WIND DATA REPORT. Savoy

Transcription

1 WIND DATA REPORT Savoy December 01, 2004 November 30, 2005 Prepared for Massachusetts Technology Collaborative 75 North Drive Westborough, MA by William L.W. Henson James F. Manwell Anthony L. Rogers Anthony F. Ellis April 3, 2006 Report template version 1.3 Renewable Energy Research Laboratory 160 Governors Drive, (413)

2 NOTICE AND ACKNOWLEDGEMENTS This report was prepared by the Renewable Energy Research Laboratory (RERL) at the in the course of performing work sponsored by the Renewable Energy Trust (RET), as administered by the Massachusetts Technology Collaborative (MTC), pursuant to work order number The opinions expressed in this report do not necessarily reflect those of MTC or the Commonwealth of Massachusetts, and reference to any specific product, service, process, or method does not constitute an implied or expressed recommendation or endorsement of it. Further, MTC, the Commonwealth of Massachusetts, and RERL make no warranties or representations, expressed or implied, as to the fitness for particular purpose or merchantability of any product, apparatus, or service, or the usefulness, completeness, or accuracy of any processes, methods or other information contained, described, disclosed, or referred to in this report. MTC, the Commonwealth of Massachusetts, and the contractor make no representation that the use of any product, apparatus, process, method, or other information will not infringe privately owned rights and will assume no liability for any loss, injury, or damage directly or indirectly resulting from, or occurring in connection with, the use of information contained, described, disclosed, or referred to in this report. April 3, 2006 Renewable Energy Research Laboratory Page 1

3 TABLE OF CONTENTS Notice and Acknowledgements... 1 Table of Contents... 2 Table of Figures... 3 Executive Summary... 4 SECTION 1 - Station Location... 5 SECTION 2 - Instrumentation and Equipment... 6 SECTION 3 - Collection and Maintenance... 7 SECTION 4 - Recovery and Validation... 8 Test Definitions... 9 Sensor Statistics SECTION 5 - Summary SECTION 6 - Graphs Wind Speed Time Series Wind Speed Distribution Monthly Average Wind Speeds Diurnal Average Wind Speeds Turbulence Intensities Wind Rose APPENDIX A - Sensor Performance Report Test Definitions Sensor Statistics: December 2004 November Sensor Statistics: December Sensor Statistics: January Sensor Statistics: February Sensor Statistics: March Sensor Statistics: April Sensor Statistics: May Sensor Statistics: June Sensor Statistics: July Sensor Statistics: August Sensor Statistics: September Sensor Statistics: October Sensor Statistics: November APPENDIX B - Plot Wind Speed Distribution Monthly Average Wind Speed Diurnal Average Wind Speed Wind Rose April 3, 2006 Renewable Energy Research Laboratory Page 2

4 TABLE OF FIGURES Figure 1 Map of Savoy site... 5 Figure 2 50m data tower in Savoy during installation... 6 Figure 3 - Wind Speed Time Series, December 2004 November Figure 4 Wind Speed Distribution, December 2004 November Figure 5 Average Monthly Wind Speeds, December 2004 November Figure 6 Diurnal Average Wind Speeds, December 2004 November Figure 7 Turbulence Intensity vs. Wind Speed, December 2004 November Figure 8 - Wind Rose, December 2004 November April 3, 2006 Renewable Energy Research Laboratory Page 3

5 EXECUTIVE SUMMARY All the work presented in this Wind Report including installation and decommissioning of the meteorological tower and instrumentation, and the data analysis and reporting was preformed by the Renewable Energy Research Laboratory (RERL) at the. Wind monitoring equipment was first installed in Savoy on November 8, Anemometers and wind direction vanes are installed at 20m, 39m, and 50m above the tower base. are collected via card swap-out on a bi-montly basis. This yearly report is the final wind data report for Savoy and summarizes data collected from December 1, 2004 to November 30, The mean recorded wind speed over this period was 5.84 m/s (13.1 mph) at 50 m; prevailing wind direction at 39 m was from the WNW. The average wind shear factor of 0.32, was calculated from data from the 50 m and 39 m anemometers. The average turbulence intensity at 50m was 0.2, well within the normal values recorded at other sites in western MA. The data logger has had trouble connecting with UMass to download the data each week. This has been addressed by periodic visits to the site to collect the data manually. The source of the problem has been unclear. Furthermore, on October 6 th the data logger ceased to be able to write to its storage media. This failure, in effect, prevented data collection until the logger was replaced on December 16 th. The cause of the data logger failure is believed to be a shorted battery. Other issues that have resulted in the loss of data include multiple icing events in January and February, a data overflow (March 20, 2005 to March 30, 2005) due to the logger being unreachable, and a spurious timestamp followed by a period of no data (August 12, 2005 to August 22, 2005) most probably caused by a nearby lightning strike. In early December 2005 the Savoy met tower was acquired by a commercial developer, therefore this report only covers data collected until the end of November. Additional information about interpreting the data presented in this report can be found in the Fact Sheet, Interpreting Your Wind Resource, produced by RERL and the Massachusetts Technology Collaborative (MTC). This document is found through the RERL website: * 1 m/s = mph. April 3, 2006 Renewable Energy Research Laboratory Page 4

