English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Wind power has experienced an enormous growth in the last years, not only in its installed capacity, but also in its great technological development. Today, wind power forecasts are necessary for the large-scale integration of wind turbines into the electrical grid due to this intermittency nature. Forecasting tools are also necessary to limit the use of spinning reserves, increasing even more the environmental benefits of this kind of energy. Wind power forecasts are necessary to participate in the electricity markets too. The work developed in this thesis is devoted to improve the performance of statistical models used in actual wind energy forecasting tools. To do so, we have begun from a particular case Sipreolico. The main contributions of this thesis are the following: first, we estimate the parameters of the models by maximum likelihood through the Kalman filter and smoother instead of imposing them ad-hoc. This parameter estimation allows us to capture the operating conditions or characteristics of each wind farm to some degree and obtain significantly lower root mean square error (RMSE). Second, we propose new specifications for the state equation, including multivariate models, to capture, through the correlation of the elements of the state vector, the effect caused by omitted inputs. The relationship between wind and generated power is highly nonlinear and timevarying, so it is very common to use alternative models based on different assumptions about the variables involved. We proposed a novel forecast combination procedure through multivariate dimension reduction techniques. We check the strategies shown throughout this thesis in the Sotavento wind farm. At last, all the work developed is included as part of a real time forecasting tool with minimal expert intervention.

Nuevos modelos de predicción eólica basados en series temporales