Loading...
HomeMy WebLinkAboutEgegik Wind Resource Assessment Report - Feb 2017 - REF Grant 7060912Egegik, Alaska Wind Resource Assessment Report Egegik met tower, photo by Douglas Vaught February 23, 2017 Douglas Vaught, P.E. V3 Energy, LLC www.v3energy.com Egegik, Alaska Wind Resource Assessment Report Page | 1 Summary The wind resource measured at the Egegik met tower site is outstanding with a mean annual wind speed of 7.43 m/s and a wind power density of 516 W/m 2 at 34 meters above ground level. In all respects the Egegik wind resource is highly suitable for wind power development. This wind resource assessment report was prepared by V3 Energy, LLC under contract to Lake and Peninsula Borough. Met tower data synopsis Data dates 8/20/2014 to 9/03/2016 (24.5 months) Wind speed mean, 34 m, annual 7.43 m/s (16.6 mph) Wind power density mean, 34 m 516 W/m 2 Wind power class Class 5 (excellent) Max. 10-min wind speed 32.6 m/s Maximum 2-sec. wind gust 41.1 m/s (91.9 mph), December 2015 Weibull distribution parameters k = 1.92, c = 8.34 m/s Wind shear power law exponent 0.216 (moderate shear) Surface roughness 0.28 meters (agricultural land with tall hedgerows) IEC 61400-1, 3rd ed. classification Class III-C Turbulence intensity, mean (at 34 m) 0.104 (at 15 m/s) Calm wind frequency (at 34 m) 19% (< 4 m/s) Test Site Location and Selection Process A 34 meter NRG Systems, Inc. tubular-type meteorological (met) tower was installed in Egegik in an open area of Becharof Corporation land on a hill approximately 2,200 ft. due east of the Egegik city office. Egegik is located on the south bank of the Egegik River on the Alaska Peninsula, 100 miles southeast of Dillingham and 326 air miles southwest of Anchorage. Egegik is incorporated as a 2nd class city in the Lake and Peninsula Borough with 85 permanent residents, but its population increases significantly during early and mid-summer when the local salmon processing plant is operational. Egegik falls within the southwest climate zone, characterized by persistently overcast skies, high winds, and frequent cyclonic storms.1 The test site was chosen after a decision was made not to lower and re-use an existing NRG Systems 30 meter, 4-inch diameter met tower located immediately across the street (due east) of the city office. This met tower had been in place for an unknown number of years and was in acceptable condition for re-use on site, but could not be moved because corrosion of the galvanized coating caused the slip-fit tower sections to fuse. It is not known who installed this tower, when, or the subsequent fate of any data collected from it. Following abandonment of the old met tower location, focus was directed to the highest point of land in the community, excluding the location of the GCI cellular communications tower 2,300 ft. south of the subsequently-chosen met tower site. The met tower site is on Becharof Corporation land and their 1 Community data obtained from State of Alaska DCCED Community and Regional Affairs website Egegik, Alaska Wind Resource Assessment Report Page | 2 approval was obtained for temporary installation of a met tower for the two-year life of the wind study. The site is a commanding location with excellent exposure and visibility in all directions. Heavy brush exists north and east of the site, which presents some concern for wind turbine operations. After approval of the chosen site by Alaska Energy Authority (AEA), they forwarded a suggestion to instead consider a site at the intersection of the airport road and the GCI tower access road to better locate the met tower with existing power infrastructure. This site was inspected during a pre- installation visit and was deemed undesirable for met tower installation due to heavy brush and an undesirable slope for a met tower. Additionally, landowner permission and an FAA obstruction determination had already been obtained for the chosen