Abstract: | This study proposes a method which makes it possible to reduce the number of tests needed to obtain the power curve of a Wind Energy Conversion System (WECS) by extracting more useful information from a test. When using the method of bins, recommended by the International Energy Agency, to determine the power curve, the pre-average occults the wind's irregularity. The study of this irregularity allows to give a weighting to each test data. Before defining this method two lines of study were developed. The first one concerns the regularity of a function and its application to the variation, for a given machine, of the wind vector module over time. The second one deals with the estimation of the mean value of a random variable by the calculation of the mean of the observed values, given a weighting by a coefficient allowing the diminution or increase then importance of a result. For a given place, whose turbulence intensity during the measurements is known, the weighted curve and the unweighted curve should be the same. Using the weighting brings either increase in the accuracy or decrease in the number of tests. This method has been tested on two WECS. The tests were carried out at the Centre National d'Essais Eoliens de Lannion (CNEEL France). Then the results are set out and commented. |