Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters in this topic directly after the session 'Wind speed predictions: Are we at the limit of our knowledge or can we improve?' taking place on Wednesday, 12 March 2014 at 11:15-12:45. The meet-the-authors will take place in the poster area.

RAFAEL ZUBIAUR BARLOVENTO RECURSOS NATURALES S.L., Spain
Co-authors:
RAFAEL ZUBIAUR (1) F P DANIEL ORTIZ (1) RICARDO MARTINEZ (1)
(1) BARLOVENTO RECURSOS NATURALES S.L., LOGROÑO, Spain

Printer friendly version: printer.gif Print

Abstract

Advanced virtual series for long term assessment

Introduction

The estimation of long term wind characteristics is crucial to develop a wind farm project. Reference meteorological stations (when it is possible), reanalysis data, and during the last years the use of virtual series, derived by means mesoscale meteorological models, has become widespread with this purpose. The traditional MCP techniques, that used these reference data and the site measurements to long term extrapolation, do this by means linear adjustment. Nevertheless meteorological phenomena are non linear and introduce uncertainty to long term estimations.

Approach

This paper presents the results of the validation of new virtual series, named “re-modeling”. The remodeling series are generated using the mesoscale information for different levels and the site measurements by means modern non linear statistical models. These new series developed by VORTEX intend to improve the accuracy and consistency of the virtual series, adjusting them to the wind site characteristics. Barlovento Recursos Naturales has analyzed and validated this product for several sites around the world.

Main body of abstract

The remodeling technology uses the current VORTEX virtual series, including great amount of simulation variables at different levels, and a year of real site measurements. These data in combination with non linear statistical techniques provide new series (re-modeled) of wind speed and direction.
Barlovento analyzed and validated the new series with actual meteorological data. For this purpose, meteorological stations with several years of quality measurements (more than five years) have been used. The new re-modeling series have been compared to actual data (VORTEX and met-mast series), as well to extrapolated series using VORTEX virtual series and standard MCP methods. The study analyzes the consistency of the new series and the best fit to the site wind conditions, analyzing the distribution of wind speed and direction, and their variation along the day. In the comparison, different attributes have been taking into account, and hourly, daily and monthly correlation, wind speed bias, wind speed distribution and direction distribution are presented.



Conclusion

The Re-modeling shows a good performance, improving in some aspects the actual virtual series and the classical MCP methods. The re-modeling shows a better adjustment to the real wind conditions at the site. Their use can be a good complement to actual techniques, and reduce the uncertainty in long term extrapolation


Learning objectives
Analyze a new long term extrapolation method that seeks to improve long term current virtual series and linear MCP techniques.
The use of the available meteorological data, not only wind speed and direction, to extrapolate the wind resource to long term.