location. But, given the strong wind resource measured in Egegik, this alternate site could be suitable wind turbines. Southwest Alaska and Egegik, Google Maps image Met Tower Information The met tower was installed in late August 2014 with highly appreciated labor and material assistance from City of Egegik. Met tower details Site number 5500 Latitude/longitude N 58° 12’ 31.50”, W 157° 21’ 49.00” Time offset -9 hours from UTC (Yukon/Alaska time zone) Site elevation 45 meters (148 ft.) Datalogger type NRG SymphoniePLUS3, 10 minute time step Tower type Guyed tubular, 15 cm (6 in.) diameter, 34 meter (112 ft.) height Egegik Anchorage Kodiak IslandPribilof Islands Egegik, Alaska Wind Resource Assessment Report Page | 3 Tower sensor information Channel Sensor type Designation SN 2 Height Multiplier Offset Orientation 1 NRG #40C anem. 34 m A 222521 33.8 m 0.744 0.40 320 T 3 2 NRG #40C anem. 34 m B 222522 34.0 m 0.758 0.34 185 T 3 NRG #40C anem. 20 m 222523 20.4 m 0.752 0.38 320 T 7 NRG #200P vane Direction 33.1 m 0.351 000 000 T 9 NRG #110S Temp C Temp 2.5 m 0.136 -86.38 030 T Met tower installation; use of bulldozer to lift the tower Tower sensor photographs North side, up tower East side, up tower South side, up tower West side, up tower 2 Anemometer serial number 3 Degrees true, or relative to Earth’s geographic North Pole Egegik, Alaska Wind Resource Assessment Report Page | 4 Met tower site photographs Site view to north Site view to northeast Site view to east Site view to southeast Site view to south Site view to southwest Egegik, Alaska Wind Resource Assessment Report Page | 5 Site view to west Site view to northwest Egegik, view north, Google Earth image Met tower Alternate site Egegik, Alaska Wind Resource Assessment Report Page | 6 Egegik, view south-southeast, Google Earth image Topographic map of Egegik Data Quality Control Data was filtered to remove presumed icing events that yield false zero wind speed data and non-variant wind direction data. Data that met criteria listed below were automatically filtered. In addition, data was manually filtered for obvious icing that the automatic filter didn’t identify, and invalid or low quality data for situations such as logger initialization and other situations. Met tower Met tower Egegik, Alaska Wind Resource Assessment Report Page | 7 x Anemometer icing – data filtered if temperature < 1°C, speed SD = 0, and speed changes < 0.25 m/s for minimum 2 hours x Vane icing – data filtered if temperature < 1°C and vane SD = 0 for minimum of 2 hours x Tower shading of 34 meter A and B paired anemometers – data filtered when winds from ± 15° of behind tower; refer to graphic below In general, icing conditions were infrequent indicating minimal concern for wind turbine energy production loss due to ice. With frequent southeasterly winds, tower shadow affected anemometer 34 m A (channel 1, oriented to 320° T) much more often than anemometer 34 m B (channel 2, oriented to 185° T), hence the significantly lower recovery rate of anemometer 34 m A. Tower shading plot Sensor data recovery table Data Column Possible Records Valid Records Recovery Rate Icing Invalid Tower shading Speed 34 m A 107,322 83,265 77.6% 1,927 197 21,793 Speed 34 m B 107,322 97,663 91.0% 1,632 198 7,880 Speed 20 m 107,322 105,077 97.9% 1,906 195 0 Direction 34 m 107,322 103,168 96.1% 3,804 206 0 Temperature 107,322 106,983 99.7% 0 195 0 Sensor data recovery rate by month (includes tower shading for 34 m A & B) Year Month 34 m A 34 m B 20 m Vane Temp 2014 Aug 82.5 83.6 88.7 88.1 88.7 2014 Sep 86.3 87.3 100.0 100.0 100.0 2014 Oct 91.1 75.1 100.0 100.0 100.0 Egegik, Alaska Wind Resource Assessment Report Page | 8 Year Month 34 m A 34 m B 20 m Vane Temp 2014 Nov 74.7 92.5 99.7 86.5 100.0 2014 Dec 75.1 96.8 95.2 86.5 100.0 2015 Jan 72.4 72.5 78.7 69.9 100.0 2015 Feb 76.5 87.7 96.4 94.9 100.0 2015 Mar 75.7 84.9 92.6 92.3 100.0 2015 Apr 80.6 87.7 100.0 100.0 