Posters

Abstract

Wind data measurements are used to calibrate and extend synthetic data to the long term. Typically, these measurements cover relatively short periods that may not be representative of the long term behavior of the site. We found a direct impact and variability of long term series when using different periods for calibrations. This methodology aims to discuss the existence of a subset of the observations to use in the calibration instead of the whole period, based on the long term representativeness. We are quantifying how choosing different periods for calibration has an impact on the resulting calibrated time series. Moreover, we propose an algorithm to choose the best observation period for calibration. It is then shown that, contrary to popular belief, using a smart selected subset rather than the full available period of the observations improves the representativeness of the long term.