6 SECTION 1 - Station Location The Savoy, MA station is located on privately owned land on a cleared hilltop ft tall trees surround the site in every direction. Several trees were removed to create a clearing for the tower approximately 250 ft in diameter. There is a 1-2ft layer of topsoil above the rock surface. Poor drainage results in the area being very wet, almost boggy. The location of the tower base is N, W (NAD 27). Tower site Figure 1 Map of Savoy site. Source: April 3, 2006 Renewable Energy Research Laboratory Page 5

7 SECTION 2 - Instrumentation and Equipment Wind monitoring equipment is mounted on a standard WindMast 50-meter tall 6in diameter tilt-up guyed tower purchased from Second Wind Inc. Four rock anchors were installed and proof tested at 7,400-8,000 lbs load for 10 minutes (equal to 160mph loading without ice). Wind vanes and anemometers are located at three heights on the tower: 20m, 39m, and 50m. Redundant anemometers exist at 39m and 50m. Additional equipment and models: NRG model 9300 Cellogger 5 #40 Anemometers, standard calibration (Slope m/s, Offset m/s) 3 - #200P Wind direction vanes 3 Sensor booms, 54 length 4 Rock anchors Lightning rod and grounding cable Shielded sensor wire Figure 2 50m data tower in Savoy during installation. April 3, 2006 Renewable Energy Research Laboratory Page 6

8 SECTION 3 - Collection and Maintenance The following maintenance/equipment problems occurred during the report period, and the following corrective actions taken: The data logger has had trouble connecting with UMass to download the data each week. This has been addressed by periodic visits to the site to collect the data manually. The source of the problem has been unclear. Furthermore, on October 6 th the data logger ceased to be able to write to its storage media. This failure, in effect, prevented data collection until the logger was replaced on December 16 th. The cause of the data logger failure is believed to be a shorted battery. Other issues that have resulted in the loss of data include multiple icing events in January and February, a data overflow (March 20, 2005 to March 30, 2005) due to the logger being unreachable, and a spurious timestamp followed by a period of no data (August 12, 2005 to August 22, 2005) most probably caused by a nearby lightning strike. Interestingly, as of mid-december 2004 both 50 m anemometers seem to be functioning correctly; up until this point, only one 50 m anemometer had been working. The 20 and 39 m vanes have evidenced recurring problems but enough data has been available to correctly determine the wind direction. The 50 m vane ceased operation on May 15, All data have been examined and faulty data has been manually flagged. Statistics Summary Date Mean Wind Speed 50 m, [m/s] Max Wind Speed 50 m, [m/s] Turbulence Intensity Prevailing Wind Direction 50 m, [ ] Good WS/WD 50 m, [%] Mean Wind Speed 39 m, [m/s] Max Wind Speed 39 m, [m/s] Turbulence Intensity * The 50 m wind vane failed in mid-may 2004, so the prevailing directions are not reported. Prevailing Wind Direction 39 m, [ ] Good WS/WD 39 m, [%] Wind Shear Exponent n/a [] Heights, units 50 m, [ ] 39 m, [ ] Dec-04 * 37/0 37/36 Jan-05 66/0 66/66 Feb-05 74/0 78/78 Mar-05 60/0 60/60 Apr / / May / / Jun / / Jul / / Aug-05 74/0 74/74 Sep / / Oct / / Nov / / Jan 05 Nov / / April 3, 2006 Renewable Energy Research Laboratory Page 7

