Wind speed extrapolation based on power law correction through a bayesian heteroskedastic model

In this study we propose a new methodology for extrapolate (in height) the wind speed given a fixed location. This problem is of great importance given that reals measurements at different heights are very costly, however, many times it is necessary to account for the complete profile of speed. In particular, this problem arises in the evaluation of wind energy projects. Although there several methods for extrapolation, such as the power law, there are no consistent methodologies that use real data, mechanistic models, and at the same time can reproduce the stochastic behavior of wind speed. The model we propose translate the wind speed probability distribution to a continuum of heights combining a statistical model (from data taken at selected points) and a theoretical model (power law). We use the Box-Cox transformation and a heteroskedastic Bayesian model that uses the physical model as a prior information. We prove by simulation that the method performs adequately and has a better fit than using just the statistical or the physical model.

• Autores: Rodríguez Ariza, Sebastián
• Fecha de publicación: 2018
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: eng
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/34843
• Acceso en línea: http://hdl.handle.net/1992/34843
• Palabra clave: Metodología en medición de la velocidad de viento; Velocidad del viento - Mediciones - Investigaciones; Velocidad del viento - Extrapolación - Investigaciones; Ingeniería
• Rights: openAccess
• License: https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf