Projects per year
With the technological and engineering developments in the design optimization and manufacturing of wind turbines, modern horizontal-axis wind turbines (HAWTs) are more flexible compared to conventional stiff designs. In addition, various new concepts involve non-straight wind turbine blades or strongly coned rotors, such as backward swept blades for passive load alleviation and downwind turbines for low specific wind speed conditions. For the design optimization and aeroelastic verification of such nonstraight blades, the aerodynamic effects should be correctly modeled. Otherwise, the design from a model that does not have the capability to correctly model the non-straight blade effects could be far from the actual optimal design. There exist high-fidelity aerodynamic models that are capable of modeling such effects, such as the liftingline method and blade geometry resolved Navier-Stokes simulations. However, these methods are computationally heavy and are not feasible for use in design optimization and repetitive aeroelastic simulations in large scales. This thesis aims to contribute to the research on the development of low-fidelity engineering aerodynamic models for the aerodynamic load calculation and design optimization of non-straight or non-planar wind turbine blades. Two special cases of blades with only in-plane sweep and blades with only out-of plane dihedral are investigated. For swept blades, the curved bound vortex influence and the changed starting position of the trailed vortex will both influence the induced velocity on the blade and consequently influence the loads. Both effects are included in a modified coupled near- and far-wake model that is the combination of analytical results and engineering approximations. The radial distribution of the loads as well as the rotor overall thrust and power predicted by the model are significantly improved compared to the conventional blade element momentum (BEM) method with a moderate increase of computational effort. For non-planar rotors, the streamwise shifted starting position of the trailed vortices are modeled with the superposition of the vortex cylinders. In addition, the consistent method of using the 2-D airfoil data for non-planar rotors is derived. The proposed blade element vortex cylinder (BEVC) method shows improved agreements with higher-fidelity models for predicting the radial distribution of loads and rotor overall thrust and power compared to the BEM method with a small increase of the computational cost. In addition, the BEVC method is used for the aerodynamic planform optimization of various non-planar rotors under different constraints. The designs are different from the designs from the BEM method and it is believed the BEVC designs are closer to the actual optimal design. The developed methods have been implemented into the aeroelastic simulation code HAWC2 and are capable of steady-state aerodynamic load calculation of rotors with either swept or dihedral blades. The current implementation enables the application of aeroelastic simulations to some extent, but future work is needed for the models to be confidently used for unsteady simulations. In addition, future work is needed to couple the two models for theapplication of arbitrary curved blade geometries with sweep and dihedral combined.