9 Wind data statistics in the table are reported when more than 90% of the data during the reporting period are valid. In cases when a larger amount of data are missing, the percent of the available data that are used to determine the data statistics is noted. No measurement of wind speed can be perfectly accurate. Errors occur due to anemometer manufacturing variability, anemometer calibration errors, the response of anemometers to turbulence and vertical air flow and due to air flows caused by the anemometer mounting system. Every effort is made to reduce the sources of these errors. Nevertheless, the values reported in this report have an expected uncertainty of about ± 2% or ± 0.2 m/s, whichever is greater. When data at multiple heights are available, shear coefficients, α, have been determined. They can be used in the following formula to estimate the average wind speed, U(z), at height z, when the average wind speed, U(z r ), at height z r is known: U ( z) = U( z r ) The change in wind speed with height is a very complicated relationship related to atmospheric conditions, wind speed, wind direction, time of day and time of year. This formula may not provide the correct answer at any given site. Nevertheless the calculated shear coefficient, based on measurements at two heights, can be used to characterize the degree of increase in wind speed with height at a site. SECTION 4 - Recovery and Validation All raw wind data are subjected to a series of tests and filters to weed out data that are faulty or corrupted. Definitions of these quality assurance (QA) controls are given below under Test Definitions and Sensor Statistics. These control filters were designed to automate the quality control process and used many of the previous hand-worked data sets made at UMass to affect a suitable emulation. The gross percentage of data recovered (ratio of the number of raw data points received to data points expected) and net percentage (ratio of raw data points which passed all QA control tests to data points expected) are shown below. z z Gross Recovered [%] Net Recovered [%] r α The high gross data recovery percentage indicates that, when operational, the logger was performing well. The lower net data recovery is the result of sensor failure and icing (as noted above in Section 3). April 3, 2006 Renewable Energy Research Laboratory Page 8

10 Test Definitions All raw data were subjected to a series of validation tests, as described below. The sensors tested and the parameters specific to each sensor are given in the Sensor Performance Report which is included in APPENDIX A. which were flagged as invalid were not included in the statistics presented in this report. MinMax Test: All sensors are expected to report data values within a range specified by the sensor and logger manufacturers. If a value falls outside this range, it is flagged as invalid. A data value from the sensor listed in Test Field 1 (TF1) is flagged if it is less than Factor 1 (F1) or greater than Factor 2. This test has been applied to the following sensors (as applicable): wind speed, wind speed standard deviation, wind direction, temperature, and solar insolation. F1 > TF1 > F2 MinMaxT Test: This is a MinMax test for wind direction standard deviation with different ranges applied for high and low wind speeds. A wind direction standard deviation data value (TF1) is flagged either if it is less than Factor 1, if the wind speed (TF2) is less than Factor 4 and the wind direction standard deviation is greater than Factor 2, or if the wind speed is greater than or equal to Factor 4 and the wind direction standard deviation is greater than Factor 3. (TF1 < F1) or (TF2 < F4 and TF1 > F2) or (TF2 F4 and TF1 > F3) Icing Test: An icing event occurs when ice collects on a sensor and degrades its performance. Icing events are characterized by the simultaneous measurements of nearzero standard deviation of wind direction, non-zero wind speed, and near- or belowfreezing temperatures. Wind speed, wind speed standard deviation, wind direction, and wind direction standard deviation data values are flagged if the wind direction standard deviation (CF1) is less than or equal to Factor 1 (F1), the wind speed (TF1) is greater than Factor 2 (F2), and the temperature (CF2) is less than Factor 3 (F3). To exit an icing event, the wind direction standard deviation must be greater than Factor 4 (F4). CF1 F1 and TF1 > F2 and CF2 < F3 CompareSensors Test: Where primary and redundant sensors are used, it is possible to determine when one of the sensors is not performing properly. For anemometers, poor performance is characterized by low data values. Therefore, if one sensor of the pair reports values significantly below the other, the low values are flagged. At low wind speeds (Test Fields 1 and 2 less than or equal to Factor 3) wind speed data are flagged if the absolute difference between the two wind speeds is greater than Factor 1. At high April 3, 2006 Renewable Energy Research Laboratory Page 9