100.0 2015 May 38.7 96.7 100.0 100.0 100.0 2015 Jun 80.7 96.7 100.0 100.0 100.0 2015 Jul 76.0 98.1 100.0 100.0 100.0 2015 Aug 89.1 89.7 100.0 100.0 100.0 2015 Sep 89.3 86.5 100.0 100.0 100.0 2015 Oct 83.9 88.5 100.0 100.0 100.0 2015 Nov 85.0 94.3 100.0 96.8 100.0 2015 Dec 77.6 91.7 98.5 93.4 100.0 2016 Jan 77.8 92.8 96.8 96.8 96.8 2016 Feb 80.9 95.7 97.3 97.4 100.0 2016 Mar 84.3 85.9 98.4 96.1 100.0 2016 Apr 61.2 97.5 100.0 100.0 100.0 2016 May 67.0 98.0 100.0 100.0 100.0 2016 Jun 83.3 97.5 100.0 100.0 100.0 2016 Jul 88.0 98.2 100.0 100.0 100.0 2016 Aug 68.4 94.3 100.0 100.0 100.0 2016 Sep 43.6 98.8 100.0 100.0 100.0 All Data 77.6 91.0 97.9 96.1 99.7 Data Synthesis Filtering removes compromised sensor readings from the data set. This is desirable for icing in that it eliminates the negative speed bias of false “zero” data. Filtering for tower shadow is more nuanced in that filtered results bias both paired anemometers, but it’s not obvious in which direction for either. One solution is to remove filtered data and fill the missing gaps with synthesized data using a gap-filling subroutine 4 contained in Windographer Pro software. Gap-filling, or data synthesis, yields more a more representative and realistic data set. This is especially true for tower shadow-filtered data in that the flagged data from one paired anemometer can be reconstructed with data from the other anemometer of the pair. The result is a true representation of wind speeds from both paired anemometers. Gap- filling icing-flagged data is more complex in that often all anemometers and/or wind vanes freeze simultaneously and hence Windographer software must use the Markov transition to create a probable result for the flagged period. For short icing periods, the inherent uncertainty of this approach is low; for long periods, it is higher. 4 First-order Markov transition matrix; described in Windographer Help Egegik, Alaska Wind Resource Assessment Report Page | 9 Wind Speed Anemometer data obtained from the met tower, from the perspectives of both mean wind speed and mean wind power density, indicate a moderate wind resource. Note that cold temperatures contributed to a higher wind power density than standard conditions would yield for the measured mean wind speeds. This is reflected in the CRMC (cubed root mean cubed) wind speed, which reflects a calculation of a steady wind speed, at the measured mean air density, that would yield the measured mean wind power density. In other words, given the cool climate in Egegik, the winds punch above their weight. A table following that below presents the same data but with anemometer icing and tower shadow data removed from the data set and then synthesized with Windographer software’s gap-filling subroutine. The advantage of gap-filling is that a more representative data set is achieved, especially with inclusion of data from the opposing anemometer (with paired anemometers) when data is filtered for tower shadow (gap-filling synthesizes tower shadow data by referencing the paired anemometer where data is not flagged). Anemometer data summary (gap-filled) Variable Speed 34 m A Speed 34 m B Speed 20 m Measurement height (m) 33.8 34.0 20.4 Mean wind speed (m/s) 7.36 7.41 6.60 MoMM 5 wind speed (m/s) 7.38 7.43 6.61 Max 10 min avg. wind speed (m/s) 32.6 32.6 30.0 Max gust wind speed (m/s) 41.1 38.7 40.8 CRMC wind speed (m/s) 9.24 9.32 8.27 Weibull k 1.94 1.92 1.97 Weibull c (m/s) 8.29 8.34 7.44 Mean power density (W/m²) 498 511 357 MoMM power density (W/m²) 503 516 361 Mean energy content (kWh/m²/yr) 4,359 4,475 3,129 MoMM energy content (kWh/m²/yr) 4,403 4,519 3,159 Energy pattern factor 1.98 1.99 1.97 Frequency of calms (<4 m/s) (%) 19.8 19.9 24.3 Time Series Time series calculations indicate higher wind speeds during the winter months compared to summer month, but this difference is not highly pronounced. The daily wind profile (annual basis) indicates