Search

In view of the continuous growth of energy demand and interest in environmental protection, the use of clean energy to replace fossil fuels is a global trend. Wind energy is the fastest growing renewable energy in the world in recent years. However, in the case of Mexico, there are still some difficulties in promoting its use in some areas of the national territory. One difficulty is knowing in advance how much energy can be injected into the grid. This paper introduces the development of artificial intelligence technology for wind power generation prediction based on multi-year meteorological information. In particular, the potential application of Bayesian network in these prediction applications is studied in detail. A weather forecasting method based on Dynamic Bayesian network (RBD) is proposed. The forecasting system was tested using meteorological data from the regional wind energy technology center (CERT) of the National Institute of Electricity and Clean Energy (INEEL) in Oaxaca, Mexico. The results are compared with the time series prediction results. The results show that dynamic Bayesian network is a promising wind power generation prediction tool.

This thesis performs the comparison on patient-reported outcomes (health-related quality of life and patients’ preferences) among radical prostatectomy, external radiotherapy and brachytherapy, in the “Spanish Multicentric Study of Clinically Localized Prostate Cancer”. This is a prospective cohort with consecutive patient recruitment at diagnosis (2003-2005) in 10 hospitals. The impact of the treatments on HRQoL at 5-year follow-up was characterized obtaining directly from the cohort patients their health-related quality of life measured with the Expanded Prostate Index Composite (EPIC), one of the most established instruments among prostate cancer-specific ones. Also patients’ preferences obtained by direct methods, including time trade off, standard gamble and willingness to pay were described. Finally, psychometric properties of the Spanish version of the Patient-Oriented Prostate Utility Scale (PORPUS), the first econometric instrument developed for measuring utilities by the indirect method on patients with localized prostate cancer, were evaluated., Esta tesis hace la comparación de los resultados percibidos por los pacientes, calidad de vida relacionada con la salud (CVRS) y preferencias, entre la prostatectomía radical, radioterapia externa y braquiterapia, en el "Estudio Multicéntrico Español de Cáncer de Próstata Localizado”, una cohorte prospectiva de pacientes reclutados en 10 centros hospitalarios (2003-2005). El impacto de los tratamientos sobre la CVRS de los pacientes fue caracterizado y medido a los 5 años de seguimiento con el “Expanded Prostate Index Composite” (EPIC), uno de los instrumentos específicos para cáncer de próstata más utilizados. Las preferencias de los pacientes también fueron estimadas, a través de métodos directos como el “time trade-off”, “standard gamble” y “willingness to pay”. Finalmente, se evaluaron las propiedades psicométricas de la versión española del “Patient-Oriented Prostate Utility Scale” (PORPUS), el cual es el primer instrumento econométrico desarrollado para la estimación indirecta de utilidades en pacientes con cáncer de próstata localizado.