Measurements for the wind industry: moving beyond the 10-minute averaged data

Introduction

Measurements for the wind resource assessment as well as the wind turbine power curve validation are based on an averaging period of 10 minutes. The representativeness of ceratin parameters averaged with this period is examined, as well as the advantages and disadvantages of using a smaller averaging period.

Approach

Wind data obtained with an averaging period of 1 minute are processed and averaged based on a period of 2, 5 and 10 minutes. The effect of using a different period than the industry standard of 10 minutes is examined, according to the changes in the resulting data distribution, general statistics and the measured wind power density.

Main body of abstract

The standard measurement practice in the wind industry has established the use of 10-minute averaged data. This practice has been based on research under the meteorological scope, which examines the time scales of the natural wind phenomena. However, this averaging period might be improperly large for the wind energy industry, as the time scales of the wind energy conversion are far different. Modern wind turbines which present improved operational characteristics, respond to changes in the wind in a fraction of this period. Yet their power curve validation relies on measurements averaged over a period in which the actual wind presents high variations in the real field, while other meteorological factors appear stable. Prior to the power curve test, the wind resource assessment of a site which is based on 10-minute averaged data might overlook variations in the wind power density which are captured by the wind turbine, but are neglected as the data is averaged for a 10-minute interval. This effect could be of higher significance especially in the cut-in wind speed region. All the above could lead to discrepancy between the assessment wind data, the calculated wind turbine yield and the actual energy yield. The possibilities of reducing this discrepancy by using a smaller averaging period, and disengaging the measurements for the wind energy industry from the 10-minute averaging practice are examined in this work.

Conclusion

Narrowing the averaging period in the wind measurements, leads to an observation of improvement in the accuracy regarding the measured wind power density and its variations. By moving the whole measurement chain to a smaller averaging period, new possibilities arise for improved accuracy and representativeness of the results, regarding both the wind resource assessment and the validation of the wind turbine power curve. Nowadays this change can be supported by the measurement technical background, as higher data storage is easily available and affordable, while modern processing power can effortlessly manipulate larger data sets.


Learning objectives
The effect of using various averaging periods on the measurements regarding the wind energy industry is examined, aimed at clarifying the alteration in the wind resource assessment parameters than can also affect the calculated energy yield.