Comunicaciones y ponencias de congresos DEIO - EIOS Biltzarretako komunikazioak eta txostenak

Permanent URI for this collection

http://hdl.handle.net/2454/10493

Browse

Browsing Comunicaciones y ponencias de congresos DEIO - EIOS Biltzarretako komunikazioak eta txostenak by Author "Estadística, Informática y Matemáticas"

Now showing 1 - 3 of 3
  • Thumbnail Image
    PublicationOpen Access
    Support vector machines in the wind energy framework: a new model for wind energy forecasting
    (European Wind Energy Association, 2009) Frías Paredes, Laura; Pascal, Edurne; Irigoyen, Uxue; Cantero Nouqueret, Elena; Loureiro, Yolanda; Lozano, Sergio; Fernandes Correia, Pedro Miguel; Martí, Ignacio; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
    In this work we expose a new model for wind energy forecasting based in Support Vector Machines. Support vector machines (SVMs) are a group of supervised learning methods that can be applied to classification or regression. They represent an extension to nonlinear models of the generalized portrait algorithm developed by Vladimir Vapnik. Support vector machines (SVMs) appeared in the early nineties as optimal margin classifiers in the context of Vapnik's statistical learning theory. Since then SVMs have been successfully applied to real-world data analysis problems, often providing improved results compared with other techniques. The SVMs operate within the framework of regularization theory by minimizing an empirical risk in a well-posed and consistent way. A clear advantage of the support vector approach is that sparse solutions to classification and regression problems are usually obtained: only a few samples are involved in the determination of the classification or regression functions. This fact facilitates the application of SVMs to problems that involve a large amount of data. Joint to that, the use of kernel functions into their algorithms allows the adaptability to nonlinear problems of SVMs. There are many "traditional" problems in the wind power forecasting systems that affect the accuracy and robustness of the forecasts, as availability of real power measures, the existence of wrong data and the nonlinearity of the relationship between meteorological variables and power. So, these robustness and adaptability of SVMs suggest that these types of learning machines could be a good set of tools to solve many problems in the wind energy context.
  • Thumbnail Image
    PublicationOpen Access
    TOUCHLESS: demonstrations of contactless haptics for affective touch
    (ACM, 2023) Chew, Sean; Dalsgaard, Tor-Salve; Maunsbach, Martin; Bergström, Joanna; Seifi, Hasti; Hornbæk, Kasper; Irisarri Erviti, Josu; Ezcurdia Aguirre, Íñigo Fermín; Iriarte Cárdenas, Naroa; Marzo Pérez, Asier; Frier, William; Georgiou, Orestis; Sheremetieva, Anna; Kwarciak, Kamil; Stroiński, Maciej; Hemmerling, Daria; Maksymenko, Mykola; Cataldo, Antonio; Obrist, Marianna; Haggard, Patrick; Subramanian, Sriram; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
    A set of demonstrators of contactless haptic principles is described in this work. The technologies are based on electrostatic piloerection, chemical compounds and ultrasound. Additionally, applications related to affective touch are presented, ranging from storytelling to biosignal transfer, accompanied with a simple application to edit dynamic tactile patterns in an easy way. The demonstrators are the result of the Touchless project, which is a H2020 european collaborative project that integrates 3 universities and 3 companies. These demostrators are contactless haptic experiences and thus facilitate the come-and-interact paradigm, where users can approach the demo booth and directly experience the applications without having to wear devices, making the experience fast and hygienic.
  • Thumbnail Image
    PublicationOpen Access
    Wind power predictability assessment from large to local scale
    (European Wind Energy Association, 2013) Rodrigo, J.; Frías Paredes, Laura; Stoffels, Nicole; Bremen, Lueder von; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
    Large wind power penetration and efficient wind farm operation require early consideration of the impact of wind power predictability from site to regional/country and continental levels. The aim of this work is to explore the added value of introducing wind power predictability in the wind resource assessment phase for large scale spatial planning and for wind farm performance evaluation. To this end, historical wind data are converted to virtual time series of wind power production on which forecasting models operate in hindcast mode to estimate predictability information. This information can be used to anticipate operational costs and lead to a better assessment of the cost-benefit of wind energy deployment. Numerical weather prediction models give us the opportunity to evaluate wind power predictability and other forecast skills. The COSMO-EU mesoscale model wind speed data is used to create a spatial dataset of different forecast skills and visualize those in form of European predictability maps at a resolution of 7×4 km. In addition to a large capacity factor, low wind variability is also desirable for smooth wind power integration. A smoothing factor determines the areas that benefit most from wind farm aggregation resulting in increased predictability and reduced variability. These forecast skill maps provide useful information for spatial planning and can be analyzed to determine the sources of forecasting errors, either by topographical effects or meteorological phenomena. At site level, wind measurements from a reference mast are converted into wind farm production time series making use of a microscale flow model. Then CENER's LocalPred wind power forecasting model is run, emulating operational conditions to assess predictability. A test case comprising two wind farms in Spain of different terrain complexity, equipped with measurements from the resource assessment and the operational phases, is presented to illustrate a methodology for local predictability assessment. Low wind power predictability is certainly a good indicator of quality-of-energy for grid integration. Hence, this type of information should be included during the development phase of a wind farm as complementary information about its performance. In conclusion, guidelines on how to produce and evaluate wind power predictability during the planning phase are provided in order to advise end-users on the use of these new layers of information.