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.

Sonia España Cuadrado ALSTOM RENEWABLE POWER, Spain
Co-authors:
Sonia España Cuadrado (1) F P
(1) ALSTOM RENEWABLE POWER, Barcelona, Spain

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Sonia España is working in the wind engineering department at Alstom Wind in Barcelona. She started in wind energy field 5 years ago, working both on commercial projects and wind assessment research. Sonia has a Master on Meteorology and Physics degree.

Abstract

Long term correction performed with reanalysis data and its impact on uncertainty analysis

Introduction

When performing a site wind resource assessment, onsite measurement data usually require correction to achieve long term representativeness. Seldomly a conventional meteorological mast with long term data is available to correlate with short term measured wind data. In other cases Reanalysis data sets are often used to apply long term corrections. However, the uncertainty associated to the long term correction remains one of the main sources of uncertainty which is difficult to estimate. This work investigates a method to characterize the uncertainty in the particular case of long term correction using Reanalysis data.

Approach

The aim of a long term correction using Reanalysis data is to reduce the uncertainty on wind variability over time. However, it introduces new uncertainties due to representativeness and consistency of the Reanalysis data series. In some cases the necessity of a long term correction can be discussed when several years of data are available or bad correlation is observed between long term reference data and on-site wind data. In general, underlying causes of uncertainties need to be better identified when a choice needs to be made in order to select the better long term correction method.

Main body of abstract

For the analysis, data sets from real meteorological masts with at least 7 years of real data have been selected. The whole data series from the masts selected has been divided into several shorter periods of different length: 3 months, 6 months, and from 1 to 7 years. These divided series’ were correlated with Reanalysis data series and corrected to achieve long term representativeness.

The mean wind speed deviation has been calculated comparing the average wind speed measured on the meteorological mast against the average wind speed from the divided series with and without long term correction applied. Also a test has been done comparing the average wind speed from the Reanalysis data against the wind speed from the meteorological mast.

The analysis focused on three different results: uncorrected short-term measured wind speed, corrected long term wind speed, and Reanalysis wind speed. Associated errors were assessed through the original data set from the meteorological mast.


Conclusion

The mean wind speed error introduced in the long term correlation has been estimated and compared for different data sets, concluding that Reanalysis is a consistent source of long term data.

Sensitivity to the short term data period duration has been analysed and it has led to characterize the statistical dependency of the Long Term correlation error with the measured data periods.



Learning objectives
Evaluate the consistency of long term correlation through Reanalysis data.
Quantify uncertainty due to long term correction.
Study the sensitivity of the available data period on the uncertainty analysis.