Environmental Engineering Reference
In-Depth Information
the other trees. However, isolated trees or those in small, widely spaced groups should be favored as
wind speed indicators. In case a comparison between several locations is to be made, trees should
be of nearly the same height and species. Near the seashore, flagging may be the result of sea spray
(salt) and not totally due to the wind.
4.6 DATA LOGGERS
Data loggers for wind resource measurements are now the norm. Data are stored on data chips, and
either chips are retrieved or data loggers send information to a base personal computer. The BASE
program monitors the phone lines, answers the call, and determines which site is calling and what
is the status of the data card and call-in schedule (card unread, first call of six tries; card partially
read, fourth call of six; etc.).
For time sequence data, the amount of data is large. For example, suppose you want to measure
wind speeds, wind direction, pressure and temperature (1 Hz sampling rate), average values, and
statistics stored every 10 min. That would be around 130 KB of data per month. A 60 min magnetic
tape will store 180 KB; however, standard data chips now store 16 MB, which is around 2 years of
data. You still need to retrieve the data at least once per month as a check on problems. With phones
or satellite connection, data should be retrieved once per week.
The logistic problems have to be taken care of to ensure high data recovery and the quality of
the data analyzed. Calibration and replacement of sensors must be part of a routine maintenance
program. For example, anemometers should be replaced once per 6 months to 2 years, depending
on the number of revolutions and the environment.
A quality assurance program for flagging suspect data is imperative. Data recovery should
be around 95%. Sensors problems are due to failure, low/no values due to icing, lightning, and
even vandalism. Data loggers and transmission problems can also lead to loss of data. Yearly
failure rates are around 25% for sensors and 10% for data loggers. Rates could be higher for
sites with very harsh conditions, for example, hail, lightning, dust or sandy areas, or extended
periods of high winds.
Generally, there will be two anemometers and one wind vane per level with two or more levels.
If one anemometer is down, there are still data from the second anemometer. If both are not operat-
ing, there is the possibility of estimating values based on data at another level and past wind shear
values. So the 95% data recovery is feasible.
Wind farm developers want the average wind speed (10 min or 1 h) so they can predict energy
production. Data analysis programs, which are fairly flexible, are available. As an example, the
monthly average, minimums, and maximums for each sensor for the month plus selected graphs
and tables are available.
EXAMPLE OF SUMMARY REPORTS AVAILABLE BY MONTH FROM AN ANALYSIS PROGRAM
Comparison of hourly wind speeds (two anemometers at same height or between different heights)
Frequency distributions (calculate wind power/area) ( Figure 4.16 )
Frequency distribution graph
Diurnal wind speed graph
Average turbulence intensity (upper level, use prevailing wind anemometer)
Wind rose graph
Average wind shear table (between two heights)
Average temperature graph
Data can be placed in spreadsheets for further analysis, as most data loggers allow export.
Another benefit is that data analysis is not tied to a proprietary program, which sometimes even
the manufacturer has trouble updating, especially if a subcontractor developed the software
program.
Search WWH ::




Custom Search