11 wind speeds (Test Fields 1 or 2 greater than Factor 3) wind speed data are flagged if the absolute value of the ratio of the two wind speeds is greater is greater than Factor 2. [ TF1 F3 and TF2 F3 and abs(tf1 - TF2) > F1 ] or [ (TF1 > F3 or TF2 > F3) and (abs(1 - TF1 / TF2) > F2 or abs(1 - TF2 / TF1) > F2) ] Sensor Statistics Expected : the total number of sample intervals between the start and end dates (inclusive). Actual : the total number of data points recorded between the start and end dates. % Recovered: the ratio of actual and expected data points (this is the gross data recovered percentage). Out of Range: total number of hours for which data were flagged according to MinMax and MinMaxT tests. These tests flag data which fall outside of an expected range. of Icing: total number of hours for which data were flagged according to Icing tests. This test uses the standard deviation of wind direction, air temperature, and wind speed to determine when sensor icing has occurred. of Fault: total number of hours for which data were flagged according to CompareSensors tests. These tests compare two sensors (e.g. primary and redundant anemometers installed at the same height) and flag data points where one sensor differs significantly from the other. % Good: the filter results are subtracted from the gross data recovery percentage to yield the net data recovered percentage. SECTION 5 - Summary This report contains several types of wind data graphs. Unless otherwise noted, each graph represents data from 1 quarter (3 months). The following graphs are included: In Figure 3, 10-minute average wind speeds are plotted against time for all data starting on December 1, 2004, through November 30, Immediately evident are the periods of missing data (early December, late March, and late August). April 3, 2006 Renewable Energy Research Laboratory Page 10

12 Wind Speed Distribution A histogram plot giving the percentage of time that the wind is at a given wind speed is shown in Figure 4. This plot shows that the wind speeds ranged between 4 and 5 m/s (8.9 and 11.2 mph) 14.5 % of the time. Monthly Averages A plot of the average monthly wind speed for each month of data is shown in Figure 5, starting December 2004, through November This graph shows that Winter average wind speeds are generally higher for Savoy. December though March, and August averages are not shown due to large amounts of missing and/or invalidated data. Diurnal Average Wind Speeds Figure 6 is a plot of the average wind speed for each hour of the day. The hourly average varied between 5.2 and 6.1 m/s, with the highest average speeds between 9 PM and 1 AM. The lowest wind speeds where between 6 AM and 11 AM. Turbulence Intensity A plot of turbulence intensity as a function of wind speed is shown in Figure 7. Turbulence Intensity is calculated as the standard deviation of the wind speed divided by the wind speed and is a measure of the gustiness of a wind resource. Lower turbulence results in lower mechanical loads on a wind turbine. In general, turbulence intensities range from 0.1 to 0.6 for Savoy. Though some intensities are higher; the average turbulence intensity was 0.2. Wind Rose Figure 8 is a plot, by compass direction, showing the percentage of time that the wind comes from a given direction and the average wind speed in that direction at the 39 m level. at 50 m is not reported, due to the failed vane at 50 m. This wind rose shows that the prevailing wind direction at the 24 m level comes from the WNW. Wind blew from this direction 20% of the time with a mean wind speed of 6.9 m/s. SECTION 6- Graphs for the wind speed histograms, monthly and diurnal average plots, and wind roses are included in APPENDIX B. April 3, 2006 Renewable Energy Research Laboratory Page 11

13 Wind Speed Time Series Figure 3 - Wind Speed Time Series, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 12

14 Wind Speed Distribution Figure 4 Wind Speed Distribution, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 13

15 Monthly Average Wind Speeds Figure 5 Average Monthly Wind Speeds, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 14

16 Diurnal Average Wind Speeds Figure 6 Diurnal Average Wind Speeds, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 15

17 Turbulence Intensities Figure 7 Turbulence Intensity vs. Wind Speed, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 16

18 Wind Rose Savoy Wind Rose, 39 m N 25 NW 20 NE W E SW 20 SE 25 S Percent Time [%] Mean Wind Speed [m/s] Plot by DQMS3 - dqms@dqms.com Figure 8 - Wind Rose, December 2004 November 2005 April 3, 2006 Renewable Energy Research Laboratory Page 17