higher wind speeds during the day with peak winds occurring between 3 and 4:00 p.m. 5 MoMM: mean of monthly means, or annual Egegik, Alaska Wind Resource Assessment Report Page | 10 34 m B anemometer data summary Raw Mean Filtered Mean Gap- filled Mean Max 10- min Avg Max Gust Std. Dev. Weibull k Weibull c Year Month (m/s)(m/s)(m/s)(m/s)(m/s)(m/s)(-)(m/s) 2014 Aug 6.68 6.79 6.71 14.4 19.0 2.72 2.62 7.53 2014 Sep 7.40 7.52 7.51 22.4 28.0 4.00 1.92 8.44 2014 Oct 6.14 6.26 6.38 15.9 19.9 3.02 2.19 7.17 2014 Nov 7.04 7.24 7.08 22.0 28.0 3.89 1.87 7.96 2014 Dec 7.77 7.94 7.88 28.0 34.6 4.78 1.74 8.88 2015 Jan 6.34 7.82 7.89 18.7 25.7 3.26 2.55 8.87 2015 Feb 7.87 7.98 8.15 23.9 31.8 3.81 2.26 9.20 2015 Mar 7.04 7.69 7.61 19.2 24.6 3.49 2.31 8.58 2015 Apr 7.65 7.87 7.74 20.2 25.7 3.90 2.09 8.74 2015 May 8.98 9.17 8.99 23.4 29.2 4.92 1.87 10.10 2015 Jun 5.79 5.87 5.81 19.6 24.6 3.17 1.90 6.54 2015 Jul 6.89 6.99 6.90 22.6 28.0 3.93 1.81 7.75 2015 Aug 7.13 7.34 7.19 16.7 20.7 3.25 2.35 8.12 2015 Sep 7.49 7.59 7.58 21.7 28.0 3.90 2.02 8.54 2015 Oct 6.19 6.16 6.29 16.0 19.9 3.22 1.91 7.00 2015 Nov 8.60 8.64 8.62 21.9 28.0 3.85 2.38 9.72 2015 Dec 8.48 8.89 8.62 32.6 41.1 5.57 1.58 9.60 2016 Jan 9.15 9.29 9.17 26.7 34.6 4.28 2.25 10.34 2016 Feb 8.41 8.64 8.58 20.4 26.8 3.79 2.40 9.66 2016 Mar 6.43 6.42 6.53 19.5 23.6 4.22 1.47 7.16 2016 Apr 8.02 8.13 8.04 22.9 28.0 4.16 2.02 9.07 2016 May 6.93 7.00 6.94 18.9 24.6 3.63 2.00 7.83 2016 Jun 5.43 5.49 5.44 20.1 25.7 3.49 1.62 6.08 2016 Jul 7.24 7.27 7.25 16.8 19.9 3.35 2.30 8.17 2016 Aug 6.31 6.41 6.34 15.9 19.9 2.98 2.24 7.15 2016 Sep 5.05 5.09 5.05 10.6 14.2 2.35 2.31 5.71 All Data 7.26 7.47 7.41 32.6 41.1 3.98 1.92 8.34 Annual 7.27 7.48 7.43 Egegik, Alaska Wind Resource Assessment Report Page | 11 Monthly time series (annual), mean wind speeds (gap-filled data set) Daily wind profile (gap-filled data set) Long-term Wind Speed Average Comparing the 24 months of measured wind speed data at the Egegik met tower is possible by reference to the nearby Egegik Airport automated weather station. Data for this station was obtained for the time period of June 1993 through Dec. 31, 2016. For this 27.5 year time period, the AWOS station recorded an average wind speed of 5.09 m/s (at a 10 meter measurement height). In 2015, which comprises the only full calendar year of the Egegik met tower operating time period, the AWOS station wind speed average was 4.89 m/s, which is 4% less than the long-term average. Note also a very slight declining trend in wind speed over the 27.5 year period, although this data trend is mostly driven by high wind speed variability from 1993 to 2000. Since 2000, winds at the Egegik Airport have been fairly constant. Egegik, Alaska Wind Resource Assessment Report Page | 12 Egegik Airport wind speed Probability Distribution Function The probability distribution function (PDF), or histogram, of the Egegik met tower site wind speed indicates a shape curve dominated by moderate wind speeds and is mostly reflective of a “normal” shape curve, known as the Rayleigh distribution (Weibull k = 2.0), which is defined as the standard wind distribution for wind power analysis. As seen below in the wind speed distribution of the 34 meter B anemometer, the most frequently-occurring wind speeds are between 4 and 8 m/s with few wind events exceeding 20 m/s, the cutout speed of most wind turbines. Also, note the accompanying cumulation distribution with respect to the infrequency of very high wind speeds in Egegik. PDF of 34 m B anemometer (gap-filled) 0 1 2 3 4 5 6 7 1993199419961998199920002001200220032004200520062007200820092010201120122013201420152016Wind Speed, m/sEgegik Airport Weather Station Egegik, Alaska Wind Resource Assessment Report Page | 13 Weibull k shape curve table Weibull values table, 34m B anemometer Weibull Proportion Power Weibull c Mean Above Density R Algorithm k (m/s)(m/s)7.41 m/s (W/m2)Squared 6 Maximum