CIMEC   24726
CENTRO DE INVESTIGACION DE METODOS COMPUTACIONALES
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Computational Models and Numerical Methods for the Design of Next Generation 15 MW HAWT
Autor/es:
FRANCK GERARDO; DORSCH, JUAN PABLO; NIGRO, NORBERTO M.; GENTILE CARLOS; COSIMO, ALEJANDRO; CARDONA ALBERTO; GIMENEZ, JUAN M.; F. CUGNON
Lugar:
Cork
Reunión:
Conferencia; 2019 Wind Energy Science Conference; 2019
Institución organizadora:
European Academy of Wind Energy
Resumen:
This work deals with the development of computational models and appropriate numerical methods to give an answer, with a good compromise between performance and accuracy, for the design of next generation 15 MW horizontal axis wind turbines. One of the most important factors to increase the power of wind turbines, which is pushing the design of next generation of wind turbines, is the use of longer blades and higher towers. This factor is driving research and development for the construction of more powerful, efficient, durable and profitable turbines. All this added to other innovations that make the manufacture of turbines easier and cheaper: the creation of turbines that collect and interpret data in real time in order to act mechanically to adapt to flow conditions, maximizing energy harvesting and taking care of the riskiest situations where the stoppage is imminent.Optimal design of wind turbines and wind energy farms requires a comprehensive understanding of the physics of multi-scale wind flows, structural mechanics and materials performance, among other things. A first requirement is knowledge of isolated phenomena on different scales and several individual components. However, wind turbines and wind energy farms are complex systems involving close interactions of various phenomena (fluid, structural, mechanical, electrical) on multiple temporal and spatial scales. A thorough understanding of these interactions is needed to optimize the performance of wind turbines and wind energy parks. It is well known that the progress in terms of computer hardware experienced in recent years has allowed to go beyond the frontiers of knowledge, with increasingly demanding challenges, including those of coupling different physical problems to have a greater realism in the simulations. It is precisely this growth of supply and demand that pushes the improvement of mathematical models and numerical methods.Although longer and slender blades can collect more energy, it is well known that they experience greater deflections compromising the efficiency of the installation and its structural integrity. On the other hand, these changes in the design increase the problem complexity because of the great disparity of flow conditions that exist between the tips of the blades with respect to the central zone.Regarding the technical aspects of the simulation, we face a problem where besides the changing flow conditions, the blades rotation and deformation make the geometry time varying which imposes the mesh to be adapted. On the other hand, the rotation and deformation of the blades depend strongly on the fluid and the blades themselves. Besides, a fine enough spatial and temporal discretization is required to capture the effects of turbulency and the fluid / structure coupling itself. In this work in progress, we present the advances in a fluid / structure coupling scheme for wind turbines, between the commercial codes StarCCM+ for the fluid and Mecano for the structural behavior of the turbine. Also, free software based on OpenFOAM with their own developments is employed, comparing techniques such as sliding mesh with simpler ones such as AMI or GDI to solve the mesh dynamics and adaption. Algorithms for generating special meshes that facilitate the transmission of force at the interfaces are also evaluated, in terms of accuracy, conservation and efficiency. All these developments are made for parallel architectures exploiting scalability.