19 APPENDIX A - Sensor Performance Report Test Definitions Test Order Test Field1 Test Field2 Test Field3 Calc Field1 Calc Field2 Calc Field3 Test Type Factor 1 Factor 2 Factor 3 Factor 4 1 TimeTest Insert 2 Itemp2aDEGC MinMax Batt2aVDC MinMax Etmp2aDEGC MinMax EtmpSD2aDEGC MinMax Anem50aMS MinMax Anem50bMS MinMax Anem39aMS MinMax Anem39bMS MinMax Anem20aMS MinMax AnemSD50aMS MinMax AnemSD50bMS MinMax AnemSD39aMS MinMax AnemSD39bMS MinMax AnemSD20aMS MinMax Vane50aDEG MinMax Vane39aDEG MinMax Vane20aDEG MinMax VaneSD50aDEG Anem50yMS MinMaxT VaneSD39aDEG Anem39yMS MinMaxT VaneSD20aDEG Anem20aMS MinMaxT Anem50aMS AnemSD50aMS Vane50aDEG VaneSD50aDEG Etmp2aDEGC Icing Anem50bMS AnemSD50bMS Vane50aDEG VaneSD50aDEG Etmp2aDEGC Icing Anem39aMS AnemSD39aMS Vane39aDEG VaneSD39aDEG Etmp2aDEGC Icing Anem39bMS AnemSD39bMS Vane50aDEG VaneSD39aDEG Etmp2aDEGC Icing Anem20aMS AnemSD20aMS Vane20aDEG VaneSD20aDEG Etmp2aDEGC Icing Anem50aMS Anem50bMS CompareSensors Anem39aMS Anem39bMS CompareSensors Anem50yMS MinMax Anem39yMS MinMax AnemSD50yMS MinMax AnemSD39yMS MinMax Turb50zNONE MinMax Turb39zNONE MinMax Turb20zNONE MinMax Pwrd50zWMC MinMax Pwrd39zWMC MinMax Pwrd20zWMC MinMax Wshr0zNone MinMax -1 4 April 3, 2006 Renewable Energy Research Laboratory Page 18

20 Sensor Sensor Statistics: December 2004 November 2005 Expected Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 19

21 Sensor Expected Sensor Statistics: December 2004 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 20

22 Sensor Expected Sensor Statistics: January 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 21

23 Sensor Expected Sensor Statistics: February 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 22

24 Sensor Expected Sensor Statistics: March 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 23

25 Sensor Expected Sensor Statistics: April 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 24

26 Sensor Expected Sensor Statistics: May 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 25

27 Sensor Expected Sensor Statistics: June 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 26

28 Sensor Expected Sensor Statistics: July 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 27

29 Sensor Expected Sensor Statistics: August 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 28

30 Sensor Expected Sensor Statistics: September 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 29

31 Sensor Expected Sensor Statistics: October 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 30

32 Sensor Expected Sensor Statistics: November 2005 Actual % Recovered Out of Range of Icing of Fault % Good Itemp2aDEGC Batt2aVDC Anem50aMS AnemSD50aMS Anem50bMS AnemSD50bMS Anem39aMS AnemSD39aMS Anem39bMS AnemSD39bMS Anem20aMS AnemSD20aMS Vane50aDEG VaneSD50aDEG Vane39aDEG VaneSD39aDEG Vane20aDEG VaneSD20aDEG Etmp2aDEGC EtmpSD2aDEGC Total April 3, 2006 Renewable Energy Research Laboratory Page 31

33 APPENDIX B - Plot Wind Speed Distribution Bin Center Wind Speed [m/s] Percent of Time [%] Table 1: Wind Speed Distribution April 3, 2006 Renewable Energy Research Laboratory Page 32

34 Monthly Average Wind Speed Date Mean [m/s] Jan Feb Mar 2005 n/a Apr May Jun Jul Aug 2005 n/a Sep Oct Nov Table 2 - Wind Speed Averages April 3, 2006 Renewable Energy Research Laboratory Page 33

35 Diurnal Average Wind Speed Hour of Day Average Wind Speed [m/s] Table 3 - Diurnal Average Wind Speeds April 3, 2006 Renewable Energy Research Laboratory Page 34

36 Wind Rose Percent Time Mean Wind Speed Direction [%] [m/s] N NNE NE ENE E ESE SE SSE S SSW SW WSW W WNW NW NNW Table 4: Wind Rose, Time Percentage and Mean Wind Speed by Direction April 3, 2006 Renewable Energy Research Laboratory Page 35