likelihood 1.92 8.34 7.39 0.451 493.1 0.981 Least squares 1.87 8.42 7.47 0.455 524.1 0.976 WAsP 1.83 8.19 7.28 0.435 495.5 0.971 Actual data 7.41 0.435 495.5 Cumulative distribution Wind Shear and Roughness Wind shear at the Egegik met tower site was calculated with the 34 m A and 20 m anemometers, both of which were oriented toward 320° T. The calculated power law exponent of 0.216 indicates a moderate 6 Relatedness or correlation of Weibull approximation algorithm with actual data Egegik, Alaska Wind Resource Assessment Report Page | 14 wind shear at the site. Calculated surface roughness at the site is 0.29 m (the height above ground where wind speed would be zero) for a roughness class of 2.88 (description: agricultural land). Note the high shear with southeasterly winds, undoubtedly due to upwind brush. Mitigation should be considered if wind turbines are located at or near this site, such as removal of the brush. Vertical wind shear profile Wind shear by direction graph Egegik, Alaska Wind Resource Assessment Report Page | 15 Extreme Winds International Electrotechnical Commission (IEC) 61400-1, 3rd edition extreme wind probability classification is one criteria – with turbulence the other – that describes a site with respect to suitability for wind turbine models. Extreme wind is described by the 50 year Vref, or reference velocity in a 50 year return period; in other words, Vref is the wind speed (10-minute average) predicted to occur once every 50 years. IEC 61400-1 extreme wind classification IEC 61400-1, 3rd ed. Class Vref, m/s I 50 II 42.5 III 37.5 S designer- specified Periodic Maxima One method to estimate Vref is a Gumbel distribution analysis modified for monthly maximum winds versus annual maximum winds, which are typically used for this calculation. Thirty-four months of wind data in the 50 meter met tower data set are acceptable for this analysis, although, as noted previously, the latter months of that data set were compromised by sensor failure and all sensors were compromised by icing during the winter months. For this analysis, the 34 meter level B anemometer is referenced because it recorded the highest wind speeds of the three anemometers on the tower. With filtered, gap-filled, and preconditioned 7 (by the Weibull k value as an exponent) data, the predicted Vref by this method is 36.6 m/s. This result meets IEC 3rd edition Class III criteria, the lowest-defined category of extreme wind probability. Note, however, the presumed substantial loss of higher speed winter anemometer data due to icing. Given the comparison of measured and filtered mean wind speed to the AWS Truepower model, it is possible that Vref may be higher than calculated. 7 Preconditioning improves the accuracy of 50-year extreme wind speed estimates; Windographer Help (references 1996 (Harris) and 2009 (Langreder et al.) studies). Egegik, Alaska Wind Resource Assessment Report Page | 16 Periodic maxima cumulative distribution, 34 m B anemometer A second technique, Method of Independent Storms, yields a V ref estimate of 41.2 m/s, higher than that predicted by the periodic maxima method but still within the classification constraint to classify as IEC 61400-1 Class II extreme wind. Method of Independent Storms Egegik, Alaska Wind Resource Assessment Report Page | 17 A third method, known as EWTS II (European Wind Turbine Standards II), ignores recorded peak wind speeds and calculates Vref from the Weibull k factor. There are three variations of this method – Exact, Gumbel and Davenport – which yields a Vref between 32.9 and 36.2 m/s for Egegik. These are in-line with the periodic maxima method and within IEC 3rd edition Class III extreme wind criteria. EWTS II plot Summary The calculated Vref wind speeds by the three methods described above all meet IEC 61400-1, 3rd edition criteria for Class III wind classification, which has a Vref limit of 42.5 m/s. The practical