Mostrar registro simples

dc.contributor.advisorGuetter, Alexandre Kolodynskiept_BR
dc.contributor.authorStival, Leandro Jose Lemespt_BR
dc.contributor.otherAndrade, Fernando Oliveirapt_BR
dc.contributor.otherUniversidade Federal do Paraná. Setor de Tecnologia. Programa de Pós-Graduação em Engenharia de Recursos Hídricos e Ambientalpt_BR
dc.date.accessioned2022-05-23T12:09:03Z
dc.date.available2022-05-23T12:09:03Z
dc.date.issued2017pt_BR
dc.identifier.urihttps://hdl.handle.net/1884/49088
dc.descriptionOrientador : Prof. Dr. Alexandre Kolodynskie Guetterpt_BR
dc.descriptionCoorientador : Prof. Dr. Fernando Oliveira Andradept_BR
dc.descriptionDissertação (mestrado) - Universidade Federal do Paraná, Setor de Tecnologia, Programa de Pós-Graduação em Engenharia de Recursos Hídricos e Ambiental. Defesa: Curitiba, 04/04/2017pt_BR
dc.descriptionInclui referências : f.91-97pt_BR
dc.description.abstractResumo: A energia eólica cresceu significativamente, porém a eficiência de geração da fonte eólica gira em torno de 30% da energia cinética disponível no vento. Por este motivo é de extrema importância que estudos sejam elaborados afim de aumentar a eficiência na geração de energia. Este estudo visa investigar odesempenho da geração de energia eólica em dois parques eólicos situadosna América do Norte, através da investigação dos dados de vento e modelagem da esteira turbulenta. Os dados analisados são provenientes da turbina, SCADA, e dados coletados pelo LiDAR. A partir das análises dos dados de vento foram estimados parâmetros como tensão de cisalhamento do vento, rosa dos ventos, perfil de velocidade do vento e intensidade de turbulência. Portanto, aumentando a intensidade de turbulência, a energia gerada é superestimada em moderadas velocidades do vento e subestimada em altas velocidades do vento. Enquanto isso, os coeficientes de cisalhamento do vento variaram entre 0 e 0.2 para altas velocidades. Além disso, coeficientes de cisalhamento com valores elevados, perto de 0.4, foram encontrados em baixas velocidades do vento. Este trabalho visou também comparar modelos de esteira turbulenta de PARK (Jensen), Frandsen, Larsen and Eddy Viscosity (Ainslie) com resultados obtidos pelo LiDAR, além disso foi realizado uma simulação numérica da esteira turbulenta utilizando Fluent CFD com as equações médias de Reynolds (RANS) que resolvem o modelos de duas equações diferenciais para obter a viscosidade turbulenta. A turbulência foi fechada pelo modelo ?????, sendo o modelo de esteira turbulenta desenvolvido para uma única turbina num terreno plano. O modelo de PARK obteve os melhores resultados para linha de centro longitudinal em relação as velocidades de 6 a 8 ms-1. Entretanto, para velocidade de 9 a 12 ms-1 , o modelo de EDDY VISCOSITY apresentou o melhor desempenho. As análises de seção transversal apresentaram o modelo de PARK como melhor resultado para 500 m. Enquanto isso, para 700 m de seção transversal, o melhor desempenho foi obtido pelo modelo de LARSEN. Palavras-chave: Energia Eólica, Modelo de Esteira Turbulenta, LiDAR, Análises do Vento.pt_BR
dc.description.abstractAbstract: Wind power has gained significant share in the global power production. However, the wind power output efficiency is only about 30% of the wind kinetic energy. Because of that, it is essential to study the efficiency of these power generation systems by assessing the effects that wind parameters and wakes will have on the whole system. Hence, a complete assessment of wind resources is crucial to retain full advantage of wind power. This study aims to investigate the efficiency of wind energy generation in two North American Wind Farms, through wind data investigation and wake modeling. The data analyzed are the SCADA data and the data collected by LiDAR measurements. The wind data analysis has estimated parameters as wind shear, wind rose, wind speed profile and turbulence intensity. Therefore increasing turbulence intensity the power output is overestimated at moderate wind speeds and underestimated at higher wind speeds. Meanwhile, the wind shear coefficients were found to vary between 0 and 0.2 at higher inflow velocities. High wind shear values, close to 0.4, were recorded for lower inflow velocities. The goal of the wake models is to simulate the turbine induced wind speed deficits and the ratio of restoration to the free stream velocity. This work has compared the PARK (Jensen), Frandsen, Larsen and Eddy Viscosity (Ainslie) models with LiDAR wake measurements, besides that it has performed a numerical simulation of the wind turbine wake using the Fluent CFD with the Reynolds Averaged Navier Stokes (RANS) equations that solves two differential equation model to obtain turbulent viscosity. The turbulence was closed by the ????? model, where the wake modeling has been developed for a single turbine on a flat terrain. In terms of centerline wake analysis, the PARK wake model yielded the best velocity simulations for inflow winds from 6 to 8 ms-1. However, the EDDY VISCOSITY wake model yielded the best performance for wind speed bins from 9 to 12 ms-1. The cross section wake analysis presented for the 500 m cross section, the domination by the PARK model. Meanwhile, along the 700 m cross section, the LARSEN wake model produced the best simulations. Key-words: Wind Power, Wake Modeling, LiDAR, Wind Analysis.pt_BR
dc.format.extent97 f. : il. algumas color., grafs., tabs.pt_BR
dc.format.mimetypeapplication/pdfpt_BR
dc.languageInglêspt_BR
dc.relationDisponível em formato digitalpt_BR
dc.subjectRecursos hídricospt_BR
dc.subjectEnergia eólicapt_BR
dc.subjectVentospt_BR
dc.subjectTesespt_BR
dc.titleA study on wind assessment on the wind power performance : wind shear and turbulence intensity effects besides the wake modeling for a single wind turbinept_BR
dc.typeDissertaçãopt_BR


Arquivos deste item

Thumbnail

Este item aparece na(s) seguinte(s) coleção(s)

Mostrar registro simples