Aeroelastic optimization of MW turbines (AeroOpt)

A new beam element has been developed to model the anisotropic structures produced by layered materials based on a cross-sectional stiffness matrix provided by a pre-processor e.g. BECAS, or VABS. The element is validated with known test cases for both static and dynamic responses. A nonlinear bar element has also been implemented to model mooring structures. The bottom contact is handled by a set of springs and dampers in each node with varying properties dependent on its distance to the bottom. Individual line systems are connected by use of a series of constraints. A linear aero-servo-elastic state-space model of a wind turbine operating at a given operating point defined by constant wind speed, rotor speed and pitch angle has been developed for open- and closed-loop aero-servo-elastic eigenvalue and frequency-domain analysis. The applications of this tool called HAWCStab2 are: closed-loop stability analysis, controller tuning based on poles placement and frequency response design, and derivation of linear first-principle reduced order models for model-based controllers. Examples of the two former applications are given in this report. The aerodynamic damping of lateral tower vibrations have been estimated by 3D CFD and unsteady BEM models. A good comparison between the two models with very different complexity supports the continuous use of unsteady BEM for aerodynamic modeling of wind turbine dynamics. Generator torque control has furthermore been used to add active damping to the lateral tower vibrations reducing the tower fatigue load by 40% with the cost of only 10% increase in drivetrain fatigue load. Yaw slip can also be used to damp out excessive lateral tower vibrations if the distance from the yaw axis to the center of gravity of the nacelle-rotor structure is sufficiently large. A viscous-inviscid model for predicting the aerodynamic behavior of airfoils subject to steady and unsteady motions has been developed. The inviscid part is modeled using a panel method whereas the viscous part is modeled using the integral form of the laminar and turbulent boundary layer equations, including a quasi-3D approach to include rotational effects. A design and optimization code based on a lifting line method coupled with a Lagrange multiplier approach has been presented. The circulation distribution which minimizes the induced loss is found, and the blade geometry is consequently derived using 2-D airfoil data. The validation of the predictive capability of EllipSys2D for flatback airfoils indicated that the drag can be captured for these airfoils. In all 2D cases, the steady state results are closer to the measured lift than the unsteady results. CFD can be used to compare the quality of different flatback designs. The investigation of using slats to improve the performance of this airfoils concluded that lift coefficients of above 3.0 can be achieved for a 40 % flatback airfoil fitted with a 30 % chord leading edge slat with a stall angle of approximately 24 degrees. The multi-element airfoil design was validated in an extensive wind tunnel campaign and comparisons be-tween the experimental results and computations showed good agreement