importance is that turbines suitable for Egegik should be IEC 61400-1 Class II certified, or possibly Class III certified if referencing only the periodic maxima and EWTS methodology. EWTS II table Vref (50 yr) Method (m/s) Periodic Maxima 36.6 Method of Independent Storms 41.3 EWTS II (Exact) 32.9 EWTS II (Gumbel) 33.4 EWTS II (Davenport) 36.2 Temperature, Density, and Relative Humidity Egegik experiences cool summers and relatively mild winters, by interior Alaska standards, with resulting higher than standard air density. Calculated mean-of-monthly-mean (or annual) air density during the met tower test period exceeds the 1.219 kg/m 3 standard air density for a 45 meter elevation by 3.6 percent. This is advantageous in wind power operations as wind turbines produce more power at low temperatures (high air density) than at standard temperature and density. Egegik, Alaska Wind Resource Assessment Report Page | 18 Temperature and density table Mean Min Max Mean Min Max Mean Min Max Month (°C)(°C)(°C)(°F)(°F)(°F)(kg/m3)(kg/m3)(kg/m3) Jan -3.3 -22.1 6.1 26.1 -7.8 43.0 1.300 1.216 1.399 Feb -1.4 -20.1 7.8 29.5 -4.2 46.0 1.292 1.248 1.388 Mar -1.0 -20.1 10.6 30.2 -4.2 51.1 1.290 1.235 1.388 Apr 3.2 -7.7 13.3 37.8 18.1 55.9 1.270 1.223 1.323 May 7.4 0.1 21.1 45.3 32.2 70.0 1.250 1.188 1.284 Jun 11.0 3.5 26.4 51.8 38.3 79.5 1.233 1.165 1.268 Jul 12.8 7.2 23.7 55.0 45.0 74.7 1.225 1.177 1.251 Aug 12.8 5.5 22.6 55.0 41.9 72.7 1.225 1.182 1.259 Sep 9.3 -0.3 19.5 48.7 31.5 67.1 1.241 1.195 1.286 Oct 3.0 -8.4 13.1 37.4 16.9 55.6 1.271 1.224 1.326 Nov 0.5 -16.6 11.6 32.9 2.1 52.9 1.283 1.231 1.369 Dec -2.0 -16.3 7.6 28.4 2.7 45.7 1.295 1.249 1.367 Annual 4.4 -22.1 26.4 39.8 -7.8 79.5 1.264 1.165 1.399 Egegik temperature boxplot graph Wind Speed Scatterplot The wind speed versus temperature scatterplot below indicates relatively cool temperatures at the Egegik met tower site but on average temperatures are above freezing, as indicated in the preceding temperate and density table. During the met tower test period, temperatures were not often colder than -20° C (-4° F), the minimum operating temperature for most standard-environment wind turbines. Egegik, Alaska Wind Resource Assessment Report Page | 19 With this, wind turbines with an arctic option, designed for -40° C operations, may not be necessary in Egegik. Wind speed/temperature (color code indicates wind direction) Wind Direction Wind frequency rose data indicates that winds at the Egegik met tower site are primarily southeasterly, with northeasterly and westerly winds frequent as well. The magnitude of the wind sector “pie” slices indicate that the strongest winds are very strongly southeasterly. Note that the measured wind rose at the met tower site mostly correlates with that winds observed by the automated weather station at the nearby Egegik Airport. The primary difference is that the airport recorded more northerly winds that was measured at the met tower. Egegik, Alaska Wind Resource Assessment Report Page | 20 Wind frequency and energy rose Egegik Airport wind rose Egegik, Alaska Wind Resource Assessment Report Page | 21 Turbulence The turbulence intensity (TI) at the Egegik met tower site is low with a mean turbulence intensity of 0.075 and a representative turbulence intensity of 0.104 at 15 m/s wind speed at the 34 meter level, indicating smooth air for wind turbine operations. This equates to an International Electrotechnical Commission (IEC) 61400-1, 3 rd Edition (2005) turbulence category C, which is the lowest defined category. Turbulence Intensity table Turbulence intensity, all direction sectors Wind Speed Sensor Mean TI SD of TI Represent ative TI Peak TI Mean TI SD of TI Represent ative TI IEC 3 ed. Turb. Cat. Speed 34 m B 0.130 0.090 0.250 1.46 0.104 0.018 0.127 C Speed 34 m A 0.130 0.080 0.240 1.74 0.104 0.023 0.133 C Speed 20 m 0.150 0.080 0.250 1.27 0.129 0.018 0.152 B All Speed Bins 15 m/s Speed Bin