENE and SSE winds and avoid tower
shadowing.
Note that a wind threshold of 3.5 m/s was selected for the definition of calm winds. This wind speed
represents the cut-in wind speed of the Northern Power NW100/21 wind turbine. Other wind turbines
in this size range have cut-in wind speeds of about 4 m/s. With a 3.5 m/s wind speed threshold, the
Marshall met tower site experienced 29 percent calm conditions during the test period.
Wind Frequency Rose (29 m vane) Mean Value Rose
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 12
Total value (power density) rose Scatterplot (30 m A power density)
Wind Shear
The power law exponent was calculated at 0.138 for all wind directions, indicating moderate wind shear
at the Marshall met tower test site. The practical application of this data is that a lower turbine tower
height may be possible at this location should wind power development occur, but of course greater
energy power generation will occur at higher hub heights.
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 13
Probability Distribution Function
The probability distribution function (PDF) provides a visual indication of measured wind speeds in one
meter per second or smaller “bins”. Note that most wind turbines do not begin to generate power until
the wind speed at hub height reaches 3.5 to 4 m/s, known as the “cut-in” wind speed and most are de-
signed to stop generating power at 25 m/s. The black line in the graph is a best fit Weibull distribution:
k=1.62, c=6.72 m/s. The PDF of wind data collected in Marshall indicates a relatively normal wind distri-
bution or Weibull curve for wind power sites with very minimal high wind events.
Air Temperature and Density
During the measurement period, Marshall had an average temperature of 0.7° C. The minimum record-
ed temperature during the measurement period was -31.1° C (February, 2009) and the maximum tem-
perature was 28.2° C (July, 2009).
Because of Marshall’s cold winter temperatures, the average air density of 1.281 kg/m3 is approximately
five percent higher than the standard air density of 1.218 kg/m
3 (14.6° C standard temperature and
100.5 kPa pressure) at 63 m elevation, indicating that Marshall has denser than standard air, resulting in
improved turbine performance as compared to a turbine in a similar wind regime in standard atmos-
pheric conditions.
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 14
Air Temperature Boxplot
Speed-Temperature Scatterplot
A scatterplot of wind speed versus temperature is of interest in order to assess possible cold tempera-
ture turbine restrictions. Note that temperatures colder than -30°C are apparently quite uncommon in
Marshall and when they do occur, winds are minimal. A wind turbine rated to -40°C would be ideal; one
rated to -30°C acceptable, and one rated to -20°C not acceptable.
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 15
Turbulence
Turbulence intensity (TI) at the Marshall met tower site during the measurement period was acceptable
with a mean TI at 15 m/s of 0.095 and a representative TI at 15 m/s of 0.131 (30 m A sensor, all sectors).
This classifies the site as turbulence category “C” by International Electrotechnical Commission (IEC)
61400-1 3rd edition (2005) criteria (note that Category C is the lowest classification of turbulence for
wind power development, indicating low turbulence at the site). In the TI wind rose, one can see low
turbulence for the predominate wind ENE wind direction and higher turbulence from the less common
NNE and S, SW and W wind directions.
Turbulence Intensity, all wind sectors
Turbulence Intensity, ENE wind sector
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 16
Turbulence Intensity Rose
Turbulence Table
Bin Bin Endpoints Records Standard
Representative Midpoint Lower Upper In Mean Deviation Peak
(m/s) (m/s) (m/s) Bin TI of TI TI TI
1 0.5 1.5 204 0.451 0.157 0.652 0.929
2 1.5 2.5 393 0.239 0.137 0.415 0.800
3 2.5 3.5 531 0.158 0.097 0.282 0.742
4 3.5 4.5 559 0.126 0.075 0.222 0.615
5 4.5 5.5 538 0.105 0.056 0.177 0.462
6 5.5 6.5 569 0.097 0.044 0.153 0.345
7 6.5 7.5 510 0.096 0.039 0.145 0.338
8 7.5 8.5 564 0.084 0.035 0.129 0.250
9 8.5 9.5 521 0.082 0.033 0.124 0.221
10 9.5 10.5 455 0.085 0.031 0.125 0.290
11 10.5 11.5 378 0.084 0.031 0.124 0.289
12 11.5 12.5 356 0.082 0.028 0.117 0.210
13 12.5 13.5 258 0.084 0.029 0.121 0.209
14 13.5 14.5 199 0.081 0.024 0.112 0.161
15 14.5 15.5 171 0.083 0.022 0.111 0.160
Marshall Wind Power Report, rev. 1
V3 Energy, LLC 17
16 15.5 16.5 89 0.079 0.021 0.106 0.162
17 16.5 17.5 70 0.075 0.016 0.096 0.124
18 17.5 18.5 45 0.071 0.017 0.093 0.110
19 18.5 19.5 24 0.068 0.017 0.089 0.101
20 19.5 20.5 10 0.082 0.020 0.107 0.122
21 20.5 21.5 11 0.061 0.010 0.073 0.073
22 21.5 22.5 2 0.067 0.017 0.089 0.079