SUMMARY Radial flow develops in the bottom of separated boundary layer over rotating blades. Therefore, the effective blade lift and drag for the local blade stations are not the same as for non-rotating airfoils. Nevertheless, wind turbines design is still based on blade-element momentum (BEM) theory, deriving the local lift and drag by 2D measurements. Results obtained are accurate at and about design conditions, but in stalled ones the aerodynamic forces are under-predicted. In this work, full-3D CFD-RANS simulations are carried out throughout different flow conditions for both a simple cylindrical blade and a more realistic geometry, by varying angle of attack and Reynolds number in a rotating framework. In order to analyse the output data, an ad-hoc post-processing tool has been developed, allowing one to evaluate within the computational domain of all of the terms in a modified form of Prandtl’s boundary layer equations. The early results for a simple blade geometry are presented, confirming that 3D loads on a rotating blade are higher than those for a non-rotating one, mostly for inboard sections and separated flow conditions. Then, the well-known NASA/NREL Phase VI wind turbine was considered. The main findings of the cylindrical blade study were confirmed, even if new complexities were introduced by the more realistic rotor model. A more detailed analysis is needed, to explain some discordance found especially for the NREL case. Different flow situations (Reynolds number, angle of attack and rotational speed) will be studied for the simpler geometry whereas pitched operation will be explored with the NREL rotor. The planned development includes the implementation of a correction law of the pure 2D airfoil characteristics, to be used in a BEM based design code.
Study of the rotational effects on wind turbine blades based on full 3-D CFD-RANS computations
MANDAS, NATALINO;CAMBULI, FRANCESCO;
2008-01-01
Abstract
SUMMARY Radial flow develops in the bottom of separated boundary layer over rotating blades. Therefore, the effective blade lift and drag for the local blade stations are not the same as for non-rotating airfoils. Nevertheless, wind turbines design is still based on blade-element momentum (BEM) theory, deriving the local lift and drag by 2D measurements. Results obtained are accurate at and about design conditions, but in stalled ones the aerodynamic forces are under-predicted. In this work, full-3D CFD-RANS simulations are carried out throughout different flow conditions for both a simple cylindrical blade and a more realistic geometry, by varying angle of attack and Reynolds number in a rotating framework. In order to analyse the output data, an ad-hoc post-processing tool has been developed, allowing one to evaluate within the computational domain of all of the terms in a modified form of Prandtl’s boundary layer equations. The early results for a simple blade geometry are presented, confirming that 3D loads on a rotating blade are higher than those for a non-rotating one, mostly for inboard sections and separated flow conditions. Then, the well-known NASA/NREL Phase VI wind turbine was considered. The main findings of the cylindrical blade study were confirmed, even if new complexities were introduced by the more realistic rotor model. A more detailed analysis is needed, to explain some discordance found especially for the NREL case. Different flow situations (Reynolds number, angle of attack and rotational speed) will be studied for the simpler geometry whereas pitched operation will be explored with the NREL rotor. The planned development includes the implementation of a correction law of the pure 2D airfoil characteristics, to be used in a BEM based design code.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
